LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATEPUE DATE DUE DATE DUE

t‘t qun
id LUUA

 

C 33
\-

ind

 

4 2 7 J‘Z-A

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCiFIC/DateDuo.p65—p.15

PUBLIC DISCLOSURE OF CORPORATE ENVIRONMENTAL PERFORMANCE:
IMPACT ON CONSUMER BEHAVIOR AND CORPORATE
POLLUTION REDUCTION IN KOREA
By

J ongyeul Moon

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Department of Resource Development and Urban Affairs Programs

2002

COP“

 

 

 

ABSTRACT
PUBLIC DISCLOSURE OF CORPORATE ENVIRONMENTAL PERFORMANCE:
IMPACT ON CONSUMER BEHAVIOR AND CORPORATE
POLLUTION REDUCTION IN KOREA
By

J ongyeul Moon

The purpose of this dissertation is to investigate the effectiveness of Public
Information Disclosure (PID) as a pollution control tool. The main research question is,
“Whether disclosure of corporate environmental performance information (CEPI)
sufficiently affects consumer behavior to influence market share and consequently
corporate interest in pollution reduction?” In order to predict consumers’ purchase
behavior change, this study tested hypotheses about three dependent variables: 1) attitude
toward corporation, 2) corporate credibility and 3) purchase intention toward products of
the corresponding corporations. This study also performed tests about hypotheses
concerning three moderator variables: 11) environmental attitude toward pollution, 2)
familiarity with corporation, and 3) information credibility.

Three hundred six Korean undergraduate students participated in the experiment
designed as a two-group random assignment combined with pre and post-tests, between
May and early J uneZOOl. Four currently existing Korean corporations and their four new
unnamed products with positive and negative CEPI were presented as stimuli in the
classroom setting of the post-test. “Within Subject Analysis” with Confidence Interval,
Inference Probability, and Significance Test was employed as a major statistical tool.

Data revealed that CEPI disclosure changes 1) consumers’ attitudes toward

specific corporations, 2) perceptions of credibility of the corporation and 3) purchase

 

 

3?“-
in:

tr;

intention toward products of the corresponding corporations in the positive direction for
non-polluting corporations and in the negative directions for polluting corporations. This
study also found 4) that positive correlation between consumer’s environmental attitude
about pollution and effect of CEPI and 5) a strong positive correlation between
information credibility and effect of CEPI exists. This study did not find a correlation
between corporate familiarity and effect of CEPI.

Based on these findings, this study predicts 1) that CEPI disclosure influences
consumers to change their purchase behavior in the negative direction for polluting
corporations and in the positive direction for non-polluting corporations, 2) CEPI
disclosure can generate market pressures or incentives for corporations to reduce
pollution voluntarily, 3) that PID can be an effective approach for pollution control, 4)
that consumers’ environmental attitudes and CEPI credibility are critical elements that
influence the degree of the effectiveness of PID, and 5) PID could be workable for even
less known brands or corporations.

Although most studies about effectiveness of PID have focused on financial
market reaction to CEPI, this study focuses on product market reaction and consumers’
sensitivity to CEPI. This study provides empirical evidence supporting the assertion that
PID can be a cost-effective pollution control tool, substituting for or complementing both
traditional command-and-control approaches and market-based instruments, especially in

developing countries.

COPyright by
JONGYEUL MOON
2002

 

{:8

P-
'1

Of?

(It

 

Pro;
COn‘
KICK]

IEV;‘

 

 

ACKNOWLEDGEMENTS

I would like to express my profound gratitude to my graduate committee members
for their support, advice, insightful questions and intellectual contributions to this
research: Dr. Eckhart Dersch - chair of the graduate committee, Dr. John H. Schweitzer -
dissertation director, Dr. Ralph Levine and Dr. J 0 Ann Beckwith — committee members.
Other than committee members, I would like to thank Dr. Kenneth Verburg, a professor
emeritus in the Department of Resource Development and my previous academic advisor.
Even after he was retired, he provided me with his intellectual insights and wisdom as
well as editorial suggestions for my dissertation writing. I could not have completed my
study without their enduring support and expertise.

For their valuable contributions to the data collection, I thank my Korean friends:
Dr. Jin-Won Lee in Department of Politics at Seoul City University in Korea, Dr. Yoon-
Sik Jung in Department of Communication at Kangwon University in Korea, J ae-Bum
Moon in Korea and Dr. Byung-Do Ahn in Korea. I extend my gratitude to my friend Dr.
Euijin Ahn in the Department of Advertising at Youngnam University in Korea. He
provided me with a great deal of insightful advice to strengthen methodological rigor of
my research.

Special thanks are extended to Dr. John H. Schweitzer and Urban Affairs
program. Without their financial support and guidance, my study could not have been
completed. I am also thankful to my family for their ongoing support: Youb Kim, Hoh
Moon and Susan Moon. Especially, Youb Kim deserves special mention for her thorough

review of my dissertation. Again, thank you all for your time and help!

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. ix

LIST OF FIGURES ........................................................................................................... xi

CHAPTER I

INTRODUCTION ............................................................................................................... 1
Introduction to the Research .......................................................................................... 1
Statement of the Problem ............................................................................................... 7
Purpose of Research ..................................................................................................... 10

CHAPTER II

LITERATURE REVIEW ................................................................................................. 12
Background of Environmental Policies ....................................................................... 12
Information Oriented Approach to Pollution Control .................................................. 20
Limitations and Advantages of Public Information Disclosure ................................... 34
Previous Cases of Public Information Disclosure for Pollution Control ..................... 43
Previous Empirical Studies .......................................................................................... 48

CHAPTER HI

CONCEPTUAL FRAMEWORK OF PUBLIC INFORMATION DISCLOUSRE .......... 58
Assumption for Corporate Behavior ............................................................................ 58
Social Norms and Corporate Profit Maximizing Behavior .......................................... 59
Corporate Reputation and Environmental Information ............................................... 60
Role of Information: Stick and Carrot ......................................................................... 62
Conceptual Framework of Public Information Disclosure .......................................... 65

CHAPTER IV

RESEARCH QUESTIONS AND HYPOTHESES ........................................................... 70
Research Questions ...................................................................................................... 70
Research Focus ............................................................................................................ 71
Research Hypothesis .................................................................................................... 71

CHAPTER V

RESEARCH METHOD ..................................................................................................... 81
Introduction .................................................................................................................. 81
Experimental Design and Participants ......................................................................... 82
Corporations and Products Selected ............................................................................ 83
Stimuli .......................................................................................................................... 84
Measures of Dependent and Moderator Variables ....................................................... 85
Procedures .................................................................................................................... 88

vi

 

 

 

 

 

 

CPI
SI

AP

 

 

 

CHAPTER VI

DATA ANALYSIS AND RESULTS ................................................................................ 92
Data and Subject Profile .............................................................................................. 92
Checking Unexpected Intervention between Pre and Post-tests .................................. 93
Stimuli Manipulation Check ........................................................................................ 93
Measurement Reliability and Validity ......................................................................... 95
Statistics for Data Analysis ........................................................................................ 112
Attitude toward Corporation ...................................................................................... 115
Credibility of Corporation .......................................................................................... 125
Purchase Intention toward Products ........................................................................... 135
Stimuli by Subject Interaction .................................................................................. 145
Environmental Attitude as a Moderator ..................................................................... 148
Corporate Familiarity as a Moderator ....................................................................... 155
Information Credibility as a Moderator ..................................................................... 163

CHAPTER VH

DISCUSSION AND INTERPRETATION ..................................................................... 170
Summary of Test of Hypothesis One, Two and Three .............................................. 170
Summary of Test of Hypothesis Four, Five and Six .................................................. 182
CEPI Disclosure and Consumer Attitude Change (H 1) ............................................ 189
CEPI Disclosure and Consumer’s Perception of Corporate Credibility (H 2) .......... 194
CEPI Disclosure and Consumer Purchase Intention Change (H 3) ........................... 196
Environmental Attitude and Sensitivity to CEPI (H 4) ............................................. 196
Corporate Familiarity and Sensitivity to CEPI (H 5) ................................................. 198
CEPI Credibility and Effectiveness of CEPI Disclosure (H 6) .................................. 208
CEPI Disclosure and Consumer Purchase Behavior Change .................................... 210
Lessons from ModeratOr Study for PID Pelicy Design ....... ........ ................ 221
Influence of PID on Long Term Business Strategy ................................................... 228
Policy Applications of Study Findings: Implementation of PII) and Suggestions for
Enhancing Effectiveness of PID ................................................................................ 229

CHAPTER VIII

SUMMARY, CONCLUSIONS AND LHVIITATIONS .................................................. 237
Summary .................................................................................................................... 237
Conclusion ................................................................................................................. 238
Limitation and Further Research ............................................................................... 240

APPENDICES ................................................................................................................. 243
APPENDIX A
Questionnaires for Experimental Tests (English Version) ........................................ 244
APPENDIX B
Experimental Stimuli: Type A and B (English Version) ........................................... 256

vii

APPENDIX C ............................................................................................................ 264
Correlation among Familiarity, Environmental Attitude and Information Credibility

Appendix D

Inference Probability .................................................................................................. 265

APPENDIX E

Korean Version of Questionnaires for Experimental Tests ....................................... 267

APPENDIX F

Korean Version of Experimental Stimuli .................................................................. 278
BIBLIOGRAPHY ............................................................................................................ 286

viii

LIST OF TABLES

Table 6.1. Stimuli Type and Mean, SD and SE of Stimuli ................................................ 94
Table 6.2. Factors and Number of Items ............................................................................ 96
Table 6.3. Inter-Correlation between Factors in the Pre-test ............................................. 98
Table 6.4. Inter-Correlation between Factors in the Post-test ........................................... 99
Table 6.5. Inter-Correlation between Factors in the Pre and Post-test ............................ 100
Table 6.6. Second-Order Factor Correlation in the Pre-test ............................................ 101
Table 6.7. Second-Order Factor Correlation in the Post-test ........................................... 101
Table 6.8. Error Rate of Heterogeneity (Samsung in the Pre-test) .................................. 103
Table 6.9. Error Rate of Heterogeneity (Hyundai in the Pre-test) ................................... 104
Table 6.10. Error Rate of Heterogeneity (Nongsim in the Pre-test) ................................ 105
Table 6.11. Error Rate of Heterogeneity (Binggrae in the Pre-test) ................................ 106
Table 6.12. Error Rate of Heterogeneity (Samsung in the Post-test) ............................... 107
Table 6.13. Error Rate of Heterogeneity (Hyundai in the Post-test) ............................... 108
Table 6.14. Error Rate of Heterogeneity (Nongsim in the Post-test) ............................... 109
Table 6.15. Error Rate of Heterogeneity (Binggrae in the Post-test) ............................... 110
Table 6.16. Measurement Reliability - Standardized or .................................................. 111
Table 6.17. Interpretation of Inference Probability ......................................................... 114
Table 6.18. Descriptive Statistics of Attitude toward Four Corporations ........................ 116
Table 6.19. Attitude Changes toward Four corporations ................................................. 117
Table 6.20. Descriptive Statistics of Credibility for Four Corporations .......................... 126
Table 6.21. Credibility Changes of Four Corporations .................................................... 127
Table 6.22. Descriptive Statistics of Purchase Intention ................................................. 136
Table 6.23. Purchase Intention Changes toward Four Products ...................................... 137

ix

Table 6.24. Stimuli by Subject Interaction ...................................................................... 147

Table 6.25. Environmental Attitude toward Pollution and

Effect Size of Three Variables ............................................................................... 149
Table 6.26. Familiarity with Corporation and Effect Size of Three Variables ................ 156
Table 6.27. Information Credibility and Effect Size of Three Variables ......................... 164
Table 7.1. Summary of Test of Hypothesis One, Two and Three ................................... 171
Table 7.2. Summary of Test of Hypothesis Four ............................................................ 183
Table 7.3. Summary of Test of Hypothesis Five ............................................................. 186
Table 7.4. Summary of Test of Hypothesis Six ............................................................... 188
Table 7.5. Attitude and Credibility toward Four Corporations before Stimuli ................ 207

 

 

 

 

Hi

Fi

LIST OF FIGURES

Figure 3.1. Public Information Disclosure as Pollution Control Tool: Carrot and Stick

Principle of Organizational Behavior Control ....................................................... 64
Figure 3.2. A Causation-Focused Model of Public Information Disclosure ..................... 67
Figure 3.3. Conceptual Framework of Public Information Disclosure .............................. 69

Figure 4.1 Hypothetical Diagram of the Effectiveness of CEPI on Consumer Purchase

Behavior ................................................................................................................. 77
Figure 6.1. Attitude Change toward Samsung ................................................................. 118
Figure 6.2. Attitude Change toward Hyundai .................................................................. 120
Figure 6.3. Attitude Change toward Nongsim ................................................................. 122
Figure 6.4. Attitude Change toward Binggrae ................................................................. 124
Figure 6.5. Credibility Change for Samsung ................................................................... 128
Figure 6.6. Credibility Change for Hyundai .................................................................... 130
Figure 6.7. Credibility Change for Nongsim ................................................................... 132
Figure 6.8. Credibility Change for Binggrae ................................................................... 134
Figure 6.9. Purchase Intention Change toward Samsung Cellular Phone ....................... 138
Figure 6.10. Purchase Intention Change toward Hyundai Cellular Phone ...................... 140
Figure 6.11. Purchase Intention Change toward Nongsim Instant Noodle ...................... 142
Figure 6.12. Purchase intention Change toward Binggrae Instant Noodle ...................... 144
Figure 6.13. Inference Probability of H4 for Attitude toward Corporation ..................... 151
Figure 6.14. Inference Probability of H4 for Corporate Credibility ................................ 153
Figure 6.15. Inference Probability of H4 for Purchase Intention .................................... 155
Figure 6.16. Inference Probability of H5 for Attitude toward Corporation ..................... 158
Figure 6.17. Inference Probability of H5 for Corporate Credibility ................................ 161
Figure 6.18. Inference Probability of H5 for Purchase Intention .................................... 163

xi

Figure 6.19. Inference Probability of H 6 for Attitude toward Corporation .................... 166

Figure 6.20. Inference Probability of H6 for Corporate Credibility ................................ 167
Figure 6.21. Inference Probability of H 6 for Purchase Intention ................................... 169
Figure 7.1. Attitude Change and Effect Direction .......................................................... 173
Figure 7.2. Stimuli Correlation of Attitude Change as Effect Size and Pattern .............. 174

Figure 7.3. Probability of Effect in the Direction Predicted (Inference Probability) ...... 175
Figure 7.4. Corporate Credibility Change and Effect Direction ..................................... 176
Figure 7.5. Stimuli Correlation of Credibility Change as Effect Size and Pattern .......... 177
Figure7.6. Probability of Effect in the Direction Predicted (Inference Probability) ....... 178
Figure 7.7. Purchase Intention Change and Effect Direction ......................................... 179
Figure 7.8. Stimuli Correlation of Purchase Intention Change as

Effect Size and Pattern .......................................................................................... 180

Figure 7.9. Probability of Effect in the Direction Predicted (Inference Probability) ...... 181

Figure7.10. Summary of Inference Probability related to H 4 ........................................ 184
Figure 7.11. Summary of Inference Probability related to H 5 ....................................... 187
Figure 7.12. Summary of Inference Probability related to H 6 ....................................... 189
Figure 7.13. Attitude and Credibility for Four Corporations ........................................... 207
Figure 7.14. Moderator Variable ..................................................................................... 222
Figure 7.15. Mediator Variable ........................................................................................ 222

xii

 

 

lntror

confi

Instr

 

 

 

 

CHAPTER I

INTRODUCTION

Introduction to the Research

Pollution control policy based on traditional command-and-control and market-
based approaches has not been completely successful. A new approach for achieving
greater pollution control is emerging.1 It could be a powerful supplement or complement
to the traditional command-and-control approach and market-based approach to pollution
control policy. A group from the World Bank named NIPR (New Ideas in Pollution
Regulation) introduced this idea as “Multiple Agents, Multiple Incentives: A New View
of Regulation” (Afsah, Laplante & Wheeler, 1996, p. 7). The new idea is grounded on
the principle that “one size no longer fits all for regulatory policy design: Optimal
combinations of regulatory tools will depend on country-specific social, economic and
institutional conditions” (Afsah, Laplante & Wheeler, 1996, p. 7). It is called the
Information Oriented Approach or Public Information Disclosure (PID) for pollution
control.

This public information disclosure strategy was adopted for pollution control in
Indonesia in 1995. Faced with growing industry and rapidly deteriorating environment,
Indonesia’s National Pollution Control Agency (BAPEDAL) and the NIPR of the World
Bank initiated the Program for Pollution Control, Evaluation and Rating (PROPER) for
rating and publicly disclosing the environmental performance of Indonesian factories.

PROPER assigned participating plants color-coded grades indicating their compliance

 

' Tietenberg (1998) stated that legal regulation is the first wave in pollution control policy, market-based
instruments are the second wave, and the third wave is information strategies. He defined that information
strategies involve public and/ or private attempts to increase the availability of information on pollution.

with pollution regulations. “Gold” meant that the plant was a world-class performer,
while “Black” signified regulatory violations causing serious damage to the environment
and human health. Disclosure of this color-rated information functioned in the same way
as credible threats of real punishment to the worst polluters (i.e., corporations with
“B1ack”). Afsah, Laplante and Wheeler (1997) reported considerable improvements in
compliance status both before and following the public announcement.2

The idea behind PROPER was to provide reliable and easily understood
information about pollution to the public because reliable and well understood
information about corporate environmental performance can create strong new
reputational incentives for corporations to voluntarily abate their pollution (Wheeler,
1997)

Our societies need a new approach for pollution control to supplement or
complement the current pollution control tools because the current tools, the command-
and-control (CAC) regulation and market-based instruments (MBI), were revealed to be
very costly in some circumstances and incapable of achieving the defined goals in others.
The pollution policy designers determine the ‘optimal pollution’ at the point where
marginal social damage is equivalent to marginal abatement cost. CAC mandates
factories not to pollute above the optimal pollution level. The pollution charge of MBI
sets pollution price at the optimal pollution level and the tradable pollution permit of MBI
allows factories to trade pollution permits within the limit of the optimal pollution. Both
CAC and MB] have the following assumptions: 1) full information availability and 2) no

transaction costs. However, transaction costs are not zero and full information is almost

 

2 As a PROPER—type programs, the Philippines launched EcoWatch in 1997, Mexico is developing PEPI
(Public Environmental Performance Indicators), and Colombia also started a public disclosure program to
complement its pollution charge system

 

 

 

COI‘.‘

P0?

 

 

£0) .

 

8d:

ei’i

never available in the real world (Afsah, Laplante & Wheeler, 1997). In other words,
CAC and MBI can be effective pollution control instruments under the right conditions
but it is unfeasible to achieve the right conditions in the pollution control practices. In this
situation, it appears that the information-oriented approach could be a supplementary or
complementary pollution control tool to both traditional CAC and MBI.

The idea of information disclosure to control pollution is highly attractive to
pollution control policy makers and designers, especially in developing countries where
government enforcement resources are limited. Because of the ineffectiveness and the
high cost of CAC (Tietenberg, 1985; 1990; 1995)3, MBI have been much more prevalent
in the practice of pollution control in the United States (Tietenberg, 1998, October).
However, in the developing countries the regulatory infrastructure for implementing MBI
is insufficiently developed. In developing countries such as Korea, Taiwan, Thailand, and
Indonesia, the monitoring problem is compounded by weak enforcement (O’Connor,
1994). Formal regulation in developing countries also has been greatly hindered by the
absence of clear and legally binding regulations, limited institutional capacity, lack of
appropriate equipment and trained personnel, and inadequate information on emissions
(Hettige, Hug, Pargal & Wheeler, 1996). In the case of either traditional regulatory
approaches or market-based approaches, developing countries are incapable of handling
adequately the burden of designing, implementing, monitoring, and enforcing an
effective pollution control system (Tietenburg, 1998). In this respect, PII) advocates
argue that PID would be a new cost-effective pollution control strategies for the

developing countries.

 

3 Tietenberg ( 1990) found that the CAC costs from twice to 22 times the least-cost alternative for given
degrees of control.

3':qu

 

 

 

10:?

em:

 

Korea is a developing country similar to the Philippines, Colombia, Mexico4 and
Indonesia which launched PID programs for pollution control. Korea has had some
experience with information disclosure programs in the public policy area, even though
they were not related to the topic of environmental protection. The Act of Disclosing
Property Information of House of Representatives and High Ranking Government
Officials was promulgated in 1993 in Korea. The Military Service Bureau disclosed the
list of people who had evaded obligatory military service. Similarly, in the national
parliamentary election held in 2000, tax information of candidates was disclosed by the
Central Election Management Agency. Korean government released lists of those related
to the crime of adolescent prostitution in public in 2001 and 2002. However, limited
empirical data has actually been developed on the topic of the effectiveness of
information disclosures. Hence, more research is needed to examine the effectiveness of
disclosing corporate environmental information and its possible impact on pollution
control policy in Korea.

Even though PID was effective in Indonesia and applications of PID have been
increasing recently in both OECD (Organization for Economic Cooperation and
Development) and developing countries, these facts do not assure the effectiveness of
PID in different social contexts such as in the context of Korea. Korea is barely familiar
with the concept of information disclosure strategies in pollution control. Measures to
enforce disclosing information to the public, such as the Community Right to Know Act
directing the release of Toxic Release Inventory (TRI) data to the public in the United

States, or pollution control program such as Indonesia’s PROPER, have never been

 

4 The success of PROPER in Indonesia has inspired a similar program in the Philippines, EcoWatch and
evoked serious interests in Colombia and Mexico. Refer to the website of NIPR at
http://www.worldbank.org/nipr/.

b

I O
ctr-mat In kcrce

”'5‘” 'n.‘.'..'_‘ .:
rebut... .a’nL ut‘
U

#43:} éEPAI .\f.

' ...‘ ‘ n 1
53.533325“ .\.;i. :

an- iw 3 ' '
I;,":a" \‘L "‘00 n

'Ht 8:15

i ' .... '

x“:

THC .30 ‘A C: C '

Damn-w .

a ..... .‘1.I\ acm't't

UH€,\.H‘:‘ I I
,I ten-0mm. -'

Hr. ~..
“~59" 9n. ‘
~¥HJgutcs :01'

.N

53216 g

~0Ven'umcfl: a:

We: - ' ‘
J: Intermauon (

‘1.
12cm?

 

conducted in Korea. Korea has not experienced a voluntary compliance and incentive
program like the “33 /50” program initiated by the US. Environmental Protection
Agency (EPA). Most pollution control policies in Korea are based on the CAC. Even
though a few MBI such as pollution charges or deposit refund systems have been
employed, both the Korean government and polluters are unfamiliar with the idea of a
MBI such as a tradable pollution permit.

The power of capital markets and community action for pollution control in Korea
is assumed to be considerably weaker than in the United States because the Korean
community activists do not seem to have been organized sufficiently with the result that
they have not developed enough power to provide a credible threat of adverse
consequences for polluters in Korea. The various information disclosure programs hosted
by the government and NGOs (N on-Govermnent Organizations) in the Korean political
and social arena have rapidly increased. However, it is inferred from the lessons of the
previous information disclosure programs that much strong and well-organized resistance
from the comprehensive coalitions of Korean industry and the political majority will
occur. Even though the environmental consciousness has continuously risen because of
severe environmental conditions, top national priorities in Korea are still the issues of
economic grth and national defense. Koreans might be lenient toward polluting
behavior of industry because of the strong desires for economic growth. People’s desires
for affluent consumption seem to be prevalent in Korea.

Thus, even though PID is asserted to be a new cost-effective pollution control
strategy for the developing countries, it is questionable whether corporate environmental

performance information (CEPI) disclosure would be effective for pollution abatement in

cu ‘
t‘,.. .‘Inv1=‘.‘\ “‘I'
Ana 3 and...» CL-

I ' '
3 I1 '9‘.‘“ V .N“ v
‘ . Li .5)....‘ ‘AaL b
a 5

DW‘W‘F‘I 9-0 -n-~
. bll\!lIA§&‘\L AAA\..

. 1.- .
Ya.“ 81"“. 'i r -r
IDS-tbs t- ‘ern‘Jn J. ..

by»! ‘ .
u...;. . AFRICA; h
| o

 

or?“ .- ~
I. .t‘atiion 0; Co"

M
..Im€nl10l1€d b

(is
“ill .
Chilge COnSU. .‘M

 

Korea’s unique context.5 Hence, this study is designed to investigate the effectiveness of
PID by testing the PID theory that disclosure of CEPI (Corporate Environmental
Performance Information) generates market pressure or incentives for corporations to
reduce pollution in Korea.

A theoretical base of PID has not been well developed because PID is an
emerging approach. Hence, more empirical studies are needed to understand the effect of
existing public disclosure programs on the corporate environmental performance. Most of
the empirical studies and theoretical assertions for PID focus on reactions of the stock
market and the community to the release of information. Few researchers have made
efforts to analyze the conceptual framework of PID. Little empirical research has focused
on the reaction of consumer and product market to the disclosure of CEPI. As Cohen
(1998) mentioned, before PID is more frequently implemented as a regulatory
mechanism, we need to understand how the theory works and what effect PID has on
corporate behavior. Thus, this study investigates the influences of CEPI disclosure on
consumer purchasing behavior, product market share, and corporate pollution
performance. It is based on the conceptual framework of PID, which suggests that CEPI
will change consumer attitude and sense of credibility about corporations, and interest in
purchasing products from polluting corporations. As a result, these changes in consumer
behavior generate market pressure or incentives that will encourage voluntary corporate

pollution abatement.

 

5 Even though Jeon (1998) asserted that disclosure of corporate violation had negative impact on the Korea
stock market, this study does not rely much on his conclusion. For details, refer to Chapter II, pp. 54-55.

I
Statement of the PI

Mr.) crap:
til-3.5.x offor’p‘ '
9.3.3201: (19‘15 l )1
1:233:35 tIRii In 1"
mar trims cc; 4:"
masons m 19:»

‘
A

\'a¢~o m V.
r O O I
’ .' . x .0
I.“ J . .. k\\ .‘ ‘t

J “a . ' . .
33x. .t'JIC .1370?"

n
H Hui] ,

5mm 115*"

- Mimi:

5.3} ”If ‘ n"
_ -u fittifmi’f‘l'fi ( '

 

 

 

 

[bx-3" .‘m' 9.
w .116 teacher“.

0,2. '
was imposed b\ .2.
. t..

Sitar-*1 ' '
t. I: In Bntzsht

h t’n i
*‘~ ~46 it‘

is " ~
3601 men:

ILLIO'A mg the I}
“ OUI’IC

Statement of the Problem

Many empirical studies have focused on the stock market’s reaction to the
disclosure of Corporate Environmental Performance Information (CEPI). For example,
Hamilton (1995) studied the reaction of stockholders and journalists to the Toxic Release
Inventory (TRI) in the U. S. and found that statistically significant negative abnormal
market returns occurred when TRI releases were first reported for publicly traded
corporations in 1989. Laplante and Lanoie (1995) investigated the reaction of the
financial market to the announcement of environmental incidents and lawsuits in Canada
and found abnormal losses of stock value of Canadian-owned corporations ranging
between 1.65% and 2% when the firms were found guilty (and fines are imposed) on the
day the settlements of lawsuits were announced. Lanoie, Laplante and Roy (1997)
observed the reaction of the capital market to the release of information such as penalties
or fines imposed by the courts or regulators and announcements of lawsuits or suit
settlements in British Colombia and provided evidence indicating that stock markets react
to the release of information based on American and Canadian data. Konar and Cohen
(1997) studied the reaction of the capital market to the disclosure of Toxic Release
Inventory (TRI) and found that corporations with the largest negative stock price effects
following the announcement of their TRI emissions were found, subsequently, to reduce
their emissions more than other firms in their industry and also to make other significant
attempts to improve their environmental performance by reducing the number and
severity of oil and chemical spills. Dasgupta, Laplante and Mamingi (1998) studied the
reaction of the capital market in Argentina, Chile, Mexico, and the Philippines to

environmental news such as the violation of permits, spills, court actions, citizen

nan-n""'" 'f‘ C ’
b‘.v-<.p--;‘c-§: lib :‘n ‘-I

I .
.:,,....v.,..1...,\ ,
hm. «\IJK’.‘ _... ..

‘

}‘\~ .3..."- Q ‘N. _' ‘-
ok“) ...... ml K'IJKA c
u

I. v

a .I F m - O
.P........‘S. L0...;‘.;_fi

. .' .L. '
.sxmc r. l.

‘4

i.
u
r

,

’N'ivenm ‘ r .
tl- .3r‘. .‘f
bn'nk'i..a.§ ctd' : .I '

. . . ..

a

1’1: gm ants abet:

Watch: i 1w

 

 

 

 

““3” I
“-Lhrlg‘i ~ 7‘ " ‘
«W. l. at. t

 

complaints and protests, agreement between government and companies, investment in
clean technologies and environmental protection, announcement of pollution abatement,
government black list of polluters, and government actions such as warnings, fines,
penalties, complaints, and company shutdowns. J eon (1998) examined the affect on the
firrn’s value in the Korean stock market from both bad news about corporate
environmental performance (e. g., the government publicizing environmental violations)
and good news about corporate environmental performance that is released voluntarily by
firms. 6

Wheeler (1999), Afsah, Laplante and Wheeler (1997), Wheeler (1997), Afsah and
Vincent (1997), and Wheeler and Afsah (1996) provided empirical evidence suggesting
that community influence increased as a result of the public announcement of
environmental information. Improvement in compliance status both before the initial
public announcement and following the public announcement, based on data from the
PROPER program in Indonesia was noticed. The research about the effectiveness of
disclosing corporate environmental performance information (CEPI) to the public for
pollution control is currently taking place in Argentina, Chile, Mexico, and the
Philippines7.

However, we need to understand how the theory works and what effect PID has
on corporate behavior before PID is used more extensively as a pollution control tool, as

Cohen (1998) mentioned. “From an empirical perspective, the impact of existing public

 

6 For the findings of previous studies related to stock market reaction by CEPI disclosure, go to Chapter II,
. 49-53.

g”PROPER-type programs began in Philippines, Mexico, and Colombia. The results from those programs

have not appeared yet (Wheeler, 1999).

. I‘
n3".‘ k‘"
r""- “'..
.»-' a
r" '\ . 7“:
5“)“ .

93 scald 3" t "

runes such as Kora

if a stud) for;
'31 in Korea is co::.

fraction oithe prod ..
COTSruction of the 1'."

cartel.

The outcom.

 

   
 
 

I

PI) works effectiu
The result of this _~
the possibility that

research “m prox':

 

disclosure programs on the environmental performance of the plants largely remains to be
tested” (Foulon, Lanoie & Laplante, 1999, p. 13).

PID could be a very cost-effective pollution control tool, especially in developing
countries where institutional capacity and resources for pollution control are limited.
However, except the studies from the case of PROPER and J eon’s study, few studies of
PID in developing countries were conducted. Little empirical work has been done on the
product market changes and reaction of the consumers to the disclosure of CEPI. The
idea of P11) is conceptually framed with two constructs: 1) CEPI creates public’s negative
attitude and behavior toward polluting corporations, and 2) the public’s negative behavior
causes corporate pollution reduction. However, few researchers formulated a theory of
PID. Thus, it is necessary to conduct an empirical study focusing on the reaction of
consumers and product market to the disclosure of CEPI in the context of developing
countries such as Korea.

If a study focusing on the reaction of products and consumers to the disclosure of
CEPI in Korea is conducted, the study will become the first empirical study about the
reaction of the product market to disclosure of CEPI and it will contribute to the
construction of the theoretical base of the information oriented approach to pollution
control.

The outcome of this study provides evidence to confirm or reject the assertion that
PID works effectively as a pollution control tool in developing and developed countries.
The results of this study supply primary data and scientific evidence for predeterrnining
the possibility that a PID program for pollution control in Korea could be effective. This

research will provide basic data and evidence that strengthen the theoretical content and

 

..
u a '9 1.00“”

t. ~ I
I. g..- .‘ob Va n-h
bx 0‘”

1 .. u 1 e
I ‘ ‘ ‘ t ' O
_ .. .. ‘ 0 . I
r... :L ’s.l‘-"\.s\.-5 .k .
-

I .‘| 1 O '
o ’,
K \ . \ \ . é
”vs no an. .9- H- b

‘r M v.. -
mmhéhgn anJ" .. I)"
’ 3 “¢- \\ ..

’Y'us. '
l W 9 .
Kt .1") 3" A, M

’ r so“
«mist-Lu. “to ‘ Li“;

‘71 OiCEPld

‘R“ ~
xx ((1‘\

 

L4

\

 

 

 

 

causal structure of the idea of the infonnation-oriented approach for pollution control.
The study gives present pollution policy makers and pollution program designers more
detailed knowledge for designing a better infonnation-oriented policy and program. The
results of this study are expected to facilitate the development of legislation for a “right-
to-know act” and to initiate public information disclosure programs elsewhere in the
world. The outcomes and data from this study can also be used in environmental studies
and practices related to public involvement or community participation for environmental
protection in areas such as environmental labeling, environmental advertising, green
certification, green consumerisms, environmental justice, environmental journalism, the

environmental movement, environmental education, and environmental attitude studies.

Purpose of Research

This study has three purposes. The most important purpose is to examine the
effeCt of CEPI disclosure on the product market. That is, it is to investigate whether
negative information disclosure of corporate environmental performance is likely to
reduce consumer demand for products of polluting corporations, and consequently shrink
the market shares of polluting corporations.

The second purpose is to examine the conceptual framework and viability of
public information disclosure (i.e., information-oriented approach) for pollution control.
This purpose can be achieved by examining the causational relationship between CEPI
disclosure and the changes of consumer’s attitude or behavior toward polluting or green
corporations. The results of the test for the second purpose will serve as a direct answer

for achieving the first purpose.

10

T' - I —
0'.»' "N
.3€“"L‘:

, wt 0
”I .v \N',‘bl EA. .‘r‘

. , ... ..
_l. M. -.-

.:..... a .. .- 9'“)
o I

._(.~.. I h.a§v

[It

{"7 .-,...J\. Ta” .
- .-a-. 5'...\. L. ..5 u

if " "‘9‘ R
Us )Ho..aiah;. \.

.. .... . t..- MS...” t.
aq~.“. I!
., ,.. hp ~ 1 m
I~o§ “.1. N45 tk‘:k\. .
l

0
‘fi. .. I ~¢
-),u
.L‘Wv " Qt
f“““‘dd C 21"".
.255...

 

The third purpose is to investigate the effectiveness of PID as a pollution control
tool in Korea. Even though PID was effective in Indonesia, this fact does not assure the
effectiveness of PID in different social, economic, political and environmental contexts
such as Korea. Hence, it is necessary to examine the effectiveness of PID in a different
social context. The third purpose also can be achieved from the results of the first tested
purpose.

In summary, the primary purposes of this study are l) to investigate the influence
of CEPI disclosure on product market share and consumers’ response to CEPI, 2) to
examine the conceptual framework of the idea of PID, and 3) to find supporting evidence

for potential effectiveness of PID for pollution control, especially in Korea.

11

Ru;

i ..

litiground oi Em ir

:V' VO‘O-m.g'.o ’
Ark ‘u...-...5..y

.

-~--»g1 up) ~ ‘v-"
. -. .

u'...__.._.\. *iu’K'. ....

t

H. NI}? 3 I‘
In. . ho .5-..‘ ‘\' . h
c

n

am ”it.-. . ,5 1) ‘I‘
HP! u‘. - \‘1 '
' ‘. ,_:§_ ||"“‘1:-‘
_. =
"-4-?“ pariah”;
. “M . ...\_ .
‘4. “it ‘
P " ‘
LA‘QI‘ h. thr\ g‘
hlfi‘m.
vb, l‘ “\‘P‘h . -
nud-K:.;l.\nz‘ of \}~ I

 

 

M310“ tom

1
01 PC;
5

"‘14.. .,
MAC“: (k. The ur‘ .-
l AIL an,

Ibrii are SUN.

lies for T

 

CHAPTER II
LITERATURE REVIEW

Background of Environmental Policies

Environmental policy for pollution control can be classified into four categories:
command-and-control approach, market-based instruments, voluntary incentive
programs, and information-oriented approach. The command-and-control approach is a
traditional legal remedy dictating environmental standards to industry, while market-
based instruments is appealing to corporate profit-maximizing motivation using market
principles. Voluntary incentive programs seek voluntary cooperative partnerships among
regulatory authorities, environmental NGOs, and industry on the basis of social norms of
industrial responsibility. The information-oriented approach seeks to impose
embarrassment or shame on polluting corporations, undermining their reputations, and

diminishing corporate sales and profits (Hoffman, 2000 a).

Editiongl Environmental Policy: Command-and-Control ApproaLh
Command-and-control (CAC) is a top-down legislative and administrative
pollution control regulation. It sets the target for the total quantity of pollution emitted by

all sources collectively and prescribes the best available technology that corporations
should use for their pollution reduction. Thus, CAC is called “technology-based
regulation” and the standards set for pollution control are called “technology-based
standards.” The uniform emission standards as targets and a specific technology
requirement are supported by effective pollution monitoring systems and by sufficiently

harsh penalties for non-compliance.

12

1"...» '7'.” 9" 'N
I ‘~ >84. ...... .
-. w‘" T» .
1 ":2... i J
‘ '

nip " ‘1'."‘ '1'
‘ " ; I“ -5 in May
_. - -

CAC IS 8 13”

'53", . ,t'n ’ '1‘ ‘
.qu‘x “db ‘L\tfi:“

V
\AOU-q ~-~ I I
“n” ‘r ~,)I v'a'fn‘ fl .
:..t~a\¥.as5ss.s 5

'5 302716 35813

The meIcht"

,5
[is

"tr" .
.i\ 3'7“.“ H' ‘ A
.-\5nsl\ail\ IOUFd

’v
its: pollutants 1hr
n l

continuous hat 6 no

his ' V
A tnuronmcnt V

Corporations

 

f
t0.“ 5 ‘
M eni'ironmcr"

hr
filament. Comor“
d. .

 

C.
AC does not prov; _

Q

~ Decause C .-\(

 

The strength of CAC lies in its simplicity, clarity, and ability to command broad
political support. The CAC system is easy to administer and very effective in the sense of
achieving large reductions in emissions quickly (Beardsley, Davies, & Hersh, 1997).
Thus, CAC has been the most widely used pollution control instrument in environmental
pohcy.

CAC is a less effective pollution control instrument than desired, however,
because it has revealed problems that include a high enforcement cost and imperfect
compliance (Tietenberg, 1985; 1990; 1995 ). For example, CAC has not perfectly
achieved industry compliance with emission standards. The intensity and rate of
corporate non-compliance with national air pollution standards is continuously increasing
in some areas.

The ineffectiveness of CAC is caused by its inflexibility and inefficiency, which
is systemically found in CAC. CAC removes the incentive for polluters to discharge
fewer pollutants than the regulatory standards require. Under certain circumstances,
corporations have no incentive to devise new technologies to decrease the pollutant level
below environmental standards.

Corporations respond to incentives because corporate managers make decisions
for their environmental performance by comparing costs and benefits for pollution
abatement. Corporate behavior may change when the costs or benefits change. However,
CAC does not provide any incentives for corporate behavior changes in terms of cost and
benefit because CAC is not flexible. In the scheme of CAC, corporations have no

freedom to choose the least expensive and the most effective technology, because the law

13

.
1x °H;‘ 'A F'- k
.- - fi‘. .
.~""'
I ' ‘
' r w I '
n \ of 3"» i

’ ‘ I
W‘ .31- ”... ‘J I
. iuu-";‘
-
r... p l“ VI '0‘
.5. . ‘ \fi aLr .h" ' ~‘ . ~
I” s
T‘s- "" .‘ H
A” “1 ~ ~

111“

. .u’n‘

‘. " ‘.
....'.;i.-.I\. i‘i‘fl‘il I"

‘.‘lb‘.~i OL.’ L - --
"*---‘-u HA5\ A54“ 6 .‘tn'\

1
-,4
I.

r . Q...- " H
‘\ mirth) :OI' p..\.-u

as; or pollution. so

at a greater cost than

5‘ h... ‘
trio. 416 cost tunctir

mtiCCilng infonnatio

In sum. \\ he:

iOCO ’r l '
no. pollution x

 

 

 

decides the technology. The requirement for specific technology thwarts the use of less
expensive and more innovative methods of achieving environmental goals.

As a result, this inflexibility of CAC imposed high cost of pollution reduction on
the corporations, and that causes the inefficiency of CAC. An empirical study conducted
by Tietenberg (1990) found that the CAC costs from twice to 22 times the least-cost
alternative for given degrees of control.

The inefficiency of CAC is promoted by asymmetric information (Perman, Ma &
McGilvray, 1996). While corporations may know their cost functions of pollution
control, they have strong incentives not to disclose the functions of their marginal costs
and benefits for pollution reduction to regulators, thus corporations may provide
misleading information to regulators (Perrnan, Ma & McGilvray, 1996). As a result, the
standards set by the pollution regulatory authority may not be the socially acceptable
level of pollution, so corporations tend to attain the required level of pollution reduction
at a greater cost than is necessary in the scheme of CAC. Thus, the inability to know the
corporate cost function because of the insufficient information and the high cost of
collecting information weaken the efficiency of CAC.

In sum, where serious public and environmental health issues exist, the best way
to control pollution would be dictating environmental standard to polluters, which is
CAC. CAC is quick and effective to achieve low level of pollution abatement because the
cost for a small amount of pollution reduction is small. However, CAC will be inefficient
at the higher level of pollution reduction because as the pollution reduction is getting
larger, the marginal cost of reduction is also sharply larger. The imposed technology

choice and arbitrary target setting for emission, that characterizes the inflexibility of

14

u
0 F-

. ' u g. I *3
“u ' "fig.“ ‘\ '
L‘s». 3.20 .5 - h

.50.

c
. ,.T:_..o " c' .
_A . ' . ‘ I
.v .‘.I>~.L-‘.\ ‘ '
u.

T. ' ,, . '

hatter. l .W l 47.
1‘." ‘ I Q

Int .\;3. mat-cs t»;

12.-(0:2:ldfit‘i‘ or o:

:1 1'0"?" e

I e
.A .
.mzcl‘ndl 6.1::

 

 

CAC, also leads to the increase of pollution reduction cost. In conclusion, CAC does not
work cost-effectively because of the high cost of pollution reduction caused by

inflexibility and insufficient information.

Market-based Instruments

Most economists assert that the market-based instruments (MBI) can overcome
the CAC limitations of inflexibility, inefficiency and ineffectiveness for pollution control.
Tietenberg (1990) argued that MBI is a cost-saving tool for pollution control compared to
CAC. MBI makes polluting corporations aware of the opportunity cost of environmental
non-compliance or of pollution reduction, and leads to the internalization of
environmental damage costs. Thus, it is asserted that MBI tends to result in pollution
control being undertaken where the control is least costly in real terms. MBI includes
practices such as pollution charges or pollution taxes, tradable pollution permits, deposit
refund systems. I

Pollution Cha_rgg Pollution charge (PC) has been used as a supplemental tool to
CAC. Pollution charges are fees levied for exceeding standards (e. g., taxes on unleaded
gasoline, heavy vehicles, etc.). Such taxes create incentives as every unit of pollution
reduction is rewarded by a tax saving. Thus, PC achieves more efficient outcomes by
internalizing pollution costs at the socially efficient pollution level through the
modification of product prices.

In the PC system, the pollution charge rate will affect the amount of abatement by
working on price. The magnitude of the economically efficient charge rate should be

equal to the magnitude of the marginal external damage of the pollution at the socially

15

II I ‘
L". 3‘ J D
2‘1. ..1.-‘L'-“' r
.0...- ”. .p. 'u') r30
' 6 bin. 51 *h‘ .u .
-. -

qr" ‘9 I “’1 ‘
7““...5». II [S It..: .\

5‘. o

If‘.‘ a "
.cfonor. 1995». l

A:

tw-epoildtzon ;‘

 

 

 

 

optimal level of pollution. However, it seems to be mostly infeasible to determine the
efficient charge rate, because of uncertain abatement cost. If the abatement cost function
is known, the control authority can determine what emission charge rate is needed to
achieve any given level of abatement. If the abatement cost function is unknown, the
amount of abatement that results from this charge will not be known. Therefore, in
practice, it is infeasible to determine the economically efficient charge rate that can
achieve the socially efficient optimal level of pollution and the quantity of pollution
abatement achieved by the PC.

Policy makers sometimes propose to raise the rate of pollution charge up to the
level of a socially efficient pollution level because the costs for non-compliance is less
costly than the cost for pollution reduction at high levels of pollution abatement
(O’Connor, 1995). However, political feasibility is an essential constraint for setting an
adequate pollution charge. The raised charge will be a burden on the economy and
people’s economic activities, and as a result it will generate strong political resistance.
Thus, it is not practical to raise the tax to achieve a socially optimal pollution level.

_Tr_adable Pollution Permit. Tradable pollution permit (TPP) predetermines the
level of “acceptable” effluent emissions or ambient concentrations, rather than acting
directly on prices like PC. The pollution permits allocate this acceptable amount among
polluters. TPP allows trading the pollution permits for money with others. Because the
corporations have the responsibility to calculate the costs and benefits of pollution
reduction, TPP induces corporations to develop least-cost strategies.

In the TPP, the price of a pollution permit that is traded in the market is not fixed.

The price is determined by the seller and the buyer so that regardless of the price of the

16

_ 9‘
ct .‘
aside Inertia») :‘
..~-'-‘
5111 3&4:ng

s-m ‘"
§

-£..:SIIICIC‘T‘.C} and
coral. The empznc
P? is last cost?) ct

Ho“ ever. I?

maxed as t’oiio‘

A~ ‘
mnem

c infonnu:

inappropnateness of

MicGiltra); 1990
hide system of TPE
The diliiculf

more flexible and C

 

 

l

mil-imam

h‘F‘dimE adequate? :

1ng it. T.

Q
\

i‘iiii‘spoinn

 

pollution permit, the target of pollution reduction will be theoretically achieved at the
pollution level that is an acceptable effluent emission level pre-set by pollution control
authorities. In the pollution charge system regulators decide the price of pollution as a
pollution charge per unit of pollution, but in TPP, regulators decide the total amount of
pollution (i.e. target for pollution control) and the market (i.e., buyers of pollution
permits) decides the price of pollution because the price is changing to achieve the target
for pollution control. As a result, it is asserted that the system of TPP generates more
flexible incentives for pollution reduction than pollution charges.

An advantage of TPP is its use as a cost-saving instrument. hr order to supplement
the inefficiency and inflexibility of CAC, TPP was introduced in the area of pollution
control. The empirical studies of air pollution conducted by Tietenberg (1990) found that
TPP is least costly compared to CAC and PC.

However, TPP also possesses several weaknesses. Disadvantages of TPP are
summarized as follows: the absence of the market including extemality and public goods,
asymmetric information, the moral problem of pollution permits, market failure, and
inappropriateness of complete reliance on markets and market instruments (Perman, Ma,
& McGilvray, 1996). In particular, the large transaction costs for information discovery
in the system of TPP hinders trading (Batie & Ervin, 1997).

The difficulty of designing and implementing TPP is another disadvantage. The
more flexible and complex the instrument is, the more it costs in terms of designing and
implementing it. Tietenberg (1998) stated that developing countries are incapable of
handling adequately the burden of designing, implementing, monitoring, and enforcing

an effective pollution control system. People, industry, and government in most

17

 

-
'
53" fl“ - ‘L “
db" '? ’
. q. "
ra-fl " a a
a . . u _-

- . -.~nO-;'~.,'e v;
~us. bk‘.ls4a‘l5ss ‘3‘ A

TEES. Some em:

I
New - ‘ ~ .
c313) and sad 1

arseret‘im o! the

 

 

developing countries have also little experience with the market for pollution control.
Establishing a market culture and constructing the market system for trading the pollution
permits in developing countries would take a long time and be very costly.
Deposit-Refund Systems. Deposit-refund systems provide incentives for
consumers to recycle or properly dispose potential pollutants or reusable resources (Batie
& Ervin, 1997). Under these systems, consumers pay a deposit when they buy an item
such as beer or bottled beverages and then receive a refund when they return the empty
beer containers or bottles. Bottle-bill programs as deposit-refund systems are prevalent in
the US. Some examples of deposit-refund systems include automobile tires, lead-acid
batteries, and used motor oil. Buyers of pesticides pay an additional fee that is refimded
on the return of the container to a designated disposal or recycling site. The deposit-
refund systems in Japan require consumers to return household appliances including

television sets, refrigerators, washing machines and air conditioners for recycling.

Voluntary Incentive Program

Voluntary incentive program (VIP) is a business-led environmental self-
regulation, driven by existing or anticipated legislation and consumer demand (Batie,
1997). The 33/50 program is the first voluntary “Industrial Toxics Emissions Reduction
Program” initiated by the US. EPA in 1991. Under this program, participating
corporations agreed to reduce emissions of seventeen priority chemicals by 33 percent
through mid-1992 and to achieve a further 50 percent reduction by 1995. The Green
Lights Program was launched in January 1991 to encourage corporations to install

energy-efficient lighting. The Common Sense Initiative (CSI) was started in 1994 as an

18

I
‘0- fl‘ "1 3-. ‘ ‘ '
buy—“w" .L \,. . . . .

’. :7." 'OP 'vn -
‘~-.‘;~‘~ s-wcb

”a-.. no" '3' V. ~

" .

m'..-...r..a..uo '.x u.
.

:czsx'rzors to \t ..

3'"-'“.nw-,3v-‘ ‘ -'
Nu -‘Cnh.-‘h;nsdd >i~1rN

I .0".U I 0
r. m: to mgr. ::

air ‘ '
Lane is to be a;

A

‘1’).W‘.
,.....Incnt \ 0'. art

31°77? ~
~ .56! tompdr‘ “

.
“L5

REE its man‘s

0"
‘- \S

tt-nnitment to the I

”aim and Safety, an

J

”'t

“1‘24
“MS and entirr

II

mnpetitit'e inceht"

Penalties 21nd 1635 r“

media

 

lS Par)E
1‘?!

Van .
nits lei els of 2

Th

5 tarr'
é€i for p01}

 

attempt to obtain industry participation in developing standards for the industry as a
whole rather than continuing a pollutant-by—pollutant approach to protecting the
environment (EPA, 1997). Project XL (Environmental Excellence and Leadership) was
established to give exemplary individual corporations greater flexibility in achieving
environmental goals. Project XL focuses on a facility and community, and it encouraged
corporations to voluntarily design their own best ways for compliance with
environmental standards (Batie, 1997; Hoffman, 2000 a).

VIP specifies the quantity of pollution but not a certain technology as the method
by which to reach the standard. It usually does not dictate how the environmental
objective is to be achieved so that VIP generates the flexible incentives for corporations.
A prominent voluntary incentive program is “Responsible Care,” which has numerous
member companies that account for the basic chemical production in the US and
Canada. Its member companies agreed voluntarily to the guiding principles that require a
commitment to the public’s right to know, process safety, pollution prevention, employee
health and safety, and product stewardship because voluntary compliances to pollution
standards and environmentally safe practices for environmental protection provide
competitive incentives for their businesses such as good image or reputation, less
penalties and less regulatory sanctions, less attacks from environmental NGOs, and mass
media.

VIP is implemented on the basis of the cooperation and partnerships among
various levels of government, environmental advocates, community and corporations.

The target for pollution control is determined by government while the compliance

19

. «H .
‘fifi ‘ ‘ a u
v. r ". -

information Orie;
The idea of
msmrmental re
my Understood ii
‘the. hi1
at .u "
Cute COIZDI
CENT“
.vnmental .\'GC

lite" .
nat'let stem to p

4
ME.“ ° '
«8~ then tradzr

 

 

schedule and technology are determined by the negotiation among government,
corporations, and community.

A disadvantage of VIP is that it requires a capability for implementation. That is,
high levels of institutional capacity and human resource are required to implement
flexible systems in an effective and low cost manner (Batie & Ervin, 1997). Successful
implementation depends upon clear performance objectives (e.g., standards), a clear
negotiation process and the credibility of participants such as credibility of government
agencies and environmental advocates because administrative discretionary power
sometimes leaves to negativity among cooperating participants in negotiation processes

(Batie & Ervin, 1997; Hoffman, 2000 a).

Information Oriented Approach to Pollution Control

The idea of an information disclosure approach to pollution control is simple.
Environmental regulatory authority, such as governments, only provides reliable and
easily understood information of corporate environmental performance to social actors
who influence corporate environmental practices. Then, social actors such as
environmental NGOs, the media, the court, community or grassroots organizations, and
the market start to press corporations to adopt environmentally safe practices and to
change their traditional business norms and strategies, which are based on profit

maximizing and unlimited resource use.

20

5 "‘
UL" :7... .5
. I. C
D -‘e ‘ ‘
__,). \J.l' t... ' ~

‘3’" '- ~
EAA ‘ arOolfi. ;

35:11: order I -. pr

v».

”'33? a
Alt.“ 6),“
9- Corporal:

 

 

Drivers of Corporate Environmental Performance

Governments and social organizations are two primary constituents that have
historically driven corporate environmental performance. Governments have regulated
corporate environmental practices through environmental laws. Social organizations,
especially environmental non-govemmental organizations (N GOs) have mobilized
diverse forms of social protests that can have a negative impact on a corporate reputation
and performance.

Environmental NGOs have used social sanctions such as protests and negative
press in order to pressure corporations to adopt more environmentally friendly
technology and to reduce pollution emission. Governments have done it by using legal
sanctions such as civil, administrative, and criminal penalties. In terms of business
management, environmental issues are structured as fundamentally external to corporate
interests. Corporations do not need to initiate environmentally safe performances
voluntarily unless the governments force them or environmental organizations damage
their image or reputation. However, these legal and social sanctions gradually force
corporations to change their dominant business norms (i.e., individualistic self-interested
profit seeking and resource utilizing beliefs) and to transform environmentalism from
something external to the market system into something that is central to the core
objectives of corporate business (i.e., internalizing environmental costs) (Hoffinan, 2000
b).

Hoffman (2000 a) breaks down these gOvemmental and social drivers into five
categories influencing corporate behavior and managerial decisions related to

environmental protection: regulatory, international, resource, market, and social drivers.

21

 

rum 9"). ‘. "x
"ANNIE...“ L\).
U

5'
"F2. ’L’aa‘g ' ~‘\
«A .mnflu'l.) \UIT

~ I. -“ I n.
«3103716313; 111.2

Mme-Know At

 

Sissies of\\‘ild r,

     
 

D it: 9

wit, the Ozone
.\' .

10“71791118 of Ha;

"Simmons;

Regulatory Drivers. A regulatory driver is the governmental environmental
actions that require corporations to adopt environmentally safe practices. These actions
are authorized by law on the assumption that without regulatory enforcement,
corporations would not voluntarily pursue environmental protection. Regulatory
government actions evolved into several environmental policy formats in the US: 1)
Command-and-control regulation that dictates the environmental standards on corporate
performance; 2) Market-based instruments that appeal to corporate profit-maximizing
motivation; 3) Voluntary incentive programs that seek cooperative partnership among
government, corporations, and environmental organizations; 4) Criminal enforcement
that threatens corporations by extreme penalties; and 5) Forcing disclosure of
environmental information to the public such as the Emergency Planning and Community
Right-to-Know Act in 1986 and the Toxic Release Inventory (Hoffman, 2000 a).

International Drivers. International drivers are defined as international
environmental agreements and international environmental standards related to products
and international trade agreements (Hoffman, 2000 a). As the globalization of
environmental issues has considerable implications for corporate management, the
influence of international drivers on corporate environmental performance is increasing.

International environmental concerns have primarily evolved into international
environmental agreements such as the Convention on International Trade in Endangered
Species of Wild Fauna and Flora (1973), the Montreal Protocol on Substances That
Deplete the Ozone Layer (1987), the Basel Convention on the Control of Trans-boundary
Movements of Hazardous Wastes and Their Disposal (1989), the International Tropical

Agreement (1983), the International Code of Conduct on the Distribution and Use of

22

manure em :.'.‘=.”.'
sec: or a set 0: x -.
3473;337:6111 and t
:20 commie n2...-
ecpxaton's 9:12;:
restarts. Corp, -;
:‘i‘. I , ‘ . 3‘ A
-.t .0 00.1er lb!)
.2'~l. k
>4... Ls the Eur-op;
Inertiatzon,
!. 7'77“ ‘
Mandate. The t
‘ i
CI ’5: . - _
m Eowming 1::
g .

(hams-on Wilt

 

 

Pesticides (1985), and the Kyoto Protocol to the United Nations Framework Convention
on Climate Change (1997).

International environmental standards for products such as ISO 14000 have driven
corporate environmental performance at the international level. ISO 14000 is a private
code or a set of voluntary standards to provide a common approach to environmental
management and eco-labeling of products and to integrate environmental responsibility
into corporate management procedures. The ISO 14000 program addresses a
corporation’s entire range of activities, from product design, planning or training, and
operations. Corporations are required to adopt the international environmental standards
and to obtain ISO certification in order to do business in certain multinational markets,
such as the European Union (Batie, 1997).

International trade agreements also influence corporate environmental
performance. The General Agreement on Tariffs and Trade (GATT) is the oldest example
of the governing institutions for most international trade. Others include the World Trade
Organization (WTO) that replaced GATT, the European Union (EU) that has explicit
power to enforce environmental regulations throughout the entire European Union, and
the North American Free Trade Agreement (NAFTA) that was constructed to facilitate
free trade between Mexico, Canada, and the United States. These agreements adopted
environmental clauses that restricte corporate polluting behaviors to promote
international environmental protection.1

Resource Drivers. Buyers, suppliers, shareholders, banks, insurance companies

and investors are defined as resource drivers for corporate environmental practice

 

' There is a strong assertion that environmental protection is incompatible with tree trade in global free
trade market (Soros, 1998; Mellor, 1993). According to this assertion, international trade agreements do not
drive corporations to do environmentally safe performance.

23

 

... en time tllor‘r’
Parasites for cont:
Mes. contractors be
one oitheir clients ‘
is sulipiier corpora:
so:

dam Were mor

Steeter, 1999 ).

 

Insurance 0|
i‘liiCi gas egcaped

Pestie

ide Plant in E

 

(Hoffman, 2000 a). Buyers control the consumption of products while suppliers control
the acquisition of raw materials. Institutions or groups that control the acquisition of raw
materials and the consumption of products can be highly influential in the way a
corporation performs its operation. Shareholders, banks, insurance companies and
investors started to apply environmental criteria to minimize risk in their investments. As
environmentalism is translated into a core business concern of resource acquisition,
processing, and sales, the influence of these resource drivers on corporate environmental
performance is increasing (Hoffman, 2000 a).

Some buyers and suppliers have developed their own environmental principles
and started to require their contractors to meet the environmental principles. For example,
Levi Strauss & Company has developed strict sourcing guidelines for selecting
contractors that are different than the traditional guidelines about price, quality, and
delivery time (Hoffman, 2000 b). It offered generous timetables, loans, and volume
guarantees for contractors who met Levi Strauss & Company’s environmental principles.
Most contractors believed that meeting these requirements and having Levi Strauss as
one of their clients was helpful for their business (Katz, 1994). In 1998, Nike required all
its supplier corporations to comply with US. air pollution standards even if the US.
standards were more stringent than their domestic air pollution standards (Goodman &
Streeter, 1999).

Insurance companies equate corporate environmentally risky operations with
increased cost of financial risk. Approximately forty-five tons of methyl isocyanate
(MIC) gas escaped from two underground storage tanks at a Union Carbide Corporation

pesticide plant in Bhopal India in 1989. Two thousand people died and another 300,000

24

' -
t _i) .... —,q q
-_\ I'EE—QDKIA ts“

‘.
t

:9» 1 9 .3 y... or
«LI—Be 32‘ Lll‘ a. "

mrszceratzon of 5

serious about the 1

Ll?“" _ I
"45 Shcn as Sago.

Cafldtla and 316 EU!

 

 

had been injured by escaped MIC gas. “The scope of the accident and the exposure of
Union Carbide’s insurance underwriters served to alter the structure of insurance liability
coverage” (Hoffman, 2000 a, p. 74). UNI Storebrand, a large Norwegian insurance
company, refused coverage to companies that failed to assume environmental
responsibilities (Deutsch, 1998). In November 1995, the insurance industry developed a
UNEP-supported Statement of Environmental Commitment with 78 official signatories to
include the environment as one of the value-drivers in their under writing decisions
(Hoffman, 2000 b).

Banks are beginning to consider corporate environmental performance in their
lending decisions to reduce the environmental liability risk. The environmental
consideration of banks has been triggered by increasing court cases in which banks have
been held liable for environmental performance of their borrowers. According to
Hoffman (2000 a), the Exxon Valdez Oil Spill disaster which occurred in 1989 altered
notions about the limits of corporate financial liability for environmental accidents. Some
banks such as Salomon Inc., the Bank of America Corporations, the Royal Bank of
Canada and the European Bank for Reconstruction and Development (EBRD) developed
a set of environmental operating principles and are beginning to examine the practices of
the applicant and to consider poor environmental performance as a financial risk
(Hoffman, 2000 a).

Shareholders and investors are also beginning to make decisions for their
financial investment based on the data or studies about corporate environmental

performance since the late 19803 because they know that there is a positive correlation

25

‘V ~»‘ 0" 'flfififl‘
.hvabvca 548‘ .o ~. .... .‘

I .‘ .
qu'v I‘fi' .1 'F' N .’-
-zbn.>~*¥ “.‘KU:
.

... - 4;" it‘s .“ '
5.-.}. Us‘;-\':“L' N
w vwawafi‘" ' ‘
un'uLaha Base. 1‘ 5:8

.‘-ka ‘ h .
3}?.¢...\ tlrottrt.

=kERESl “as 31.x) 1

(II
N
3,.
r:
(4.
r-1
1’
J
‘

prone a set of ter.
0. maestors. Accort
esteemed that

com vi ~
f» SHOP 4 ."
:5 1m. III:

..o:.."nan. 1996. p

will 19961. Ben

 

if.“

1:» .
M the Sinsinaw.

Star .
es Bolts-m Gu

OCCldemaI

P€ll0i{-_

 

 

between environmental and economic performance.2 Shareholders and investors used
their power through shareholder voting or by directing capital investment. For example,
in June, 1988 the Social Investment Forum, a collection of socially oriented investment
groups, developed criteria by which investors could assess the sufficiency of a corporate
environmental practices in order to direct capital and resources toward those that behaved
responsibly (Hoffman, 1996). The Council for Environmentally Responsible Economics
(CERES) was also formed in 1989 as a coalition between socially responsible investors
(SRIs) and representatives of several prominent environmental organizations in order to
provide a set of ten guidelines for environmentally responsible behaviors to shareholders
or investors. According to an early CERES mission statement, “SRIs advisers are
concerned that environmentally unsound practices will undermine the economic health of
corporations and therefore reduce the risk-adjusted return of investments in that firm”
(Hoffman, 1996, p. 53). The CERES have solicited corporations to endorse the principles
(Nash, 1996). Even without such outside pressures, shareholders have exerted
environmental pressures on the companies in which they own stock. For example, in June
1999 the Sinsinawa Dominican Sisters (a religious order with operatives in the United
States, Bolivia, Guatemala, and Trinidad) used their 100 shares of stock to force the

Occidental Petroleum Corporation to reconsider its oil exploration program (Waldman,

 

2 “Research by Innovest identified a number of industries in which companies that rate higher on its scale
of environmental performance produced better returns for stock holders than did less environmentally
conscious competitors” (Deutsch, 1998, EU 7). [CF Kaiser International, Inc. found that companies with
high scores on enviromnental criteria were considered as being less risky for investments and would thus
enjoy a lower cost of capital and ultimately a higher stock price (Feldman, Soyka & Ameer, 1996 in
Hoffman, 2000 a, p. 79). Cohen, Fenn and Nairnon (1995) found that good environmental performers also
tend to be good financial performers. Dasgupta, Laplante and Mamingi (1998) claimed that a high level of
pollution intensity may be a signal to investors the inefficiency of the firm’s production process.

26

.'.' . .. -
Acct-ran to Rust

5

3a: :13: a cor?“

., .131
sat-i is} where con.
Mentsaid the) u
ESDOHSC to a July 1

tempts} 's emimr:

The dema"

green products is ‘0'
Irradiation. S)mhet.
hctnones. Sales of

bison) to 1994 (T;

gailons) to 1995 t"

 

meltNd from 195

 

1999). Thus, shareholders and investors appear as powerful forces for environmentally
safe performance of corporations since the late 19805 (Hoffman, 2000 b).

Market Drivers. Consumers include environmental concerns in their purchasing
decisions. The environmental behavior of consumerism has been termed “green
consumerism.” Green consumerism is an outcome of public awareness of environmental
problems and of support for environmental protection. The advocates for green
consumerism believe that products should be more environmentally accountable in use.
According to Rosendahl’s (1990) findings from a 1989 survey, 77 percent of Americans
said that a company’s environmental reputation affected what they bought, 89 percent
said they were concerned about the environmental impact of products purchased, and 78
percent said they were willing to pay more for recyclable or biodegradable packaging. In
response to a July 1989 survey, 77 percent of Americans answered that they consider a
company’s environmental reputation when they made purchases (Krupp, 1990).

The demand for green products is continuously growing. One example of such
green products is organic foods that are free from artificial preservatives, coloring,
irradiation, synthetic pesticides, fungicides, ripening agents, fumigants, and growth
hormones. Sales of organic foods doubled in the five years period from 1989 ($ 3.9
billion) to 1994 (7.6 billion). Sales of bottled water tripled from 1984 (933 million
gallons) to 1995 (2.87 billion gallons). Health-food supermarkets across the US.
increased from 195 stores in 1991 to 650 stores in 1994 (Burros, 1996).

Organizational expression of green consumerism is a boycott to reject the
purchase of particular corporations’ products. For example, consumers boycotted Burger

King in order to pressure the company to modify its purchase of imported beef from

27

{3:21. Amend. L

. -
. 9‘7.‘. 1... "'3\-
56%-”; .0 b5.\ . »_ ...

Etta: receix ct. ox c
~~~4m<a

“‘FI“F 5
b\.bb~o$~ ..sku. h

.""”m 1 #-
CO“:H-al’~rb ‘

COW-\‘Tatior‘

“x

L031 COIIS’JTEL’TS

\

r
f l s
— —.—

 

 

Central America, claiming that foreign production of beef for fast-food burgers was
leading to deforestation practices in countries such as Costa Rica. In another example,
Exxon received over 20,000 out up Exxon credit cards after the Alaskan Valdez oil spill
incidence which occurred in 1989 (Batie, 1997).

Consumers also influence corporate environmental performance by participating
as shareholders or investors in corporations and actively forcing the corporations to
change their business practices. For example, following the Valdez oil spill,
environmental resolutions appeared on many stockholders’ ballots and were introduced at
corporate annual meetings (Batie, 1997).

Corporations have responded to attacks or demands from green consumers by
obtaining coo—labeling or green certification, which several organizations established to
inform consumers of which products passed their environmental product standards. Eco-
label are “seals of approval” certified by either public or private organizations. They
provide chances for businesses to advertise their existence to consumers. Most the
European countries have adopted eco-labeling programs. The oldest program is
Germany’s Blue Angel seal that was established in 1988, and now is applied to over
3,500 products (Batie, 1997). Some of the more prominent certification programs include
the Green Seal program in the United States (1990), the European Union Eco-label
program (1992), Taiwan’s Green Mark program (1992), the Korean Environmental
Labelling Association’s Eco-Mark program (1992), the Eco-Market Program of the Japan
Environment Association (1989), Environmental Choice New Zealand (1992), and so

forth (Hoffman, 2000 a). Although these programs are designed to encourage existing

28

-'I-JQ" . \"'
:3“ "is-rum U -.»
a; .19
“ALMS.

 

BLAZE ’3?!) EYOI’JIX‘;

....~...~O-§“- .‘ i

l a ’ '

3...-..»116. ofi¥-\\
...

n- n ‘n - 0cm I 9

I. . ' - - h

5b....L mLBi-ALAAb. 5;:

KL-u a; L " - '

b.‘.55 It": “ d\ £11

pit-tectzon triot‘t‘rt‘..-
Em 1:0:

mm emironment.

 

‘4 IO “ork [0"C1

green markets, corporations have attempted to use green certification to create new
markets.

Social Drivers. Social drivers influencing corporate environmental performance
include environmental NGOs, the community, the press, courts, academia and religious
institutions. These social drivers that are constituents in the social system can mobilize
public sentiment, alter accepted social norms, attitudes, and beliefs, and consequently
change the way of thinking about the role of corporations related to environmental
protection (Hoffman, 2000 a).

Enviromnental NGOs are the most prominent actors pressing corporations to
adopt environmentally safe practices. They have employed diverse methods, strategies
and channels to press corporations to reduce pollution emissions. They undertake
scientific research and conduct public protests such as the boycott of Mitsubishi Motor
Sales of America, Inc. by the Rainforest Action Network in 1993 (McCoy, 1998).
Environmental NGOs engage in alliances between business leaders and NGO activists in
order to work together cooperatively to find solutions that make both economic and
environmental sense. Environmental NGOs have funded technological and market
innovation projects. They also influence press coverage, public opinion, and politics.
Environmental NGOs initiated environmental lawsuits as plaintiff against both the US.
Environmental Protection Agency (EPA) and industry.

Local communities were aware of the negative impact of local pollution and
began pressing local industry to adopt environmentally safe practices. Some communities
have participated in the process of assessment, identifying problems, planning, and

implementing corrective programs related to local industry pollution. They also

29

.0-
J ‘.I .,

- _...~ 9 .( ..-
'C’C...L- m‘.-

I "

I
"Y“ W”"- ‘
m‘thmktnhdfl II“N

2:93:11 to DC“ 5 ed;
*t‘mer. 1995). F

We Canal d: 59;"

.‘~

1071; in 19’81 on t

Court dec:

5
”by? ‘ - - ,.
«are Inleld‘u’dil

mere environment;

1.... . .
"4° befn 1111(1er b)

 

 

 

 

conducted negotiations for pollution reduction among local corporations, community
representatives and government. A form of community pressure on local industrial
pollution is the environmental movement of grassroots organizations such as the
“Environmental Justice’” movement in the US.

The news media have played a critical role in raising environmental concern in
both the public and industry. One study found that media coverage has a strong effect on
the agenda setting of the public and attitude of policy makers about toxic waste issues
(Protess, Cook, Curtin, Gordon, Leff, McCombs & Miller, 1987). The radical
environmental movement group, Green Peace has developed diverse media strategies to
appeal to news editors and thereby attract public attention and sway public opinion
(Wapner, 1995). For another example, the media were strongly influential in placing the
Love Canal disaster (i.e., toxic waste dumps near residential areas in Love Canal, New
York in 1978) on the national agenda (Hoffinan, 1993).

Court decisions have forced corporations to change traditional business attributes
that are individualistic, self-interested, profit seeking, and resource utilizing, and to adopt
more environmentally friendly business strategies and product procedures. A person who
has been injured by environmentally hazardous substances can pursue common—law
remedies through four types of legal action: trespass, nuisance, negligence, and liability.
Environmental lawsuit cases have been growing continuously in the US. since 1982.

When the Exxon Valdez oil tanker spilled 10.8 million gallons of crude oil in Alaska in

 

3 Environmental injustice is defined as an unfair distribution or discriminatory allocation of environmental
hazards on minorities in terms of race and class such as African Americans, Native Americans, and Latinos
who are mostly poor as well as poor white people. The disproportionately concentrated pollutants on
minority classes and races raised the issue of environmental justice and the corrective activities are called
the environmental justice movement (Bryant, 1995; Bullard, 1993; Hofrichter, 1993).

30

1989, the court forced the Exxon Corporation to pay more than S 5 billion in legal
settlements. After this court’s decision, most oil companies started to look carefully at the
safety records of their oil transport procedures and invested greater attention to the
prevention of environmental incidents (Hoffinan, 2000 3). Therefore, the court is also a
powerful actor driving corporate environmental performances.

Academia is an indirect actor driving corporate environmental performances.
Academic study on environmental issues and concern for environmental education are
growing (Hoffman, 2000 a). Academia has provided basic and fiindamental ideas,
knowledge, technologies, methods, human resources, and education needed for effective
corporate environmental management (Hoffman, 2000 a).

Religious institutions change the values and norms of business and corporate
managers or workers related to environmental protection because religious institutions
help alter values, beliefs and morals at both individual and societal levels (Hoffman, 2000
a). Christianity is changing its religious context from humankinds’ superiority over nature
to humankinds’ stewardship for nature. Christian maltreatment of the environment comes
from an interpretation of the Bible4 that has resulted in a relationship to nature called
“Christian arrogance” (White, 1967). Giving up Christian arrogance, religious institutions
begin to acknowledge the intrinsic value of nature and to adopt more environmentally
fiiendly doctrines. For example, the Presbyterian Church placed environmental concerns
directly into the church canon in 1991 (Associated Press, 1991). In 1994, the Roman
Catholic Church equated environmental degradation with the thefi from future

generations (Woodward & Nordland, 1992). His All Holiness Ecumenical Patriarch

 

’ “Fill the earth and subdue it. Rule over the fish of the sea and the birds of the air and over every living
creature that moves on the groun ” (Genesis, 1:28).

31

. .
" “" '-

‘ I
P-nrv .a'.
JEN-1L,“

CP""""‘-
“'-“«|a

a4..

S" ~vsgw.’
bo.‘ _ .__”

idea 01 L'L.
32“: -

Bartholomew 1, spiritual leader of the world’s 300 million Orthodox Christians, linked
specific ecological problems with sinful behavior in 1997 (Stammer, 1997). The religious
institutions’ adoption of environmentally friendly context changes societal attitudes,
values and morals about nature, changes beliefs and worldviews about the relationship
between human beings and nature, and consequently alters corporate business morals and
behaviors, and pressures corporations to take responsibility for protecting the
environment. Therefore, religious institutions are indirect social actors driving corporate

environmental performance.

Critical Players of Public Information Disclosure5

Different from the command-and-control approach and the market-based
instruments for pollution control, the information oriented approach emphasizes the
importance of the roles of the community and the market as pollution regulatory agents,
in addition to the traditional role of government. The govemment has played its
traditional roles of setting and enforcing rules of environmental behavior. However, the
idea of the information oriented approach focuses on the links of interaction among four
agents as sources of pressure on polluters to improve their environmental performance:
plants, government, community and the market (Afsah, Laplante & Wheeler, 1996).

The local community is a key force influencing corporate environmental
performances. A study examining plant pollution in four Asian countries revealed that

community pressure played an important role in some factories having high levels of

 

5 Public Information Disclosure is defined as a idea for pollution control that seeks to disclose information
of corporate environmental performance to the public in terms of pollution control tool or strategies. In this
paper, the public information disclosure is also referred to the Information Oriented Approach.

32

abatement (Hettige, Hug, Pargal & Wheeler, 1996).6 Recent evidence from Asia, Latin
America and North America suggests that neighboring communities can have a powerful
influence on factories’ environmental performance.7 “Factories negotiate directly with
local communities, responding to social norms and /or explicit or implicit threats of
social, political or physical sanctions if they fail to reduce the damages caused by their

emissions” (Afsah, Laplante & Wheeler, 1996, p. 6). Thus, the community appears to be
an informal regulator for pollution control because it forces local corporations to
voluntarily reduce their pollution emissions. The community pressures concerning
corporate pollution reduction are produced by activities of environmental NGOs,
religious institutions, social organizations, community leaders, the news media, the .
courts, academia or politicians (Hoffman, 2000 a).

The market is another key force affecting corporate environmental performance,
in addition to the government and the community. Several studies exploring the
relationship between capital market and plant environmental behavior in the US provide
evidence to support the argument that financial markets provide strong incentives for
corporations to reduce voluntarily their pollution.8 A study examining the relationship
between pollution and capital markets in developing countries found that capital markets

responded positively (e. g., increase in the corporate values) to the announcement of good

 

6 Additional evidence from Asia can be found in the following papers: Pargal & Wheeler, 1996; Huq &
Wheeler, 1993.

7 Refer to Huq & Wheeler, 1993; Hettige, Hug, Pargal & Wheeler, 1996; Pargal & Wheeler, 1996;
Hartman, Huq & Wheeler, 1997; Dasgupta, Lucas & Wheeler, 1997.

Evidence related to market power for pollution management can be found in the following papers:
Muoghalu & Robison , 1990; Lanoie & Laplante, 1994; Hamilton, 1995; Arora & Cason, 1996; Badrinath
& Bolster, 1996; Khanna & Quimio, 1997; Konar & Cohen, 1997; Dasgupta, Laplante & Mamingi, 1998;
Lanoie, Laplante & Roy, 1997; Fouloan, Lanoi & Laplante, 1999; Hoffman, 2000 a; Hoffman, 2000 b. For
example, Konar & Cohen (1997)’s study found that corporations that received the largest stock price
declined immediately following the release of the Toxic Release Inventory to the public, and subsequently
reduced their emissions more than their industry peers.

33

l
i _ . v ‘
.«r ‘ .. “u‘ 1 La
BkJu‘l“ ' “-

u
i i V.
i

f..rbsu

' .
”c'q'qp-W_1“O 1:
Hr r. Vinny-nu:

rim of Loaf
statements] L
:tjwrats em in
mute repaid
art: the} have 1
more internal ca.
f‘il’ieeler. 199~ i
Carrol adds a n:-
33d enforcing

p0

[1111113110115 and

PUbilC Inf]

 

conduct awards and negatively (e. g., decrease in the corporate value) to citizens’
complaints (Dasgupta, Laplante & Mamingi, 1998).9 Thus, it appears that markets can
pressure corporations to voluntarily reduce pollution.

The role of a regulatory authority in the scheme of the information approach for
pollution control is to provide information of corporate environmental performance to the
public (i.e., the international, resource, market, and social drivers of corporate
environmental performance). Information is a critical actor in the context of the public
information disclosure strategy. Information empowers communities to participate in the
process of local industry pollution control and provides markets with reliable
environmental data or knowledge related to corporations. Negative information about
corporate environmental performance embarrasses polluting corporations and hurts
corporate reputations. Thus, regulators could manage pollution more cost-effectively
once they have 1) high-quality information, 2) more integrated information systems, 3)
more internal capacity for priority-setting, and 4) especially stronger public participation
(Wheeler, 1997). Thus, the idea of disclosing environmental information for pollution
control adds a new role for government in addition to the traditional role of establishing

and enforcing pollution standards or charges.

Limitations and Advantages of Public Information Disclosure
Public Information Disclosure (PID) offers strong advantages for pollution
control, however, PID also has some limitations for pollution control even though

shortcomings of PID have not been formally reported.

 

9 See Wheeler, 1999, P 62. It summarizes the findings related to stock value changes caused by
environmental information.

34

”Xi'an”,
5.; ....’

7‘
'i":
A)
[1

u- ¢
i.“ “3.3 pv
u-s.b..

' ”'Aln-V‘
‘Ll‘..“‘.

,
.L‘

.
Q” :4
f ‘32:)“.

.
“~v“ 2 -l
““k‘iuh \.
I; I .
1.: 4 a
o. e

Limitations of PID

PID would not be workable in under-developed countries. For PID to work
effectively, at least three conditions need to be satisfied. First, governments should have
an internal capacity for collecting credible and hi gh-quality information about corporate
environmental performance, an integrated information system, a capacity for priority-
setting and a capacity for translating raw information into easily readable information
(Wheeler, 1997). Especially, pollution control authorities of government, which disclose
environmental information of corporations, should retain credibility. If the authorities are
perceived as ‘not credible,’ the public will not accept the information so that the
information will not work for pollution control. Second, publics must have a sufficient
level of environmental attitude or consciousness that leads to public participation in
environmental protection because PID requires strong public participation for its success
(Wheeler, 1997). Even though a well-developed stock or product market system affecting
corporate environmental performances does not exist, PID could work effectively
because where formal regulators were absent or ineffective, the informed communities
with CEPI directly pressured neighborhood corporations to reduce pollution (Pargal &
Wheeler, 1996; Blackman & Bannister, 1998). That is, PID requires strong public
participation for its success. Third, social organizations and media need to be well-
developed for disseminating information to the public. However, all three conditions
would not be easily satisfied in the under-developed countries. No PID advocates have
asserted that PII) could be effective for pollution control in under-developed counties,
even though they asserted that PID could be effective in both developing countries and

developed countries.

35

Without the consent of policy makers and industry, it is impossible for
government to adopt PID strategy for pollution control. Industry or businesses
continuously lobby policy makers to lower national environmental standards or to
weaken environmental law enforcement. Basically, polluting corporations have strong
incentives not to disclosure their environmental performance information because the
information will damage their image or reputations. Even though it is not an example of
PID, Korean govemment’s attempts to release lists of those related to the crime of
adolescent prostitution failed due to strong opposition from political leaders, then, the
first release was administrated in 2001 and the second was conducted on March 19, 2002.
Thus, PID requires, even at the minimal level, the consent for information disclosure.

PID could be a supplementary or complementary tool to CAC, while it could not
be a major pollution control tool. Different from CAC, PID is a voluntary pollution
control tool, not an enforcing tool for pollution reduction. Different from voluntary
incentive programs such as the 33/ 50 program and the Response Care program, pollution
regulatory authorities (i.e. government) can not predict the quantity and time schedule of
corporate pollution reduction in the system of PID. The effectiveness of CEPI disclosure
could be temporary because a great deal of anecdotal cases showed that public
environmental activities such as a consumers’ avoidance of buying products of polluting
corporation is mostly temporary. Thus, PID will be an ineffective and unreliable
pollution control tool in the situation where serious public and environmental health
issues exist and in the situation where pollution reduction should be rapidly achieved. In
other words, CAC will be a more reliable and better pollution control tool rather PID in

an emergency situation of environmental problems.

36

‘1 ‘M‘ ‘.I
MT). .‘..]v,

7.21: is. a x

.Q
‘. “L
Y‘ P
p‘”“-“)-A n

PID would not be effective for corporations that do not care about maintaining
their image or reputation. It is asserted that PID could work well for reputationally
sensitive companies (Wheeler & Afsah, 1996). Large corporations seem to be more
sensitive to their images or reputation for their businesses. However, it seems that some
small corporations do not care about maintaining their reputation. These small
corporations seem to have less sense of social responsibility for a clean environment.
That is, a voluntary pollution control program such as PID will not be effective for

pollution reduction of the reputationallly insensitive corporations.

Advantages of PID

Cost-Effectiveness. The first advantage of PID is cost-effectiveness, especially in
developing countries. Tietenberg (1985; 1990; 1995 b) asserted that command-and-
control (CAC) regulation is an inefficient and expensive pollution control tool. Tradable
pollution permit (TPP) has also been revealed as an expensive tool. There are large
transaction costs for information discovery in the system of TPP, and this hinders trading
(Batie & Ervin, 1997). The difficulty of designing and implementing TPP is another cost.
In the developing countries, the regulatory infrastructure for implementing Market-based
Instruments (MBI) is insufficiently developed. As a result, developing countries are
incapable of handling adequately the burden of designing, implementing, monitoring, and
enforcing an effective pollution control system (Tietenberg, 1998). Formal regulation in
developing countries also has been greatly hindered by the absence of clear and legally

binding regulations, limited institutional capacity, lack of appropriate equipment and

37

.
i ‘ a...

W‘s-\— y 5

an...- a

.
.'Fp"..jn’)'
fl 4.5 .50 \

fry-Liv 7,31.
mt L.1.0L 5M

..,i' . .
Cuffi‘t In:
‘ .

trained personnel, and inadequate information on emissions (Hettige, Hug, Pargal &
Wheeler, 1996).

A study of the cost and effectiveness of PID has not yet been conducted. As a
result, the supporting data or evidence for cost—effectiveness of PID is not available.
However, it is assumed that costs for designing, implementing, and monitoring PID
would be smaller than CAC and MBI. PID simply consists of disclosing high-quality
information about corporate environmental performance. It leaves everything to social
drivers (i.e., community and market) to control polluting behaviors. With respect to
PROPER, several teams of BAPEDAL (Indonesia’s National Pollution Control Agency)
and NIPR (World Bank’s New Ideas in Pollution Regulation) designed, implemented,
monitored and assessed PROPER with the cooperation of mass media. Cost for high-
quality information in terms of credibility, presentation, and integrated information
system are considered. Data for CEPI is mostly already available from the database of
pollution regulatory authority. The cost for information collection, aggregation and
dissemination is also decreasing (Tietenberg & Wheeler, 1998). All social drivers (e. g.,
mass media, grassroots organizations, NGOs, financial markets, products markets, courts,
etc.) are also freely available for PID implementation. Thus, PID would be a less
expensive and more easily implemented pollution control approach, compared to both
CAC and MBI.

System Stability. PII) is a systematically more stable pollution control tool than
CAC and MBI because it is a multi-agent pollution control tool. The primary agent of
CAC is the government environmental agency such as the US. EPA (Environmental

Protection Agency). The result is that effectiveness of CAC relies entirely on

38

 

" V “a 1».o

. , o .

“~-..Lus
-

u
“A. ‘ .3
'\ e b\ ..
g .
‘h‘|'.Ph A‘

government. When government does not play its role well, pollution control fails. Critical
players in the MBI approach are government and market. For example, the US. EPA and
the trading market for Pollution Permits play a key role in the system of TPP. Thus,
effectiveness of MBI relies on the government and market. If the government or the
market fails to play their role well, pollution control will not be achieved. In contrast, PID
relies on multi-agents: government, corporations, financial industry and stock market

(e. g., banks, insurance companies, share holders, investors), product market (e. g.,
consumer, big corporations of buyer and supplier), community (e. g., grassroots
organizations), mass media, NGOs, courts, and others. Thus, when government plays its
role well (i.e., providing a credible and easily understood information to the public), even
though some of key agents fail to play their roles, a variety of agents are motivated to act.
As Wheeler (2000) emphasized, where formal regulators (i.e., government’s
environmental officials) are ineffective, informed NGOs and community groups (e.g.,
religious institutions, social organizations, citizens’ movements, and politicians) may
pursue informal regulation based on convincing polluters or negotiating with them
polluters to conform to social norms. In other words, it is assumed that PID operates with
relative stability in any situation in the system of industry pollution control tools.

Public Participation. PID is a public participation-based approach and is highly
appropriate in the current trend of academic involvement and practice in public policy.
CAC employs a power-based pollution control procedures. Power is the force of
coercion. Power is defined as the ability to make others do something they are not
otherwise willing to do. CAC imposes some type of regulatory action and associated

costs upon polluters. As a public participation-based approach, PID is a tool or strategy

39

l/i

. ‘ .
5’- ~-b\ \ .

for pollution control using power of public, rather than relying solely on governments’
power. For the public to participate in the pollution control arena and to generate power
against polluters, people need to be informed (Spano, 2001). Community involvement or
public participation often starts by informing and educating people about a proposed
decision or action through press releases and public meetings because knowing is the
beginning of attitude and behavior change (McKinney, 2000). According to McKinney,
collaborative action begins on the basis of common understanding that pollution places
them at risk and in danger. A common understanding builds consensus among people.
Consensus facilitates voluntary engagement of public. Information ignites the process of
clarifying, discussing, and seeking a common understanding of environmental issues
which community faces. Thus, disclosure of information about corporate environmental
performance is assumed to ignite public’s voluntary engagement in the pollution control.

A concern about public participation in the public policy process such as
environmental or community development policy has been rising in the academic and
practical arenas. Advocates for public participation assert that the public and
communities should participate in the process of identifying and assessing problems,
planning and implementing public policies, and advocating pollution reduction among
corporations, community representatives and government. Community and public
participation can be effective in achieving sustainable improvements in environmental
conditions and health (Y accob, 1995). In conclusion, PID is a pollution control strategy
based on public participation.

Information Socicfi PID is an information-based approach for pollution control,

and thus it has the advantage of using the infrastructure of a modern information society.

40

We are living in the information and technology era and a salient feature of this era is far
greater capacity to collect, process, and disseminate information (Afsah, Laplante &
Wheeler, 1997). The role of mass media and the Internet is increasing rapidly as modern
technology progresses. Information moves through the Internet at the speed of light and is
easily transmitted to the public by mass media. The cost for information collection,
aggregation and dissemination continuously falls in this era (Tietenberg & Wheeler,
1998). Thus, disclosing and transmitting CEPI can be easy, inexpensive, and fast. PID
can fully use all the resources of this information and technology era and employ them to
help further reduce the cost of PID and increase its effectiveness. Afsah, Laplante and
Wheeler (1997) assert that regulation should change fundamentally in the new
information age and they suggest that governments should allocate fewer resources to
setting rules imposing standards of behavior, and more effort in collecting and
disseminating appropriate information. In this respect, PID is very appropriate to the
trend of our modern information society.

Ethics and Right to Know. PID would avoid an ethical criticism, that Pollution
Permit (TPP) might face it. Environmentalists associated with deep ecology criticize
utilitarian moral principles (Perman, Ma & McGilvray, 1996). The moral position of the
deep ecology is likely to be that polluting is a basically bad behavior and government
should not allow corporations to purchase permits to pollute. With TPP, government
allows undesired behaviors that can have severely negative impacts on human health and
the ecosystem. Thus, PID would be relatively more ethical than TPP, even from the

perspective of deep ecologists.

41

Communities and people have a right to know the conditions that affect them and
have a right to defend themselves against it. They have a right to participate in decisions
that affect the fundamental conditions of their lives. Thus, government should provide the
information about the conditions that affect peoples negatively. PID fulfills the ethical
responsibility for “people’s right to know.”

Enhancing Transparency of Government Environmental Protection Activities.
PID promotes transparency of government data collection or evaluation of corporate
environmental performance (Afsah, Laplante and Wheeler, 1996; Afsah, 2002) because
PID allows corporate feedback related to their data released. Disclosing CEPI opens
governments’ data to the public. When released information about corporate
enviromnental information is incorrect, the corresponding corporations can counter the
information because it damages their image or reputation. When corporations are
polluting, but government evaluates them as non-polluting corporations and the
evaluative information is released to the public, competing corporations, environmental
NGOs, or mass media can disclose the facts regarding polluting corporations. Thus, PID
allows assessment and feedback from various social groups to oversee and monitor
government environmental activities related to data collection and evaluation of corporate
environmental performance. That is, when CEPI is released to the public, the informed
public can assess and monitor government’s oversight activities. As a result, PID can
prevent corruption of public environmental officials and enhance transparency of a

government’s environmental protection activities.

42

Protons C ms 0

1 g. . ' I
553110.130 .\
iv:- «; .
“Q\.'du\e prl: :.
matinee of

A fl? A ‘ u
t0.\~.~g0fled :.
S

for
3- Bit

the

or AL. .

" '41 ;
Laniama 1: 0f;
~ L & Wiley

_ ‘.c‘.-. ."r u -

Previous Cases of Public Information Disclosure for Pollution Control
To obtain a clearer understanding about PID, this section introduces previous

studies that implemented the idea of PID to the practice of pollution control.

Program for Pollution Control, Evaluation and Rating (PROPER)10

A pilot program using a strategy of public information disclosure for pollution
abatement was performed in Indonesia in 1995. Faced with industrial expansion and a
rapidly deteriorating environment, BAPEDAL (Indonesia’s National Pollution Control
Agency) and NIPR (New Ideas in Pollution Regulation in the World Bank) initiated an
innovative program called PROPER for rating and publicly disclosing the environmental
performance of Indonesian factories. PROPER evaluated participating plants with five

color-coded “grades” indicating their compliance with pollution regulations as follows:

1. Black: Factories or business activities that apply no environmental
management effort and whose activities cause environmental degradation and
serious damage to nature and human beings.

2. Red: Factories or business activities that apply some environmental
management effort but not sufficiently to comply with the national standards
for air, water, solid and toxic wastes.

3. Blue: Factories or business activities that apply effort sufficient only to meet

the national standards for air, water, solid and toxic wastes.

\

'0 For details of PROPER, refer to Afsah, Laplante & Wheeler, 1996; Afsah & Vincent, 1997; Afsah,
Laplante & Wheeler, 1997; Wheeler, 1997.

43

BAPED

\II‘ _ J.)
all: Juli mdUs F

nitration. In (
r1329}: sources
computer progr;
treating propt‘
multiple retictt
Ytt‘tetters V\ ere
BAPEDAL (cg

State Minister c

mini Smem \\
1iSit the offices
computer prttgr
BAPED

l‘OiUniar}! pIOLq

Initiated in 193

 

4. Green: Factories or business activities that conduct environmental impact
management efforts and achieve better than the national standards for air,
water, solid and toxic wastes.

5. Gold: Factories or business activities that use the best available clean
technology; promote zero discharge of pollutants; and conduct environmental

impact management efforts with very satisfactory results.

BAPEDAL worried about mistakes that could be exposed after the color grades
had been made public because mistakes made in public could weaken credibility of
information. In order to reduce mistakes, BAPEDAL based the grading system upon
multiple sources of data, included independent inspections, developed a user friendly
computer program for analyzing the data, and designed a multiple step process for
reviewing proposed grades before disclosing them to the public. For an example of the
multiple reviewing processes, BAPEDAL employed a three-step review process. The
reviewers were 1) a special advisory board that included members from outside
BAPEDAL (e. g., the Department of Health, business associations, and NGOs), 2) the
State Minister of Environment, and 3) the President. To ensure that press reports on the
rating system were accurate, BAPEDAL arranged for officers involved in the project to
visit the offices of major local newspapers, to explain the system, and to demonstrate the
computer program.

BAPEDAL surveyed initially 350 facilities participating in PROKASIH, a semi-
voluntary program for controlling the discharge of industrial pollution in waterways

initiated in 1989, formally labeled the Clean River Program. Through the survey,

44

- “I? 5..“

...“ ’
d-

‘ .
. V ‘ _
-# ‘3; .g _

'..~~~‘
U

.

' ...-.11.)! If:

u "‘ H
.‘..

1372111165 (

and 6 53:11
wired “
is more tli

Sit

ptallc an:

1997‘). BB

all faciliti:

BAPEDAL judged that 176 facilities had sufficient data to be graded. Thus, BAPEDAL
invited facilities not included in PROKASIH to volunteer to be graded. As a result, the
initial number of participants in the grading system was 187 facilities.

BAPEDAL decided initially to release the names of only those facilities earning
green or gold grades. For the remaining facilities, it decided to initially release just the
number in each color category. Finally, it decided to allow facilities six months to
improve their environmental performance before it disclosed their names and grades.
BAPEDAL also decided not to release all the blue, red, and black facilities
simultaneously in order to sustain interest in the news. Instead, it disclosed their names
and grades industry by industry.

The Minister of Environment announced the names of five facilities graded as
“Green” in June 1995. It also disclosed the distribution of the grades, but not the
identities of the remaining 182 facilities. Most facilities, 115 facilities, obtained “Red”
and 6 facilities received “Black.” None of 187 facilities gained “Gold” and 61 facilities
received “Blue.” The percentage of facilities receiving “Blue” or “Green” was 36 %. That
is more than a third of the total.

Significant improvements in compliance status, both before and following the
public announcement, were observed. The information disclosure worked as a credible
threat of real punishment to the polluters rated as “Black” (Afsah, Laplante & Wheeler,
1997). By December 1995, when the Minister started disclosing the names and grades of
all facilities, the number of facilities graded as “Black” dropped from 6 to 3. The number
of facilities graded as “Red” fell from 115 to 108. Consequently, the number of facilities

graded as “Blue” meeting the basic requirements for compliance increased from 61 to 72.

45

LCD 5.0qu 1 I 1"
441‘ .A‘nh L\.LL..
F'm ‘ “by";
1: ‘r..):‘“;.;b

cfPRQ'PER to.

mi.

(1
(I)

mmmd

(11“?on em. .

ofa lined the

The firgz 1600

mEWNEdL

That is, in six months, PROPER raised the compliance rate from 36 % to 41 %. More
than a quarter of the facilities graded as “Red” or “Black” in December 1995 improved
their grade to “Blue” or “Green” by September 1996.

PROPER embodied a power strategy for strengthening relationships with three
important external stakeholders: NGOs, communities, and the press. These three groups
are empowered when they are informed of corporate environmental performance. A study
of PROPER found that armed with credible corporate environmental information, NGOs,
communities, and the media effectively exerted pressure on polluters to improve

performance (Afsah, Laplante & Wheeler, 1997).

Toxic Release Inventory

In 1986 the US. Congress passed the Emergency Planning and Community Right
to Know Act (EPCRA) that includes the Toxic Release Inventory (TRI) designed to
provide information to the public on the release of toxic substances into the environment.
EPCRA required manufacturing establishments with certain threshold sizes of chemical
emissions to disclose publicly the quantity and type of toxic chemicals released into the
environment. Consequently, TRI lists those corporations that use 10,000 pounds or more
of a listed chemical in a given calendar year. Corporations that import, process, and
manufacture 25,000 pounds or more of a listed chemical must file a report on each of the
chemicals in existence within the facilities if they have 10 or more full time employees.
The first reports were on July 1, 1988 for toxic emissions in the calendar year 1987. The

first public disclosure of TRI data occurred on June 19, 1989.

46

"__ .l- ‘..
o- "

...‘u - -

_____'——.-—

\t
1“
-.

K

The 30 / 50 Program

In 1991, the US. EPA initiated the 30 / 50 program, the first voluntary Industrial
Toxics Emissions Reduction Program. Under this program, participating corporations
agreed to reduce emissions of seventeen priority chemicals by 33 percent through mid-
1992 and to achieve a further 50 percent reduction by 1995. The number of participating
corporations was expanded from 555 to 5,000. Participating corporations decreased their
emissions by more than 50 percent (a total of 757 million pounds of pollutants) by 1994,
a year ahead of schedule. A primary factor attracting chemical corporations to participate
in this program was a national media campaign promoting better images for them (Arora

& Cason, 1996; Batie, 1997; Tietenburg, 1998; Hoffman, 2000 a).

Pollutant Release and Transfer Registerll

Inspired by the successes of TRI, the OECD initiated a program of Pollutant
Release and Transfer Register (PRTR). Agenda 21 of the United Nations Conference on
Environment and Development (UNCED) in Rio de J aneiro in 1992 set a specific
objective. By the turn of the century governments were to implement and improve
databases and inventories on chemical emissions. PRTR can be defined as an
environmental database or inventory of potentially harmful chemicals that are released to
air, water and soil as well as water for their treatment and disposal. PRTR uses the same
format as TRI but restricts listed chemicals to those with relatively high hazard ratings.

PRTR is currently implemented in the United Kingdom”, Egypt, the Czech Republic,

 

” Environmental NGOs such as the Environmental Defense Fund have used an Internet site (e.g.,
http://www.scorecard.org) to inform the public of the risks of chemicals and to assist communities in
identifying polluters and assessing corrrmunities’ overall pollution problems.

'2 PRTR is implemented in the name of the Cherrrical Release Inventory in the United Kingdom.

47

31:513.;

.Qd‘n"

Proious

T'
.1.
”al.— ‘ 3'
”HA? .5».

'l.’
c'La

Ft

amp}? Ct

BOD «Ba
Tiler;

- f0U;

L

”Fact or

and Mexico. The results of those programs, however, have not been reported (Wheeler,

1999).

Previous Empirical Studies
The New Ideas in Pollution Regulation (N IPR) in the World Bank provided
empirical evidence about the effectiveness of the information disclosure program named

Pollution Control, Evaluation and Rating (PROPER) initiated in Indonesia in 1995.

Effectiveness of Public Information Disclosure in Canada

In addition to evidence from PROPER in Indonesia, Foulon, Lanoie and Laplante
(1999) provided empirical evidence about the effectiveness of the information disclosure
program in Canada by comparing the existing regulatory enforcement actions.

The Ministry of Environment, Lands, and Parks of British Columbia, Canada
(MOE) publishes twice a year a list of firms that either do not comply with the existing
regulation or whose environmental performance is of concern to MOE since July 1990.
The information revealed by MOE includes the names of the firm, the location, the nature
of concern (e. g. mining operation effluent, pulp mill effluent, sawmill emissions), the
reason for which the firm is on the list, and the number of times the firm has been on the
list.

Foulon, Lanoie and Laplante conducted the empirical investigation based on a
sample covering 15 plants in the pulp and paper industry over the period 1987 — 1996.
BOD (Biological Oxygen Demand) and TSS (Total Suspended Solids) were considered.
They found that the adoption of stricter standards and higher penalties had a significant

impact on emissions levels, but the public disclosure strategy adopted by the province of

48

British Columbia (Canada) has a larger impact on both emission levels and compliance
status than orders, fines, and penalties traditionally imposed by the Ministry of the
Environment and courts. They concluded that the public disclosure of environmental

performance does create additional and strong incentives for pollution control.

Financial Market and Corporate Environmental Performance Information

Several empirical studies that define the association between stock value change
and CEPI release have been performed. They mostly focus on the effect of TRI reports on
the stock values of corporations listed in the reports.

Klassen and McLaughlin (1993) investigated the stock market reaction to positive
or negative environmental information using event study methodology. They found that
positive environmental news in the form of the announcement of an environmental award
resulted in positive excess market returns and negative news in the form of a crisis such
as an oil spill resulted in negative excess returns.

Cohen, Penn and Nairnon (1995) examined the stock market response to new
information of TRI emission on the environmental performance of individual firms, and
found that corporations with lower TRI emissions outperformed their industry
competitors in the stock market between 1987 and 1989. They also presented evidence
for the assertion that green investors need not pay a premium for their convictions and
argued “the fact that greener firms are doing as well or better than their more pollution-
intensive counterparts may be due to the fact that firms with more efficient

manufacturing processes also pollute less” (p. 14).

49

. Oh .
I arr" . .
rifts. ..... t ...~

g | I
00;»n‘ '7.

3? ”-3.
.03.“. uru‘. u

mt: tater \\
cert-rate ‘.C-\ 1.
i722 toxic rc
:ozpaics th:
mien-ration it
Striated that
Stock \‘(liUC or
“Wk price
about poliutit

Lapla
announceme;
They found a
hem“cert 1.65

[hill the Settle

“.9?an t

Hamilton (1995) found that publicly traded corporations whose TRI releases were
first reported on that date experienced statistically significant negative abnormal market
returns and 2) TRI provided news to investors and journalists. He investigated stock
values of the corporations listed in the reports released by the US. EPA in 1989 in order
to examine the reaction of journalists and stockholders to the TRI reports. Hamilton
found that the higher the pollution figures (such as air emissions or offsite shipments of
toxic waste) were in TRI reports, the more likely print journalists were to write about the
corporate toxic releases and the more likely print journalists were to write about the
firm's toxic release, and this would be more likely to affect investors’ behavior. For those
companies that reported TRI data to the EPA, the average abnormal return on the day this
information was made public was negative and statistically significant. Hamilton
estimated that corporations reporting TRI information lost on average $ 4.1 million in
stock value on the first day the data were released. The abnormal negative return effects
of stock price were reduced for corporations where investors had previous information
about pollution patterns such as companies with exposure at Superfund sites.

Laplante and Lanoie (1995) observed the reaction of the stock market to the
announcement of environmental incidents, environmental lawsuits, or suit settlements.
They found abnormal losses in stock value of Canadian-owned corporations ranging
between 1.65% and 2% when the firms were found guilty (and fines imposed) on the day
that the settlements of lawsuits were announced. No losses in stock value were
experienced when lawsuits were first initiated. They analyzed the impact of the

announcement of 47 environmental incidents, investment in emission control equipment

50

9......

nfiuonai
16732011 of :h.
Honexer.tf.
tales: of pr"
by tour-.5 at;
£02301 been
{compared tt
the costs and
hwmmr
regulators fa
1991P- 1’).
r01c: in treati
an impact or

Kmu
Mitts bac

”CUM 0n 1}

on the equity value of Canadian corporations, 9 announcements of lawsuits and 13
announcements of suit settlements in Canada, using data from 1982 to 1991.

Lanoie, Laplante and Roy (1997) provided evidence indicating that stock markets
react to the release of information based on American and Canadian data. Every six
months since July 1990, the Ministry of the Environment of British Columbia (BC,
Canada) publishes a list of polluters identified in two categories: “not complying” and “of
concern to the Ministry.” They found that appearing on the BC polluters' list had no
impact on a firm's equity value in Canada, using the SIMM (single-index market model)
version of the standard event-study technique (i.e., DAY -1, DAY 0, and DAY +1).
However, they also found that large polluters were affected more significantly from the
release of polluter lists than smaller polluters. It was also observed that penalties imposed
by courts and regulators did not provide enough incentive for investment in pollution
control because fines or penalties (as imposed by regulators and courts) were too low
(compared to pollution abatement cost) to act as effective deterrents. “Upon trading-off
the costs and benefits of pollution control, profit-maximizing firms may choose not to
invest their resources in pollution abatement since the expected penalty imposed by
regulators falls considerably short of the investment cost” (Lanoie, Laplante & Roy,
1997, p. 17). Hence, Lanoie, Laplante and Roy asserted that capital markets may play a
role in creating the incentive for pollution control. They concluded that information has
an impact on market valuation, and it affects a polluter's environmental performance.

Konar and Cohen (1997) studied an impact of CEPI disclosure on financial

markets based on the TRI data by comparing 1989 emission to 1992 levels. Their study

focused on frrrn behavior in response to a significant stock market reaction to new

51

information on toxic chemical emissions. They observed that corporations with the
largest negative stock price effects following the announcement of their TRI emissions
were found subsequently to reduce their emissions more than other firms in their industry
and also to make other significant attempts at improving their environmental performance
by reducing the number and severity of oil and chemical spills (e.g., top 40 polluting
firms subsequently reduced their emissions more than other firms.). They also found that
media attention and resultant stock price effects have more of an effect on subsequent
firm behavior than simply being known as the largest emitters in the US. It is also found
that the market reacted more to unexpected TRI disclosures than to those that were
already expected to be very large. Konar and Cohen provided a rationality for the
negative stock price effect. That is, firms high on the TRI list can be expected to spend
resources to catch up with their competitors who are not polluting as much and to defend
themselves in costly litigation. Investors who learn that a firm is high on the TRI list may
rationally react to this information by bidding down that firm's stock price. Thus,
investors use TRI emissions as a signal of the corporate productive efficiency. Financial
markets provide a strong incentive for firms to change their environmental behavior.
Thus, it can be asserted that under the efficient markets hypothesis, we can expect any
abnormal movement in stock prices to be the result of new information that changes the
expectations of the investing public about the future prospects of a firm. Konar and
Cohen argued that if the provision of this information negatively impacts the financial
performance of the firm, it will provide strong incentive to the firm to become a better
environmental actor. They concluded that providing information to the public may be an

effective remedy to reduce environmental extemalities beyond a regulatory standard.

52

' x.
L}, .L....._ L“
‘\Afi.‘.u‘rl.

“if” “"1 ~}-
‘uA' .‘..

K .
up

“0“; c

an“ 5“.“

Some studies focused on an association between corporate environmental
performance and economic performance in the stock market, not focusing on the effect of
disclosure of CEPI on stock market. The Alliance for Environmental Innovation reviewed
seventy studies and found a positive correlation between corporate environmental
performance and economic practice in stock market, meaning that corporations that
environmentally outperform their peers outperform them on the stock market by as much
as two percentage points (Hoffman, 2000 a). ICF Kaiser International, Inc. found a
similar correlation in a study of 300 of the largest public corporations in the US
(Hoffman, 2000 a, p. 79). Feldman, Soyka, and Ameer (1996) found that corporations
with high scores on environmental criteria were considered less risky for investments and
would thus enjoy a lower cost of capital and ultimately a higher stock price (Hoffman,
2000 a, p. 79).

Dasgupta, Laplante and Mamingi (1998) found that capital markets in Argentina,
Chile, Mexico, and the Philippines appeared to react to the announcement of
environmental events (news) involving publicly traded companies. They investigated the
reaction of the capital market in Argentina, Chile, Mexico, and the Philippines to
environmental news such as violation of permits, spills, court actions, citizen complaints
and protests, agreement between government and companies, investment in clean
technologies and environmental protection, the announcement of pollution abatement,
government black list of polluters, and government actions such as warnings, fines,

penalties, complaints, and company shutdowns. Their investigation proved that capital
markets react negatively (i.e., decrease corporate value) to citizens' complaints targeted at

specific firms. Markets reacted positively (i.e., increase corporate market value) to the

53

«nu-v- -.~ «- ~90“ ,xr-O

Lam J.ahbi..~ru

b

’ r -- 334 . ..
'3‘” "a“; LOT-TM

Sm lines ‘3

s

“25 intenttor

.
n.

V "ru I
liOlddOflS rE'X

announcement of rewards and explicit recognition of superior environmental
performance. Thus, Dasgupta, Laplante and Mamingi (1998) concluded that capital
markets have penalized corporations suffering from adverse environmental events, and

rewarded corporations with positive environmental news.

Effectiveness of Public Information Disclosure in Korea

The study performed by Byung-Jun J eon (1998) is the first and only study related
to Public Information Disclosure as a pollution control strategy in the Korean stock
market. J eon examined the effects of both good news and bad news about corporations on
stock values between 1992 and 1995. The good news was environmental information that
was intentionally disclosed by corporations. The bad news was about environmental
violations reported by government. He used event test and multiple regression for data
analysis. J eon found that 1) investors responded to information on environmental
violations, and 2) environment-related self-disclosure news did not have significant
effects on the stock market because of the weak reliability of information. Based on his
findings, he concluded: “the use of information by the Korean government as an
environmental policy tool already has had effects” (J eon, 1998, p. 155).

Cautious interpretation of J eon’s research is advised for the following reasons:
First, 53.5 %13 of the good news that was intentionally disclosed by corporations was
business-related data rather than environment-related data. These data could provide

information about business practices, not about corporate environmental performance.”

 

'3 Data items include “Environmental Business (42.4 %),” “Entry in Spring Water Business (9.7%),”
“Construct of golf Business (0.7%),” and “Construct of Natural Park (0.7%).” See Jeon, 1998, p103.
” Jeon’s data analysis result of t-statistics also shows that the variable of golf course did not have
significant influence on the abnormal return of stock value. Refer to Jeon, 1998, p 147.

54

In

N.
“123+.
”Vii Q

Therefore, more than half of the data may not have been useful in predicting
environmental performance. This problem could weaken the validity of his conclusions.
Second, the bad news used for J eon’s study was not easily understood information
by the public. The bad news was available only to a few people concerned in contrast to
the case of TRI. The bad news adopted for the event test in J eon’s research was not
disseminated to the Korean public by mass media or environmental and social NGO’s.
This means that the bad news used in Jeon’s research could not create the pressures
necessary to increase the potential cost incurred by corporate management, and
consequently creating incentives for polluting corporations to reduce their pollution.15
Therefore, the bad news used in J eon’s research does not seem to be valid data for

measuring what J eon intended to measure.

Product Market and Environmental Information Disclosure

Consumers could react by not buying the products of corporations with very large
emissions. So far no study directly examined the effect of CEPI disclosure on product
markets. However, some studies about tendencies of consumer behavior related to
corporate environmental information and tendencies of corporate behavior related to
green consumerism have been done.

Rosendahl (1990) found evidence supporting the existence of green consumerism

from a 1989 survey. Eighty nine percent responded that they were concerned about the

 

'5 Investors paid attention to the increase in community pressure, citizen lawsuits, product boycotts of green
consumers or environmental NGOs, and enforcement action of government caused by TRI release.
Investors assessed the potential cost that firrrrs on the TRI list were expected to incur in order to catch up
with their competitors (Konar and Cohen, 1997).

55

”i-

P~“’\‘~’.5c‘ :;

b-Kx.

.

.‘w-q—n‘ ,.
l a

w...._.\ 3

environmental impact of products purchased, and 78 percent said they were willing to
pay for more recyclable or biodegradable packaging.

Krupp (1990) also observed the same consumer tendency as Rosendahl. In
response to a July 1989 survey, 77 percent of Americans said that they consider a
company’s reputation when they buy a product.

Magat and Viscusi (1992) examined the potential effectiveness of hazard
warnings and found consumers do react to warning labels. For example, in the case of
insect spray, half of consumers stopped using insect spray. However, information used in
their study was about the product, not corporate environmental performance.

Arora and Gangopadhyay (1995) found that industries that relate more closely to
consumers were more likely to participate in voluntary pollution prevention efforts. This
finding supports the claim that consumer power in the market can be assumed to threaten
polluters.

Arora and Cason (1996) investigated participation in the EPA’s 33 / 50 to
determine the potential that such voluntary environmental regulation has for improving
environmental quality. They found that industries with greater consumer contact
participated more in this program than industries with less consumer contact.

Wheeler (1997) addressed four factors for impact of public information: citizens,
NGOs, investors, and consumer. Relying on results of the PROPER program in Indonesia
he asserted that environmental reputation does matter to corporations whose expected
costs or revenues are affected by judgments of environmental performance by customers,

suppliers, and stockholders.

56

 

In one o.
titted or. the t

' r
.'.."... -o. \ 5.
u.l‘.‘..i.&..\v:‘l, \\ ..

MCCITL'KC. l:

i

R?“ i- a i ' ‘
vaL'tha} 0\ 2:1,.
. I

In one of the most recent studies, Lanoie, Laplante and Roy’s (1997) study
focused on the capital market. Based on their findings, the authors asserted that collective
information, which compares firms with bad performance to those with good
performance, is more likely to allow consumers to switch away from the products

produced by the firms with bad environmental performance.

57

CONCEI

Assumptior

L05:
pent-writ;
reduce the:
(Pill. This

behatior:

'0‘)

(’J

As
that make
detlde the
corpora,
{Tn-”011m:

mlol‘tmc

CHAPTER [II

CONCEPTUAL FRAMEWORK OF PUBLIC INFORMATION DISCLOUSRE

Assumption for Corporate Behavior
Loss of market shares or stock values by disclosure of corporate environmental
performance information (CEPI) is asserted to be a key factor motivating corporations to
reduce their total quantity of pollution in the scheme of Public Information Disclosure
(PID). This assertion is made on the basis of the following assumptions about corporate
behavior:
1. The corporation is a profit-maximizing behavioral entity.
2. The quantity of corporate pollution is decided based on cost and benefit
analysis.
3. Potential cost from CEPI disclosure causes loss of market shares or stock
values for polluting corporations.
4. Loss of market shares or stock values motivate corporations to reduce the

pollution of polluting corporations.

As is well known in economics, the corporations are profit-maximizing entities
that make decisions based on self-interest. Corporate managers are actually the ones who
decide the amount of corporate pollution. They can change their behaviors related to
corporate pollution without changing their attitudes, beliefs, or values for the
environment. As Green and Cowden (1992) have argued, the prospect of behavioral

involvement (unlike the request for an opinion) forces people to consider cost and hence,

58

_ \L‘f“

Social .\'0ri

SOC
11537124130?
pram. in
absent or in
polluzton tl
corporation
tuironmer
tie tommu
Staked'ji

Hut

manner

institutions

In Oihel “C

\

i

v“

113i 1
:Rt‘
Htamnu
‘ “an it».

ditto

prompts self-interest reflection. Miller and Ratner (1998) asserted that the link between
self-interest and behavior is stronger than that between self-interest and attitudes.
Increase of cost because of the disclosure of CEPI forces managers to reduce the amount
of pollution to satisfy their self-interest, which is maximizing their profit, even though

they still keep the negative attitudes, beliefs, or values toward the environment.

Social Norms and Corporate Profit Maximizing Behavior

Social norms or desires also drive polluting corporations to abate pollution when
information that they are severely polluting is revealed to the public. In the PROPER
program in Indonesia, some factories were located in areas where formal regulators are
absent or ineffective. However those factories voluntarily reduced the amount of
pollution (Hettige, Hug, Pargal & Wheeler, 1996).1 It is explained that polluting
corporations complied with social norms or community desires that the community
environment should be protected and factories should not deteriorate the air and water of
the community. This sort of pollution control through social norms or community desires
is called “informal regulation” in pollution control studies.

However, the motivation of compliance to social norms also resulted from the
potential cost by violation of social norms. The potential cost of violating social norms is
driven by community grassroots organizations or environmental NGOs: local religious
institutions, social organizations, community leaders, citizens’ movements or politicians,2

in other words, the international, resource, market, and social drivers of corporate

 

' Additional evidences from Asia can be found in the following papers: Pargal & Wheeler, 1996; Huq &
Wheeler, 1993.

2 Refer to Huq & Wheeler, 1993; Hettige, Hug, Pargal & Wheeler, 1996; Pargal & Wheeler, 1996;
Hartman, Huq & Wheeler, 1997; Dasgupta, Lucas & Wheeler, 1997.

59

I
I 1"
53703730..

. . ,..
C4‘”‘t.‘.'tti.t‘..>
"r

beta. tor.

Corporate
1:11»:

' '0...
m

:0: Dre-“KM...

I

”id- Iht inc:

Sort
E3331] fem;
kmlailm
companies
Care bOUrit
P650171] an;
chemlciil ii
”Cider“ at

cmi”3mm

Shalegles
11151

little in

Texiiie M

E

environmental performance. Therefore, it appears that compliance of polluting
corporations with social norms or community desires is also a profit-maximizing

behavior.

Corporate Reputation and Environmental Information

Information about corporate environmental performance disclosed to the public
for promoting pollution control would be very effective, especially in the situation where
environmental reputation plays an important role in determining business strategies,
consumers’ purchase behavior, and investors’ financial investment decisions. Damage to
the image and reputation of polluting corporations becomes the pressure or motivation
and the incentive for pollution reduction.

Some corporations became concerned about their environmental reputations (i.e.,
green reputation) in their business strategies. For example, the Chemical Manufacturers
Association (CMA) developed a set of environmental standards for all member
companies to adopt and instituted a program, “Responsible Care,” in 1990. Responsible
Care bound its members to a set of principles designed to improve environmental
performance. The primary goal of this program was to mend the public image of the
chemical industry following events such as the disaster of Union Carbide’s pesticide gas
accident at Bhopal, India, in 1989, and eventually to improve member companies’
environmental reputation (Nash & Ehrenfeld, 1996). Similarly, programs designed to
improve environmental reputation were flourishing in other industries, such as the
Strategies for Today’s Environmental Partnership (STEP) of the American Petroleum
Institute in 1990, the Encouraging Environmental Excellence (E3) of the American

Textile Manufacturers Institute in 1992, the Enviromnental Management Program (EMP)

60

o: in: Pm:
Prat rattan
1991. etc. t

12312 area

v" Q, L '
£1.31. mhll\

bums m
"Wt' ‘

”Esme .

orlendg , .

Correlatio;

Dag-um

“1165105 .
Compmic
“Fitted
[Hat also
compilin

WORD a

‘ ”“10 t1

R313}; to L

of the Printing Industries of America, Inc. in 1992, the Great Lakes Automotive Pollution
Prevention Project (APPP) of the American Automobile Manufacturers Association in
1991, etc. (Hoffman, 2000 a). Thus, environmental reputation is becoming more. critical
in the area of business.

A critical factor encouraging corporations to make stronger efforts for improving
their environmental reputation is consumers’ environmental concerns related to corporate
business performances. It is because consumers began to consider a corporations’ green
reputation in making their purchase decision. In other words, consumers began to demand
environmentally friendly attributes in products and companies.3 Another critical factor is
financial market drivers. Financial investors including bankers, shareholders, and
insurance companies also began to consider environmental reputation in their investment
or lending decisions in order to reduce investment risks because of the possible positive
correlation between environmental performance and economic practices.4 According to
Dasgupta, Laplante, and Mamingi (1998), a high level of pollution intensity may signal to
investors the inefficiency of the firm’s production process. “For reputationally-sensitive
companies, public recognition of good or bad performance may translate to large
expected gains or losses over time. These can affect lending decisions by bankers, who
may also be concerned about legal or financial liability for polluters who are not
complying with regulations” (Wheeler, 1997, p 8). Thus, investors began to make
decisions for their financial investment based on the data about corporate environmental

performance. Environmental reputation matters for firms whose expected costs or

¥

3 Refer to the “Market Drivers” in Chapter II, p. 27.
4 Refer to the “Resource Drivers” in Chapter II, p. 23.

61

"'1, f“ 1
n
TQr-ln: lain..-

mra next re;
Ear-merit of
lilfi pUbilC Chg

Pollution red;

info:
bum,
as the Ihreat
bttauSe int}
Share and lo
C0515 “hm
litigation re

”“Smedte

revenues are affected by judgments of environmental performance by customers,
suppliers, and stockholders (Cohen, 1998).

CEPI influences the image and reputation of related corporations. Dasgupta,
Laplante, and Mamingi (1998) claimed, “If properly informed, capital markets may
provide the appropriate reputational and financial incentive” (p. 5). Similarly, Wheeler
and Afsah, (1996) stated, “Public knowledge of environmental performance has
important implications for reputaionally sensitive companies” (p. 2) and “Good PROPER
ratings will enhance business reputations with investors and consumers” (p. 4). Thus,
reliable and understood information of corporate environmental performance can create
strong new reputational incentives to polluters to move beyond compliance and toward
attainment of higher performance ratings (Wheeler, 1997). In sum, CEPI disclosures to
the public change corporate image and reputation and they would be effective for

pollution reduction, especially for reputationally sensitive companies.

Role of Information: Carrot and Stick

Information disclosure of corporate environmental performance to the public may
be seen as a “Carrot” or “Stick” approach where the carrot is the incentive and the stick
as the thl’eat of enforcement. In other words, the donkey or corporations moves forward
because information disclosure causes a decrease in revenue due to a shrinking market
Share and loss of stock value. This in turn causes an increase of corporate management
costs when boycotts by green consumers occurs, or when citizen complaints and
litigation results in stronger environmental law enforcement by government and results in

In - - '
ass media exposure of corporate envrronmental practice.

62

hr" '5‘
ll5§5

- «Luff-f :1 Cl“ '
A. *

.‘.....u
I

.R; 1.; :‘H
.15 ..sth. u I.
o

";l"\“ "

4‘; 1n-

n r I

b.~.~.bl~ £5.
1

333355 0! CI

.1 .
'I—r . .
trad. ., f EI'OI‘.

Utertompfx

p. .
u; \

"I6.
I! 1‘

W316 foils.

These potential losses and increased management costs would weaken the
corporate competitive position and eventually cause them to lose competitive standing in
the industry. That is, the new industrial competitive business environment is driven by
corporate environmental information. Therefore, for polluters, the financial loss and
increase of costs are nothing more than a stick strongly threatening their survival in their
new business environment. For green firms, the relative benefit from polluters’ financial
loss and increased costs are a carrot granting competitive advantages. Thus, it is assumed
that the stronger the firms perceive threats, the more voluntarily they will comply or
over—comply to environmental standards.

The carrot and stick principle implied in the idea of PID is visually summarized

into the following Figure3.1.

63

Figure 3.1. Public Information Disclosure as Pollution Control Tool: Carrot and Stick

Principle of Organizational Behavior Control

 

 

Environmental Information Disclosure

l

 

 

 

 

 

 

 

 

 

 

 

Public
(MARKED
Carrot (Prize) for Non-Polluting Stick (Threat) to Polluting Corporation
Corporation
Relative Benefit Loss of corporation & Brand credibility
Revenue Loss: Decrease of Market Shara
Financial Loss: Decrease of Stock Value

 

\ /

INDUSTRY: New Competitive Environment)

 

 

Threats to Survival of Polluting Corporations
Encouragement for Non-polluting corporations

 

 

 

[\

Voluntary Compliance to Environmental Standards: Polluting Corporations

 

Over-Compliance: Non-polluting Corporation
\

64

 

h“

Conceptual Fl

WT. so (

pressures for o
”.‘.., 9

... nut 2:10:
or 0: market

at stock talc:-

“m ? ii ‘1—
tt..l.01 15 L142:

C01

pit

.N
.—

.L’ J
m

Conceptual Framework of Public Information Disclosure

When Corporate Environmental Performance Information (CEPI) is disclosed to

the public in an easily understood form, the information may generate motivation5 and

pressures for profit-maXimizing corporations to comply with environmental standards.

The motivation and pressure for voluntary compliance basically results from potential

loss of market share, potential increase of cost by threats of divers social drivers, and loss

of stock value. Thus, the idea of Public Information Disclosure (PID) for pollution

control is claimed to have the following content for polluting corporations:

1.

CEPI can turn consumer attitude in a negative direction against the polluting
corporations, and the negative attitude will decrease the demand for the
products. Decreased product demand causes a loss of market share. CEPI,
therefore, motivates corporations to comply voluntarily with environmental
standards.

CEPI motivates the public to pressure politicians, government, environmental
NGOs, the community, and the market to act against polluting corporations.
These pressures result in a strengthening of environmental law enforcement,
increasing environmental litigation, stimulating product boycotts, and
strengthening mass media exposures. The threats can constitute potential
costs for the polluting corporations and the potential cost is a reason for
corporations to comply voluntarily with environmental standards.

CEPI influences investors to predict that the stock prices of the polluting

corporations will drop because of the potential effects of negative information

Economists generally use the term “incentive” rather than motivation or pressure.

65

h Nani... r...—

D25;

““4" ‘3. MIL

" ‘3“..‘1' ,3
Or “1.“.th

 

such as the potential loss of market share and the increased costs of corrective
action by the polluting corporations. This results in the reduction of stock

value and motivates polluting corporations to comply voluntarily with

environmental standards.

Disclosure of CEPI for non-polluting corporations creates the opposite effect.
That is, the potential increase of market shares and stock values produces the motivation
or incentive for non-polluting corporations to over-comply with the environmental

standards.

The idea of PID is visually summarized in the following Figure 3.2: A Causation-

Focused Model of Public Information Disclosure.

66

Figure 3.2. A Causation-Focused Model of Public Information Disclosure

 

[Corporate Environmental Information disclosure by Governmentj

J

(Public)

 

 

 
 
     
   
 

  

GOVERNMENT

 

 
  

  
       
   
     

COMMUNITY
Civil Complaints

 

 

 

Consumer Attitude
and Purchase
Behavioral Change

 

 

 

 

. ./

Tighter Law Law Suits
Enforcement: ‘—_ Boycott by NGOs
F mes/Regulations Political Pressure
Direct Negotiation
Mass Media Exposure
MARKET

     
   
   

  
  

 

INDUSTRY
Effect on Stock Value
Effect on Market

Share

 

 

 

 

 

Pollutin Cor orations
Reputation Loss

Profit Loss

Financial Loss

 

 

 

 

\

l

Non-polluting
corporations
Relative Benefit

./

 

 

 

 

 

(IND USTR Y)

 

Voluntary Compliance / Over-Compliance for Survival
in the New Competitive Business Environment

 

 

67

 

Tum.
... 5k

.
*H- ,ar'

50:19.. t...~...\

. i .. l
scram. C

3:45.) i‘H" ‘
La :5 D
I

M Hi 11\

It; q 'h’j - to.

”>U»ll\l

pilot: for p

is its first 5:.-
Em;

“‘h L
“fists the

10“,}.de poll,

it aoitude .

berattoml L

ahalorient, '

1011mm Ct

_r\)
A

“7930
(”551mm anii ‘

Three actors are at work in the framework of PID: government, public or diverse
social drivers, and corporations. Government should provide highly credible and easily
understood CEPI to public. CEPI disclosures may create negative attitude and behavior
of the public toward polluting corporations and create positive attitude and behavior of
the public toward non-polluting corporations. Then, negative attitude and behavior of
public for polluting corporations or positive attitude and behavior of public to non-
polluting corporations force corporations to reduce the total quantity of pollution
emissions. Thus, the government’s CEPI disclosure is the initial stage, the public’s
attitude and behavior change is the intermediate stage, and corporate pollution reduction
is the last stage in the fi'amework of PID.

Environmental information about corporate pollution disclosed to the public
Changes the public’s behaviors (i.e., consumer purchase behavior or investors’ behavior)
toward polluting or non-polluting corporations and the behavior changes are preceded by
the attitude changes.6 The resulting is expressed in altered purchase practices, and this
behavioral change creates incentives (i.e., pressure or motivation) for corporate pollution
abatement. Therefore, the conceptual framework of PID can be examined in terms of the

following components (Refer to Figure 3.3):

1. Development and release of CEPI to the public
2. CEPI influence on public’s negative (or positive) attitude and behavior toward

polluting corporations (or toward non-polluting corporations).

\

6
Assumption about the relationship between behavior and attitude is discussed in the Chapter IV Research
estion and Hypothesis.

68

3. Negative attitude and behavior toward polluting corporations causes loss of
market share of polluting corporations.
4. Loss of market share or the fear of lost market share leads to corporate

pollution reduction.

Figure 3.3

Conceptual Framework of Public Information Disclosure

 

 

 

 

GOVERNMENT PUBLIC MARKET CORPORATION
CEPI Attitude & L f Pollution
Disclosure Behavior 055 0 Reduction

’ Change . Market ’
Share

 

 

 

 

 

 

 

 

 

 

 

 

69

Research

TI.

0. _
1‘“wa ‘
“l‘\l\cl“\l\

no,

‘I
L A

5. than”: I
U

"“‘-.\::;

$ 0
ifi‘w in
i" \u “1‘

CHAPTER IV

RESEARCH QUESTIONS AND HYPOTHESES

Research Questions
The primary research question is: “Does corporate environmental performance
information (CEPI), when disclosed to the public, influence the product market by
creating incentives (i.e., pressures or motivations) for corporations to reduce their
pollution?”
To answer this primary question, the following three subsidiary questions have
been developed:
1. Do negative CEPI disclosures generate negative consumer attitudes and
behavior toward polluting corporations and, conversely, do positive CEPI
disclosures cause positive consumer attitude and behavior toward non-

polluting corporations?
2. Do CEPI disclosures shrink the product market of polluting corporations and,
conversely, do the CEPI disclosures expand the product market of non-

polluting corporations?

3. Could public information disclosure (PID) as a pollution control tool be

effective in Korea?

70

Research Foctt
T3315 re
..es 171 to:
'c m er
The c

fit‘fig- ..
hue)“;

"J

835
conceptual
”mama:
Meow
mflui‘nce tl

“Otpottum

Research H

%

Research Focus

This research focuses on the causal relationship between CEPI disclosure and
changes in consumer attitude and purchase behavior. An empirical approach is employed
to answer the research questions.

The conceptual framework of PID related to product markets is defined as

follows:

1. CEPI leads consumers to develop negative attitudes and behavior toward
polluting corporations and positive attitudes and behavior toward non-
polluting corporations.

2. The negative or positive attitude and behavior toward polluting corporations

or non-polluting corporations cause corporate pollution reduction.

Based on the findings discussed in the Chapter H, the second stage of the
conceptual framework has been demonstrated to be true. Now, the first stage of the
conceptual framework must be examined in order to determine the truth of the conceptual
framework of PID. Hence, the research focus is on the question “Does CEPI disclosure
influence the public’s attitudes and purchase behavior toward polluting corporations and

non-polluting corporations?”

Research Hypothesis

CEPi_md Attitude Change

71

“S
{I
..‘I

about the pol
taxation. C
rte-rot es the
masks it.
the} also b0}
'30166 8; A:
elect or. cor
{hugged by (
Pill, CEPI c3

Attitt
iFlSilbem S;
that is “pre
(£3318; Ci
toward a SPt
lime ShOUld
coneSPOI‘ids

[Ward bu“.

Ex"mood

It is hypothesized that CEPI disclosures change consumers’ attitude toward
polluting and non-polluting corporations. Most PID advocates assert that disclosing CEPI
to the public creates reputational incentives for pollution reduction because information
about the polluting actions and practices of corporations can damage corporate image and
reputation. Conversely, environmental information of non-polluting corporations
improves their image and reputation. When the Exxon Valdez oil spill disaster occurred
in Alaska in 1989, consumers not only developed unfavorable attitudes toward Exxon,
they also boycotted Exxon products and returned thousands of the company’s credit cards
(Bovee & Arens, 1992). Assuming that corporate image or reputation can have a direct
effect on consumer attitude toward corporations, it is inferred that an image or reputation
changed by CEPI disclosure can change consumer attitude toward corporations. Simply
put, CEPI can change consumer attitude toward corporations.

Attitudes directly influence behavior or influence behavior through intentions
(F ishbein & Ajzen, 1975).l “Attitude is conceptually defined as a psychological tendency
that is expressed by evaluating a particular entity with some degree of favor or disfavor”
(Eagly & Chaiken, 1993, p. 1). According to Eagly and Chaiken (1993), “An attitude
toward a specific behavior directed toward a given target in a given context at a given
time should predict the specific behavior quite well because this attitude exactly
corresponds to the specific behavior” (p. 167). For example, the changes of attitude
toward buying Exxon products strongly influences the behavioral changes of purchasing

Exxon products because Exxon event released 11 million gallons of crude oil that

 

' The theory of reasoned action asserts that intentions have a direct effect on behavior and attitudes have a
direct effect only on behavioral intentions, not on behavior.

72

\u—‘q

 

”.37.th '53
m, '

E: 2‘
(it. com
compost
racism: 5
enconsunt
cope-razor
lessen at
I.“ :consu:
his study. (
trimming
adienisem
fal‘Orable.
Were away,
Th:

dlSClosum

fir“ hlpoi

polluted the Alaskan coastline in 1989. Thus, attitude toward purchase behavior strongly
influences the purchase behavior (or purchase intention).

Even though it is not attitudes toward a specific behaviorz, attitude toward targets
(i.e., corporations) also could strongly influence the behavior of purchasing products of
corresponding corporations. So far, environmental studies have not provided any
evidence supporting the assertion that attitude toward polluting corporations influences
on consumer purchase behaviors (or purchase intention) against products of polluting
corporation. However, some studies of advertising have identified the relationship
between attitude toward corporations and purchase behavior. Winters (1989) suggests
that consumers' attitude toward a corporation has a direct effect on brand purchasing. In
his study, Chevron Corporation ran corporate commercials with the objective of
improving consumers’ unfavorable attitudes toward the company. Through these
advertisements, corporate credibility improved and attitude toward Chevron became more
favorable. The company experienced increased brand sales among those consumers who
were aware of the campaign.

Therefore, based on the inference that a positive relationship exists between
attitude toward corporation and purchase behavior (or intention), it is inferred that CEPI
disclosure has a direct effect on the consumers’ attitude3 toward corporations. Thus, the

first hypothesis is as follows:

 

2 Attitude toward behavior (e.g., buying Samsung Cellular phone) is a distinctly different class of attitude
than the attitude toward targets (e.g., Samsung Co.).

3 Like hypothetical constructs, attitudes are not directly observable but can be inferred from observable
responses. Social sciences often have assumed that people’s attitudes can be or should be divided into three
classes: cognition, affect, and behavior. Thus, observable responses for attitude inference are cognitive
responses, affective responses, and behavior responses (Eagly & Chaiken, 1993).

73

H1.“

CE". 351 CW

ll 15.:

axiom-:1:
and cm iron
mutation.

The
credibility.
credibility.
Someone 01
calm-alien
WCClled e
Hence. the
Olatademic
lNEWCll. 19
“Opt. and c
“fined that

“Mavens.

images.

H 1: CEPI disclosures change consumer attitude toward corporations positively

for non-polluting corporations and negatively for polluting corporations.

CEPLand Corporate Credibility

It is assumed that CEPI disclosures change consumers’ perception of credibility
about polluting and non-polluting corporations. Previously mentioned, PID advocates
assert that CEPI creates reputational incentives for pollution reduction because
environmental information of polluting corporations damages their image and reputation
and environmental information of non-polluting corporations improves their image and
reputation.

The change of corporate images or reputation means the change of corporate
credibility. The terms “reputation” or “image” are conceptually very similar to the term
credibility. According to the definitions from Newell (1993), image is the concept of
someone or something that is held or projected by the public and reputation is the general
estimation by which one is held by the public. “Corporate credibility” is defined as the
perceived expertise, truthfulness, and/or honesty of the firm (Mackenzie & Lutz, 1989).
Hence, the term "credibility" seems to be definitionally more specific and for the purpose
of academic research, more appropriate, than either of the terms "image" or "reputation"
(Newell, 1993). Image and reputation, on the other hand, seem to be much broader in
scope, and encompass many other dimensions including credibility. Therefore, it is
asserted that CEPI disclosures influence consumers’ perception of credibility toward
corporations, instead of asserting that CEPI disclosures influence corporate reputations or

images.

74

H 0

. '. .
l 4
,. ihvlsul

sat-“:7”
a ..

I
EL.» 5..

"4“".7‘.’
5M. . . u.
.

‘7‘”,‘Jfl‘c 3.
hiub.d..‘v..

. . .
”’1‘ k,
“italic ..

""th -.‘
“din-Ml.

it 'm

Hui.

L;

:71

Ti

‘.‘c

\

“Ftp:

9‘ ‘
iii ton

‘ 1
‘1‘ w.
..‘potlfi

More recently, Newell (1993) found that corporate credibility had a positive effect
on attitude toward brands (i.e., corporations). Fombrun (1996) asserted that high
corporate credibility is important in producing positive attitudinal changes toward a
corporate advertisement and toward the brand as well as in influencing purchase
intentions. Lafferty and Goldsmith’s (1999) experiment also proved that corporate
credibility or reputation influences consumers’ attitudes toward brand and purchase
intentions.

Based on these findings and the identification of the conceptual relation among

’9 6‘

the terms “image, reputation” and “credibility,” it is asserted that CEPI disclosure
influences the corporate images or reputations, and consequently changes consumer
perception of credibility toward corporations. Thus, the relationship between CEPI and

the consumers’ perception of corporate credibility is reflected in the following

hypothesis:

H 2: CEPI disclosures change the consumers’ perception of corporate credibility
positively for non-polluting corporations and negatively for polluting

corporations.

CEPI and Purchase Intention
PID advocates assert that CEPI changes consumer purchase behavior negatively
for polluting corporations and positively for non-polluting corporations that in turn create

incentives for corporations to reduce pollution. A number of the cases of consumer

75

u
IVA-a- .<
ll
i..~s.-c I.
v
Q‘r-Op an
...‘n J»

V
R“ ”r ‘ l '\
v§‘51.k‘lk ‘
l u .
‘.‘w 9
121303.310

' «
Hi ‘65:

Rffa

s ”10'

tier {0
19h; '

 

boycott or green consumerism support this assertion.4 Thus, the research should observe
purchase behavioral change to prove the causal relationship between CEPI and behavior
change. However, a number of variables influence behavior change. Therefore,
constructing an experiment in which the subject shows behavioral changes caused only
by the CEPI is extremely costly.

The purchase behavior can be well predicted from purchase intention that is a
psychological construct distinct from attitude. Intention represents the person’s
motivation or conscious plan to exert effort to carry out. Purchase intention, therefore,
has an effect on the purchase behavior (Eagly & Chaiken, 1993).5 Thus, it can be
assumed that purchase behavior and purchase intention are comparable for the purpose of

this research. Thus, the third hypothesis is developed as follows:

H 3: CEPI disclosures increase consumer purchase intention for the products of

non-polluting corporations and decrease consumer purchase intention for the

products of polluting corporations.

Hypotheses 1, 2, and 3 are illustrated in Figure 4.1.

\

4

5 Refer to “Market Drivers” in the Chapter II, p. 27.

19 efer to the Fishbein and Ajzen’s theory of reasoned action or planned behavior (Ajzen, 1985; Ajzen,

F‘ 88; Ajzen, 1991; Ajzen & Fishbein, 1980; Fishbein, 1967; Fishbein, 1980; Fishbein & Ajzen, 1975;
lShbein & Ajzen, 1977).

76

:1 fir?)

53;. -,
'4...

. x

Figure 4.1. Hypothetical Diagram of the Effectiveness of CEPI on Consumer Purchase

 

 

 

 

 

 

 

 

Behavior".
Environmental Information
V
Consumer Corporate Consumer
Attitude Credibility Purchase
toward . to Intention
Corporations Consumers

 

 

 

 

 

 

 

 

 

Purchase Behavior

 

 
  
    
   
   

 
  
   
     
 

     

Decreasing
Demand for
Product of
Polluting
Corporations

Increasing
Demand for
Product of Non-
polluting

Corporations

 
 

\

6
This research does not focus on the causal relationship among the three variables but focuses on the causal
relatronship between CEPI and the three variables: attitude, credibility, and purchase intention.

77

 

 

['1'- "
L—L‘ ”1

mm '3
cm i:
i”;.:

“my;

cons.

Environmental Attitude as a Moderator7 Variable

The public’s environmental attitude toward pollution is assumed to influence the
effectiveness of the public information disclosure as a pollution control tool because it is
reasoned that the extent of changes in consumers’ purchase behavior by CEPI disclosure
varies depending on the intensity of consumers’ environmental attitude toward pollution.
Chan (1996) found that respondents who were more concerned about environmental
issues tended to purchase more environmentally friendly products. As is well known, the
movements for green consumerism or boycotts have been initiated mostly by
environmentalists who apparently are high on the environmental attitude scale. Thus, it is
inferred that there is a relationship between consumers’ environmental attitude and

consumer purchase behavior, related to CEPI disclosure so that the fourth hypotheses is:

H 4: Consumers who have strong environmental attitudes about pollution (EAP)
will change the three dependent variables more than consumers who have
weak EAP8. In other words, a positive relationship exists between EAP and
the three dependent variables: attitude toward corporation (AC), corporate

credibility (CC), and purchase intention (PI).

 

7 Moderator is defined as a variable that intervenes between independent and dependent variable and
influences the intensity of effect in the dependent variable. However, it is neither an independent variable
nor a mediator variable. Moderator variable is a special case of the masking variable. Masking variables are
unknown variables that affect the defendant variable. Meanwhile, Mediator variable is defined as a middle
variable located between the first and the final variable in the causal diagram so that mediator is both
independent and dependent variable. For more details, see chapter VII, p. 221.

8 In this research, environmental attitude specifies the environmental attitude toward pollution (EAP)
because the research is related to industry pollution. Thus, high EAP is defined as the status of a strong
concern, awareness, and belief about pollution. Low EAP is defined as the status of a weak concern,
awareness and belief about pollution.

78

cow

It»!

Or brzx

Corporate F ainiliaritxfl Moderator Variabl_e

There may also be a relationship between consumers’ familiarity with
corporations and consumers’ purchase attitude and/or behavior changes. According to
Eagle and Chaiken (1993), the amount of stored information or knowledge that is
available and accessible to people (i.e., familiarityg) moderates attitude-behavior
correspondence. The less information possessed by an individual, the greater the change
induced by any new piece of information. Then, Eagle and Chaiken’s assertion is
consistent with the finding of Lavine, Huff, Wagner and Sweeney (1998) that one with
more information has a stronger and more stable attitude than one with less information.
Many consumers do not change their purchase behavior even though they are exposed to
CEPI because they already have strong familiarity (or preferences) for specific products

or brands. Thus, the fiflh hypothesis is developed:

H 5: Consumers that are highly familiar with a corporation (CFC) change
consumer attitude, corporate credibility, and purchase intention less than one
with low familiarity. In other words, a negative relationship exists between
CFC and the three dependent variables: attitude toward corporation (AC),

corporate credibility (CC), and purchase intention (PI).

CEPI Credibility as a Moderator Variable
The credibility of CEPI in the mind of the consumer moderates the change of
consumer attitude, corporate credibility, and purchase intention. The credibility of

information is a critical element for the success of the P11) program aimed to reduce

 

9 Additionally, for a conceptual understanding about “familiarity”, refer to Brucks (1985).

79

o“ “
LoLU .A

YH‘V’j

...x.‘ 5

m--'

”nigg

'fi “v.3
um. ,,

d,‘re

d

industry pollution. BAPEDAL (Indonesia’s National Pollution Control Agency) and
NIPR (the New Ideas in Pollution Regulation: a group of the World Bank) initiated an
Indonesian PID program “PROPER.” In 1995, PROPER made considerable effort to
increase the credibility of information which they disclosed to the public because
mistakes made in public could destroy the credibility of the grading system of CEPI. To
increase the credibility of CEPI, they based the grading system upon multiple sources of
data, conducted independent inspections, developed a user friendly computer program for
analyzing the data, and designed a multiple step process for reviewing proposed grades
before disclosing CEPI to the public. Aronson, Turner and Carlsmith (1963) found that
for messages from a high credible source, there was a substantial increase in opinion
change but when the source had only a moderate level of credibility, opinion change

decreased. Thus, the sixth hypothesis is introduced:
H 6: The greater the credibility of CEPI (CCI), the greater the effect of the CEPI

rating on the dependent variables. That is, there is a positive relationship

between CCI and the three dependent variables: AC, CC, and PI.

80

 

lntr

Dar
5.. ‘

Rein
LY I‘

56?.

CHAPTER V

RESEARCH METHOD'

Introduction

The primary purpose of this experiment is to test the hypothesis that corporate
environmental performance information (CEPI) disclosures change consumers’ attitude
and purchase behavior negatively against polluting corporations or positively for non-
polluting corporations.

The experiment is designed to observe causal effects of CEPI on purchase
behavior changes. Hence, CEPI is an independent variable and the dependent variables
serving to predict purchase behavior are as follows:

1. Attitude toward corporation (AC)
2. Consumer corporate credibility (CC)

3. Consumer purchase intention (PI)

The following three variables are moderator2 variables that are used to test
hypothesis H 4, H 5, and H 6:

4. Environmental attitude toward pollution (EAP)3

5. Consumer familiarity with a corporation (CFC)

6. Credibility of Corporate Environmental Performance Information (CCI)

 

' Images in this dissertation are presented in color.

2 Moderator is defined as a variable that intervenes between independent and dependent variable and
influences the degree of dependent variable. However, it is neither an independent variable nor a mediator
variable. Masking variables are unknown variables that affect the defendant variable. Meanwhile, Mediator
variable is a middle variable located between the first and the final variable in the causal diagram so that
mediator is both independent and dependent variable. For more details, see chapter VII, p. 221.

3 High EAP is defined as the status of a strong concern, awareness, and belief about pollution. Low EAP is
defined as the status of a weak concern, awareness and belief about pollution.

81

 

 

M,

P...

am

011

Experimental Design and Participants

An experiment was conducted to collect data identifying the causal relationship
between CEPI and consumers’ purchase behavior change.

The over-arching the research design is the two-group random assignment method
combined with pre-post tests in the laboratory environment (i.e., university classroom). 4
The post-test was performed approximately one week after the initial experiment (i.e.,
pre-test) to track individual attitude and behavior change caused by a stimulus, which was
CEPI. This test was used to observe differences of individual attitude or behavior before
and afler exposure to the stimuli information.

Both of the two groups in the post-test were defined as treatment groups because
they were provided with stimuli, but subjects in the pre-test were not given any stimuli so
that subjects in the pre-test were a control group. In the post-test, each of the two groups
were treated with inverse information (e. g., positive versus negative CEPI) in order to
comparatively observe how the subjects responded to inverse information about the same
corporations concerning corporate pollution emissions. For example, Group A in the
post-test was given negative information and Group B in the post-test was given positive
information about pollution emissions of Samsung Electronics Co.

Three hundred six Korean undergraduate students participated in this experiment

on a voluntary basis.

 

4 All data were collected during regular class session. A laboratory setting for the experiment was selected
rather than field experiment because it is desirable to control the experimental situation and to manipulate
experimental stimuli more precisely for the causation study (Wirnmer & Dominick, 1994). Therefore,
subjects were invited into the laboratory (i.e., classroom) rather than having experimenter go to the subjects
in a real world setting.

82

Cort

,,
‘=n-n
EL. L
V

1“Ir
it)“.

Riny

lllSial

Mtlf

coax

Prod:
PTOdL

exper

Corporations and Products Selected

The four real5 Korean corporations and their new unnamed products presented to
subjects were 1) Samsung Electronics Cellular Phone Co. and its new cellular phone, 2)
Hyundai Electronics Cellular Phone Co. and its new cellular phone, 3) Nongsim
Ramyeun Co. and its new instant noodle, and 4) Binggrae Ramyeun Co. and its new
instant noodle.

Four corporations, two of which are cellular phone companies and two of which
are instant noodle companies, were selected to measure attitude and credibility toward
corporations. Their four new unnamed products were presented to subjects to measure
purchase intention.

To eliminate the influence of established brands, this experiment employed new
unnamed products that were fictitious. Consumers’ purchase intention for particular
products is influenced by previous knowledge, experiences, or preferences for those
products. By using a new unnamed product, the influence of previous knowledge,
experiences, and/ or preferences on purchase intention can be reduce.

The market share of Samsung and Hyundai for cellular phones and Nongsim and
Binggrae for instant noodles were almost equal at this time. Dominant market share
means a pre-existing strong preference by consumers. The strong pre-exiting preference
of subjects hinders constructing a strong experiment because the strong preference
hinders changes of consumer purchase behavior, even though external stimuli are strong.

Cellular phones and instant noodles are considerably salient for Korean university

 

5 Using real corporation names is expected to diminish issues resulting from the artificial nature of the
experimental environment.

83

Ct

students.6 Student subjects know enough about cellular phones and instant noodles to
formulate meaningful estimates of brand7 attitude and purchase intention. Korean

university students also are very familiar with the above four corporate names.

Stimuli

One of two colors ( green and red)8 was assigned to a corporation to show the
status of corporate compliance with Korean national pollution standards. “Green” was
assigned to Samsung and Nongsim, and “Red” was assigned to Hyundai and Binggrae for
Group A in the post-test. For Group B in the post-test, “Green” was assigned to Hyundai
and Binggrae, and “Red” was assigned to Samsung and Nongsim. That is, two booklets
of stimuli distributed to subjects contained inverse CEPI.9

The graded CEPI was disclosed under the name of the Bureau of Environmental
Policy in the Korean Ministry of Environment (KME) to increase the credibility of
information provided to subjects.10 Subjects were informed that the Bureau of
Environmental Policy in the KME gathered data, and rated the status of corporate
compliance with the pollution standards.

The meaning of the color grades of four corporations were repeatedly presented to

subjects in the format of simple and colorful written sentences, a table, and a graph

 

6 For example, a golf club is not a salient item for Korean university students because golf is a very
luxurious sport in Korea. However, according to an informal survey, most (or al most all) university
students have cellular phones and almost all students frequently enjoy instant noodles.
7 “Brand name” is synonymous with “corporate name” in this study. When the brand name is more familiar
to consumers than the corporate name, brand name instead of corporate name can be presented to
experimental subjects.
8 A five-color grade system is used for PROPER in Indonesia to rate the extent of corporate compliance to
national pollution regulation standards. This system already effectively worked without any problems in
Indonesia. Therefore, the color-coded grades as an evaluation system for corporate environmental
rformance is expected to work effectively in Korea.

Assigned color grade is arbitrarily produced for the experiment. That is, the grades are not true.

'0 The credibility of KME was known to be high for the Korean public.

84

 

Ca.
(V.

E:

lrr‘
la3‘

.‘le

Tale

adje

depicting the corporate environmental index” to help subjects easily remember the
information provided in the booklets. The corporate environmental index was provided in
the name of the Department of Environmental Assessment in the Bureau of
Environmental Policy in the Korean Ministry of Environment (KME) in order to give
subjects the understanding that the information provided to subjects was produced by
scientific professionals. Two photos of industrial pollution were placed on the cover page
of the booklets to attract subjects’ attention. All colors in the booklets for stimuli were
color-printed.

Except environmental performance information of four corporations, no other
information relating to the four corporations was provided to the subjects because
information not related to the environmental performance of the four corporations could
be confounding variables.

Except two types of CEPI (i.e., each the inverse information), all contents in
stimuli material (e. g., cover photos in the front pages of two stimuli printing materials)
were exactly the same. This helped remove influences from unexpected factors on

dependent variables.

Measures of Dependent and Moderator Variables
The subjects were asked to respond to several questions. First, they were asked to
rate their overall impression of the four corporations on a four- tem and 7-point, bipolar

adjective scale (i.e., Semantic Differential type scale) to measure consumer attitudes

 

” The index was also artificial.

85

 

 

if]

:4

C01
Sul

lUlt‘

toward each corporation. The scales are anchored with “unreputable / reputable,”
“untrustworthy / trustworthy,” “negative / positive,” and “dislike / like.”12

Second, the subjects were asked to rate their overall evaluation about corporate
credibility on a four-item and 7-point bipolar adjective scale. Two of the four items
anchored with “overall low quality products / overall high quality products” and with
“not at all good at manufacturing / very good manufacturing” measured corporate
expertise. The other two items anchored with “not at all dependable / very dependable”
and with “not at all concerned about customers / very concerned about customers,”
measured trustworthiness.”

Third, subjects were asked how likely they would be to consider buying that
product and that they would purchase it. A three-item and 7-point Semantic Differential
type scale was employed. The three items were anchored by “very likely / very unlikely,”
“probable / improbable,” and “possible / impossible” (Yi, 1990).

The scales used for all of the three dependent variables were the same so that each
variable would contribute equally to the distance measure. It was, therefore, not
necessary to standardize the data.

A moderate variable of familiarity that was subjective knowledge about
corporations used in the stimuli was measured in the pre-test. One of two measures for
subjective knowledge asked subjects to respond to the following statement: “Rate Your

knowledge of this item, as compared to the average undergraduate students.” This

measure was anchored by “one of the most knowledgeable / one of the least

 

'2 Measures for attitude toward corporations was developed based on the measure used by Bruner & Hensel
eds., 1992, # 29 (Boulding and Kirmani) .

‘3 Bruner & Hensel eds., 1992, # 72 (Keller & Asker). Four of six items used by Keller and Asker were
adopted for the measure of corporate credibility in this experiment.

86

 

I""

 

Si.

7:1

V‘
6..

(1g
ti;

knowledgeable.” The second measure asked subjects to respond to the following
statement: “Circle one of the numbers below to describe your familiarity to this item.”
The anchor for this scale was “Extremely Familiar / Not at all familiar.” For both
measures, 7—point semantic differential scales were used (Brucks, 1985).

Subjects were asked to rate their overall concern, consciousness, and belief
toward Korean pollution on a 5-point Likert type scale. Eight items measuring
environmental attitude consisted of the verbal and behavioral commitment, affect and
cognitive dimension (i.e., knowledge about environment), and government action for
environmental protection. '4

The questions for environmental attitude were not placed before the questions for
three dependent variables in order to prevent subjects from guessing the experimental
hypothesis fiom the measure of environmental attitude.

Questions to check manipulation of CEPI were presented in the post—test because
subjects received stimuli in the post-test. To assess the effectiveness of the experimental
CEPI manipulation, subjects were asked whether the four corporations complied with
national environmental standards “very much” or “not at all,” and whether the four
corporations had severely negative or positive impacts on the environment and the health
of human beings. The rating scale was a 7-point Semantic Differential type scale.

Credibility of stimuli information presented by the KME name was measured on a
7-point, bipolar adjective scale, anchored by “not credible / credible” and “scientific /

unscientific.”

 

'4 See Appendix A. Questionnaire for Experimental Tests. Referring to Leeming, Dwyer and Bracken
(1995), Berberoglu and Tosunoglu (1995), the measure for environmental attitude was developed for this
experiment.

87

00nd

Prob

”16:15
llkel)

and ..

All measures for variables mentioned above were sealed with a high number (e. g.,
5 or 7) denoting a positive response (e.g., “like”) and a lower number (e. g., l) denoting a
negative response (e.g., “dislike”).

The final questions in the booklet were constructed for demographic data. They
were placed in the pre-test. Demographic data consisted of “Gender,” “Age,” “Major,”
“Hometown,” and “Average monthly family income.” Demographic data was

additionally collected to obtain basic information about subjects.

Procedures
Translation and Back-Translation Test

A translation and back-translation test for measurement and CEPI stimuli was
conducted before implementing the pilot test to ensure reliable and valid translation from
English to Korean. The original questionnaires and content of stimuli drafted first in
English were translated into Korean, and then back translated into English using external
translators blind to the hypotheses. As a final process, the original English and back-
translated English version of the measurement and stimuli were compared to find
problematic differences between them that could weaken their reliability and validity.

A difference found from the translation and back-translation test was about the
measure for the variable of “purchase intention,” anchored with “very unlikely / very
likely,” “improbable / probable,” and “impossible / possible.” The meanings of “likely”
and “probable” in the original questionnaires were back-translated into “probable” in the

last process because the meanings of both terms are almost the same in Korean language.

88

Thus, “very unlikely / very likely” was translated into “not want to / want to” in the

Korean version.

Pilot Test

The pilot test was conducted during January 2001 to achieve its three purposes: 1)
manipulation check for stimuli, 2) measurement check, and 3) checking Korean wording
in Korean questionnaires.

Twenty five Korean Kyeung-Gi university undergraduate students participated in
the pilot test. They had been at Michigan State University since December 2000 as
visiting students. Their age, major, and sex were diverse.

A manipulation check for stimuli confirmed that the manipulation of CEPI was
successful as expected. Negative information was perceived as negatively and positive
information was perceived as positively in the pilot test.

Except the measure of environmental attitude, all measures were revealed to be
highly reliable and valid. Exploratory factor analysis performed on the eight items for the
measure of environmental attitude postulated a three-factor model so that confirmatory
factor analysis based on both the posited three factors and one factor were conducted on
its eight items to observe the heterogeneity (i.e., parallelism) and homogeneity (i.e.,
internal consistency) among the three factors or within one factor. As a result, a
significant reason for the three-factor model was not found. Afier inspecting the factor
loadings and errors generated from the discrepancy between the obtained and predicted
correlations, it was concluded that it was not necessary to exclude any item of the

environmental attitude measure.

89

Based on comments from participants, one of 8 questions for environmental
attitudes was redeveloped. Except comments about an item of environmental attitude,
there were no comments about the Korean wording from participants so that it was
assumed that the questionnaires have no Korean wording problems in the Korean version

of the questionnaires and the stimuli materials.

Administration

The experiment was implemented from May 17, 2001 to June 12, 2001 in the
classrooms of Seoul City University located in Seoul and in the Kangwon University
located in Chun-Chen15 in Korea. For the pre-test of the experiment, participants first
received an “Informed Consent Form” containing assurance of confidentiality of data and
statement of agreement. Participants also received an instruction asking to read each page
of the booklet carefully without being allowed to turn back and without being allowed to
talk to each other. In addition to the written instruction, administrators gave very detailed
verbal instructions on how to complete the questionnaire. Then, stimuli booklets were
randomly distributed to three hundred six participants and they were asked to write down
their student ID number in the first page of the questionnaire booklets, in order to know
the identification of subjects between pre-post tests. They were asked to keep the booklet
face down, when the questions were finished.

The post-test with the same participants was performed approximately one week

after the pre-test in the same classroom during normal class-time. ‘6

 

'5 The city of Chun-Chen is located around 150 miles west from Seoul.
'6 Time interval between pre-post tests is M (Mean) = 8.7 and SD (Standard Deviation) = 4.0 days (7
days=68. 1% and 8 days=15.5% of cases). All subjects answered to experimental questionnaires just after

90

lead“
333;,

In the post-test, two types of stimuli booklets that contain the manipulated CEPI
were distributed randomly to participants and they were asked to read the booklets very
carefully. Participants were asked to write down their student II) on the first page of the
questionnaire booklets to figure out the participants’ identification to observe the
individual differences of responses to the stimuli. They also were asked to write the type
of stimuli (e.g., “A” or “B”) on the first page of the questionnaire booklets for the post-
test to identify who belonged to which stimuli Group A or Group B.

Other instructions given in the post-test were the same as the ones in the pre-test.

Debriefing

As a final step in the process, when participants finished the questions for the
post-test, they received notes of debriefing saying that all the information about the four
corporations given to the subjects was not true and was arbitrarily created for the
experimental purposes. The debriefing also addressed the true purpose of this experiment

and contained a thank you note for participation.

 

reading a print material for stimuli in the post-test so the time interval between reading the material and
answering questionnaires of post-test is constant.

91

CHAPTER VI

DATA ANALYSIS AND RESULTS

Data and Subject Profile

Three hundred six Korean undergraduate students participated in this experiment
on a voluntary basis; 469 participated in the pre-test and 439 participated in the post-test.
Some students did not write down the stimuli type A or B and student identification
numbers. The cases not indicating the stimuli type and student identification numbers
were discarded. The final cases matched by student identification numbers totaled 306
cases.

Of the total number of participants, 52.6 percent were males and 47.4 percent
were females. Their ages ranged from 18 to 33 years of age (M = 21.5, SD = 2.78)‘. Most
subjects were majoring in the social sciences such as: Politics (49.2 %), Business
Administration (9.3%), Mass Media (8.6 %), Psychology (6.6 %) of the participants, etc.
Of the 306 participants, 57.5 percent were students of the Seoul City University in Seoul,
Korea and 42.5 percent of the participants were students of the Kangwon University in
Choon-Chun, a small city located 150 miles west of Seoul. Three fourths of the subjects
grew up in urban areas (big city = 37.2%, small city = 39.1% and rural area = 22.7%).
The Family Monthly Average Income of subjects was 2,160,000 Won or approximately

1,674 US dollars (SD = 1,260,000 Won / approximately 976 US. $)’.

 

' M = Mean, so = Standard Deviation

2 “Won” is a unit of Korean money. The exchange rate was US. $ 1 = approximately 1,290 Won in May or
June, 2001 when the experiments were conducted. Thus, most subjects would belong to a middle class in
terms of Korean social economics.

92

Che

bep
5110

the:

Stil

mtg

C01]

3511

Checking Unexpected Intervention between Pre and Post-tests

All subjects completed experimental questionnaires immediately after reading
printed materials for stimuli in the post-test. Thus, the time interval between reading the
material and answering questionnaires in the post-test is constant and very short.
Therefore, the probability of unexpected factors influencing subjects’ responses to the
stimuli would be low.

However, probability that unexpected factors intervened between the pre and
post-test exists because the time interval between pre and post-tests was on average 8.7
days. The tests were implemented from May 17, 2001 to June 12, 2001. In order to check ‘
the intervention of unexpected factors between pre and post-tests, the researcher observed
negative or positive news related to the four corporations Samsung, Hyundai, Nongsim
and Binggrae, which were released in several major newspapers at the national level
between May 17 and June 12, 2001. Negative or positive news that could influence each
subject’s attitude, credibility and purchase intention related to the four corporations and

their products was not found.

Stimuli Manipulation Check

Stimuli (i.e., treatment) were artificially manipulated. Stimuli consisted of
negative and positive information concerning the environmental performances of the four
companies: Samsung & Hyundai Electronics, and Nongsim & Binggrae Ramyeun. The
research is needed to check whether subjects perceived the stimuli in the same manner as
the researcher intended. That is, did the subjects perceive negative information of stimuli

as negative information, and positive information of stimuli as positive information?

93

A variable “stimuli” was measured with two items in a 7-point Semantic
Differential type scale. One item of the measurement of stimuli asked whether the four
corporations complied with national environmental standards and another item asked
whether the four corporations had severely negative or positive impacts on the
environment and the health of human beings3 . Subjects of A and B groups were given
reverse information of each other. That is, if group A was given positive environmental
performance information for Samsung, group B was given negative performance
information for Samsung. When group A was given negative environmental performance
information for Hyundai, group B was given with positive environmental performance

information for Hyundai. Table 6.1 includes basic statistics related to the manipulation

 

 

test.
Table 6.1.Stimuli Type and Mean, SD and SE of Stimuli

Treatment N Mean Std. Std. Error

Type Deviation Mean
SAMSUNG 1 154 5.4318 1.1520 9.283E—02

2 152 2.3224 1.2387 .1005

HYUNDAI 1 154 2.7370 1.2475 .1005
2 152 5.3684 1.1901 9.653E-02
NONGSIM 1 154 5.3669 1.1896 9.586E-02
2 152 2.4145 1.2067 9.788E-02

BINGGRAE 1 154 2.8961 1.3811 .1113
2 152 5.5132 1.1068 8.977E-02

 

- Group A was exposed to Stimuli (i.e., treatment) Type 1
- Group B was exposed to Stimuli Type 2

 

3 The questionnaires for stimuli is as follows:

* Rate the extent to which the following corporations comply to the national pollution standards.
1. Samsung Electronics Cellular Phone Co. has complied with national environmental standards

Not at all 1 2 3 4 5 6 7 Very well

2. The environmental performance of Samsung Electronics Cellular Phone Co. has created on the health
of human being and nature
Very severe l 2 3 4 5 6 7 Very Positive
Adverse effects Effects

94

T-tests suggested that manipulation of the stimuli for corporate environmental

performance information (CEPI) about four corporations was successful. To check the

success of manipulation, independent sample T-tests were conducted to verify statistical

significance for mean differences between Group A and Group B in the variable

“stimuli.” Significant differences were found for the stimuli of Samsung (M = 3.1094) (t

(304): 22.74, p < .001), Hyundai (M = -2.6314) (t (304) = -18. 86, p < .001), Nongsim

(M = 2.9524) (t (304) = 21.55, p < .001), and Binggrae (M = -2.171) (t (292) = -l8.30, p

< .001).4

Measurement Reliability and Validity

Measurements of five variables with twenty-one items were used in the pre-test.

The variables were:

1.

2.

Attitude toward the four corporations (four items)

Credibility of the four corporations (four items)

Purchase Intention toward products of the four corporations (three items)
Familiarity with the four corporations (two items)

Environmental Attitude (eight items)

Measurements for five variables with fifteen items were used in the post-test. The

variables were:

1.

Attitude toward the four corporations (four items)

 

4 T = t-value, p = p-value, figure in a parenthesis (i.e., t (304)) is a degree of freedom.

95

2. Credibility of the four corporations (four items)
3. Purchase Intention toward products of the four corporations (three items)5
4. Stimuli for manipulation check (two items)

5. Information Credibility of Stimuli (two items)

Factors of measurements and number of items (i.e. questions) of each factor are

summarized in Table 6.2.

Table 6.2

Factors and Number of Items (i.e., questions)

 

 

 

 

 

 

Pre-Test
Credibility Attitude Intention Familiarity Environmental Attitude
Samsung 4 4 3 2
Hyundai 4 4 3 2
Nongsim 4 4 3 2
Binggrae 4 4 3 2
All subjects 8
Post-Test
Credibility Attitude Intention Stimuli Information Credibility
Samsung 4 4 4 2
Hyundai 4 4 4 2
Nongsim 4 4 4 2
Binggrae 4 4 4 2
All subjects 2

 

- “4” is the number of item of a factor. For example: “4 ” in the cell of “Samsung” by
“Credibility” in pre-test means that the factor of Credibility of Samsung is constructed by four
items.

5 Questions were the same in measuring three variables “attitude,” “credibility,” and “purchase intention”
in the pre and post-test.

96

r5.
. .
. . ‘

US

Pa

11:

Confirrnatory factor analysis (CF A) 6 was performed in order to assess the uni-
dimensionality of each scale (i.e., internal consistency and discrimination of each
measurement) because each measurement should retain internal consistency (i.e.,
homogeneity) within each factor and should maintain discrimination (i.e., heterogeneity)
among factors (Hunter & Gerbing, 1982). The first-order CF A set each scale toward
different corporations as a factor (e.g., “attitude” to Samsung is a factor and “attitude” to
Hyundai is another factor) so that the first-order CFA included 17 factors with 60 items
of the pre-test and 17 factors with 54 items of the post-test (Refer to table 6.1). The
second-order CFA set each scale of four corporation as a factor (e.g., setting attitude to
all of Samsung, Hyundai, Nongsim and Binggrae as a factor) so that the second-order
CFA included 5 factors with 60 items of pre-test and 5 factors with 54 items of post-test.
The CF A by a unit of corporation was also conducted. For example, CFA for the scales
used for Samsung included 5 factors (Attitude, Credibility, Purchase Intention,
Familiarity, and Environmental Attitude) with 21 items in the pre-test and 5 factors
(Attitude, Credibility, Purchase Intention, Stimuli, and Information Credibility) with 15

items in the post-test.

g

6 A confirmatory factor analysis was conducted by the computer statistics program “CFA” (1992)
developed by Mark A. Hamilton and John E. Hunter, “M-MODEL” (version 1.0, 1988) developed by
Mark A. Hamilton, “PACKET” (version 1.0, 1988) developed by John E. Hunter and “HTZ (Tests of
Homogeneity and Heterogeneity for Confirmatory Factor Analysis)” (version 1.0) developed by Vernon
Miller and Joyce Lanning. These programs are working on MS-DOS. Dr. Mark A. Hamilton is a professor
at University of Connecticut. John E. Hunter is a professor in the Department of Psychology at Michigan
State University. Vernon Miller and Joyce Lanning are professors in the Department of Communication at
Michigan State University.

97

23.—.33-.-.30 973.92. $.12}; .2072: 12.3323. 99...: 0.: a... —.U\...:2.:::.I. «..-—7. <..~. v L.u~u.:ut~m...~: 0.: C.- .nhagnvrwu w:3:=... 0.23..:...~U-~.~C.01L.u~:\

 

Hobo 3682582: .3 338 983328... we 83.35% Ho.“ 388:8 mud—Na «$0
8383 3:835 8m $5 Samoa .3 380.80 meets—obooéza .mé ES «.6 £6 035 2: 5 .33: .825: can HHS—gm 3 pogo—26p 88on
«£0 05 £5 wean—no.8 macaw—obooaafi 05 55 C032 and 880pr biwzm :5 .5333». 808? 2a mmmm .3 pea—:28 macaw—atoning: -

 

 

oo.H vo. mH. mo. HN. Ho. mo. oH. mo. NH. NH. «0. mo. No. HH. mo. NH. oeauHuue HeuaoaaouH>eunNH
eo. oo.H me. mN. Hm. we. mo. mH. «0.- mm. Ho.- NH. mo. Nm. mo. NH. 0H. oeummaHn nuH: uuHueHHaauuuuH
NH. me. oo.H oN. me. No. we. 4H.- NH. mo. mm. eo.- eN. NH. Hm. 4o. NN. .aHnmaoz nuHa NuHueHHaaeuumH
mo. NN. 0N. oo.H Hm. NH. «0. NH. vN. NH. mo. HH. mN. NH. mo. mH. ON. Heeeaum nth NuHueHHHauu-cH
HN. Hm. me. Hm. oo.H HH. NH. 0H.- Ne. NH. oN. «0.- Nm. NH. mH. mo. om. mannaem nuH: NuHueHHHaeuan
Ho. me. No. NH. HH. oo.H NN. NH. NH.- HN. 40. NN. mo. me. 0H. wN. mo. oHeooe uneuueH oeummch uou aoHuaoueH onenouamuNH
mo. mo. me. «O. NH. NN. oo.H oo. mo. mH. em. NH. mH. mN. mm. NH. 0N. oHeoou uneuaaH aHnmcoz uou aoHueouaH onenuunmuHH
0H. mH. 4H.- NH. 0H.- NH. oo. co.H NN. 4N. mo.- me. mo.- NN. mo. me. No. ocean ueHaHHou Heeeaxa u0u eoHueouaH onenousmuoH
mo. eo.- mH. 4N. Ne. NH.- mo. NN. co.H Ho.- mN. «0. me. mo. mN. mH. me. oeona ueHaHHou meanaum uou aoHueouuH onenousmum
NH. mm. mo. NH. MH. HN. 3. en. Hof coin mH. mm. NH. mm. mH. oe. NN. ouummafln ou opauuuudum
MH. 3.. mm. mo. om. vo. em. 8.. mm. mH. oo.H No. we. mm. mm. mo. em. .5393: cu opauauufluh
vo. mH. eo.- HH. eo.- mN. NH. mm. eo. mm. No. oo.H oN. Nm. mH. mm. 0N. Haven»: 0» ovauHuueuo
mo. mo. 4N. mN. Nm. mo. mH. mo.- mm. NH. Na. ON. oo.H mN. Ne. 0N. mm. mauuaum ou oeauHuueum
No. mm. NH. NH. 3. mm. mm. mm. mo. mm. mm. Nm. mm. oo.H mm. Hv. om. oaummnum no .HuHHHnHvouonv
HH. mo. Hm. mo. mH. 0H. mm. mo. mN. NH. mm. mH. Ne. mm. co.H NN. Ne. aHumaoz uo auHHHnHeouu-m
mo. NH. 40. mH. mo. mN. NH. we. mH. ow. mo. mm. 0N. He. NN. oo.H mm. Haveuhu no NHHHHaHeouUIN
NH. 0H. NN. 0N. om. mo. 0N. No.- me. NN. an. 0N. mm. om. Ne. mm. oo.H masaaem uo NuHHHnHeououH
NH eH mH «H NH NH HH oH m m N u m v N N H

 

um0ut0hnm 0st Gm EOHQGW $002503 QCBM—PCOUJSHE U0u00b0mv ..MO 0—969

.mfio 58:00

5:28.80 2322: £023 .833 wEBozom 025 2: E BntmEfiSm 0.8 <m0 8296.5 05 5 E880 wcofim mcocsmhouéfifi

98

.19. 3......— UZ. a... .....L:_.J:..— ...J.J>)~U£ 23.3.2.9FNC. v|.-.t.v:: 33.3.JLLCTV .VNV U~C~NF

 

Eugen confluence-SHE 33030 05 wfiEZc E 33:23 2N We can v.0 .00 633. 05 E macaw—oboe 382.80 805-

.N 28 H H98.“ .3 5 BNEEEHNE 05 mo 858m 05 mo 88 98:3 2: .3 N 88$ 28 H ~3on

 

 

oo.H mo. mo. No. mo. NH. mo. 0H. mo. «o. mo. mo. mo. No. NH. mo. 3. 533396 5333373
mo. 8H 8.- 3. HN.- mm. 8.- om. NH..- NN. 3.- He. emf mm. Ne: He. mm..- oeummeHm no HHHfiHum-nn
mo. no.1 oo.H vwf mm. mm.- Nb. Nvf mm. mm.1 bb. nmf ww. omf bb. vm.- mm. Sumac! no «ngmlnn
bo. vm. vm... oo.H ww.- Nm. bm.t mm. mvf 0b. mm.- mm. Nm.- mm. wm.- mm. mm..- «avg no «ngmldn
mo. Hmi mm. emf 3H 3.. ON. 3.- 3. 3.- «N. 3.- HN. mm: 3.. Ne... 0N. manna-m no HHsaHum-on
NH. mm. mm.- mm. mm.- ago-H .3..- Hw. bm.- mm. mvf mm. mm.- om. mef mm. mm.- OHQOOHH ufiduunfl onugfldn How acaunouaH Gannon—Hula“
mo. 3.- NN. Nmi ON. HY- 3H 3.- 3. 3.- Nm. 3.- oN. 3.- mm. 3.- NN. oHeooa uneunnH Sumac: uou noHueouaH 3335-:
OH. om. mv.- mm. 3.- am. mv.- coin mm: mm. ovf 0b. mm.- mm. mm.- Mb. bm.- OGOAQ HIHHHHHOU «$0; How GOHuflOuGH 0030.35le
mo. NH..- Na. 3..- we. NN.- em. NN.- oo.H NT- Nm. 3..- mN. NN.- we. 3.- NN. 0:93 .3338 manna—um now 33:35 3335-:
eo. Nb. Nm.- 0b. mm..- mm. mv.- mm. Nvf coin mm.- bb. mwf vm. vmf bb. mvf Odummflfin Ou ovfluwuudnmu
mo. mm.- bb. mm.- vb. m¢.- bm. ovf bm. mm.- oc.H Hm.- mm. mm.- mm. mwf mm. adamfloz 0.... o—Xfifl-ufllvu
mo. Hm. Nm.- mm. vm.- mm. wvf 0b. «v.1 bb. Hm.t coin bv.- mb. om.t mm. 3..- .nflvg O» ovfluduudlnu
mo. vm.- mm. mm... Hb. mm.- 0b. mmf mb. mvf mm. b¢.- oo.H Hv.- mm. vvf mm. madam Ou Ovfluauudluu
No. mm. emf mm. mm.- om. mv.- mm. mm.- vm. mm.- mb. va coin mm.- «m. mvf Gage—"m uo hudflwnwuouoldn
MH. Nmf bb. mmf Vb. mvf mm. mm.- mm. emf mm. om.- mm. mm.- cc...” om..- mm. 6.3932 no enuHHHnHvouolou
mo. Hm. vm.- mm. bm.- mw. mv.- Mb. va bb. av... mm. 3.- vm. om.t coin mvf «dug uo %uaddndflouuvlad
H. mm:- me. mm.- 0N. mmf NN. NN.- NN. 3.- mm. 3..- mm. 3.- mm. m3..- ooH manna-Hm no .SHHHnHeduUamH
on mm mm Hm on an an bu on mu w" nu an Ha on a." «H

 

um0uuum0m 05“ GM thuomnm CO0EOD Gowns-€001.52: avg—09:00 ...VC gawk.

99

.vé 2an 9 £8 .38 BE 05 E Vm 8 3 88m $0283: 2: Son-N -

 

 

8. 8.-8.-8.-No.-wo.-8.-mo.-S.-oo. 8.-8. 8.-8.-8. why-NN.- 3333 830531683
8. Z. 8.- 8. 8.- Nm. 8.- E. 8.- 3.. 8.- S. 8.- mm. 8.- 3. 8.- ouummfim n»? 3333573
2. ma. 8.- 3. 8.- 8. NH. mo. mo. 8. NH. 8.- No. 8. ma. 8. 8. 33.82 5!. 3333583
2. 8. 8. S. No.- 2. No. 3. 3. 3. 8. 3. 8. NN. 8. Z. 3. «355m A»? 53:338.:
8. No. 8. mo. 8. 8. 8. No. NN. E. 8. 8. 3. NH. 3. 88. NN. manna—um n3: huauufiaufiunna
NH. ma. mo.- 0H. Ho.- .3. .3. NH. mo.- Nw. No. NN. No.- 3‘. Ho.- bN. oo. Bacon 9533:" ouummaam you noduaouan oudnouflmINH
ON. 2. 8.- NH. 8.- ON. 3. 8. 8.- NN. S. 2. No. 3. S. 8. 8. 0300: panama.“ 38982 uou noflaouan manna-Snug
NH. 8. 8. 3. No. HN. 8.- 8. 8. S. No.- 3. 8.- ON. 8.- N... 8.- 23am 8.333 805$: new aoflnounn 0.3338."
8.- 8. mo. 80. 8. 8. 8. NN. mm. 8.- ma. 8. 8. 8.- 3. 8. «m. 39:— uugfioo 259:3 new 53:35 3333-8
mo. 8. 8. 8.- NH. NN. 8. 8. 8. av. 8. S. 8. mm. 8. 3. 3. 09.533 ou 3333.8
8. 8. mo. 8. 8. 88. NN. 8. NH. 8. on. 8.- 3. 8. mN. 8.- NN. E8982 3 333::
3. mo. 8. S. 88. NN. 8. 8. 00. «N. 8. Nm. mo. om. 8. mm. mo. 8055.. 3 3332-8
8. 8.- S. 8.- ma. 88. NN. 8.- Nv. 8. NM. 8.- 8. 8. Nm. 8. 8. 98956 3 33338
3. «0. 0H. .3.- oa. mm. HH. NH. bo. mm. NH. 5. ca. mm. ma. 0N. ma. ouummaum no huaafinwvounulv
E. No. 2. 8. 8. mo. NN. 8. ON. 8. mm. 8. MN. 8. «n. 8. mN. 5352 no 533305..”
bH. «0. mo. mo. mo. ma. 3. mm. 0H. «N. .3. om. ma. mN. 0H. cm. 3. «avg no huwawflvouonn
2. 8.- NH. 8.- 3: S. «N. 8.- om. mo. 8N. 8.- 3. 8.- 8. No. Nm. 95955 No 3333265
In 8 Na 8 on «N «N 8N 8N 8 «N NN NN NN oN 3 3

 

«mow-“mom was 8m 05 5 £8on 52509 scum—280-885 .m6 2an

100

CO:

 

pk. nu. nnv .\~ -.\

O Q O D Q I

Tak

sec

9: Cu :~ .3 ......\ \
§

. n
vIL

 

h...

Second-order factor correlations (i.e., inter-correlation among factors across

corporations) in the pre and post-test are presented in the following two tables7.

Table 6.68

Second-Order Factor Correlation in the Pre-test

 

 

Factor 501 502 503 504 505
501 100 97 70 42 12
502 97 100 76 45 13
503 70 76 100 38 12
504 42 45 38 100 17
505 12 13 12 17 100

 

- The correlation “97” in the table is actually “.97.”

- Factor 501= Credibility of 4 corporations: —— Samsung, Hyundai, Nongsim and Binggrae
- Factor 502= Attitude of 4 corporations

- Factor 503= Purchase intention of 4 corporations

- Factor 504: Familiarity of 4 corporations

- Factor 505: Environmental Attitude

Table 6.7

Second-Order Factor Correlation in the Post-test

 

 

Factor 501 502 503 506 507
501 100 122 101 54 21
502 122 100 100 58 15
503 101 100 100 65 22
506 54 58 65 100 22
507 21 15 22 22 100

 

 

7 Unfortunately, second-order correlation between factors in the pre-test and factors in the post-test is not

presented because the CFA program developed by Hamilton and Hunter (1992) is incapable of handling a
large size of correlation matrix with 114 items of 34factors, which are 17 factors with 60 items in the pre-
test plus 17 factors with 54 items in the post-test.

8 Correlations in Table 6.6 and 6.7 are generated by program CFA.

101

l“

«U

 

' ¥

- The correlation “58” in the table refers to “.58.”

- Factor 501= Credibility of 4 corporations: Samsung, Hyundai, Nongsim and Binggrae
- Factor 502= Attitude of 4 corporations

- Factor 503= Purchase intention of 4 corporations

- Factor 506= Stimuli of 4 corporations

- Factor 507= Information Credibility

Discrimination among factors was checked based on the error matrix of factor
loadings generated by a statistical computer programs HT 2 (Tests of Homogeneity and
Heterogeneity for Confirrnatory Factor Analysis). At present time HT 2 is not capable of
handling a size of correlation matrix with 60 items in the pre-test or/ and 54 items in the
post-test, thus, HT 2 was performed by a unit of corporations in the pre and post-test
separately. The following error tables of factor of loading by a unit of corporations show
error rate that is the discrepancy between the obtained and predicted correlations (i.e.,

observed factor loading — expected factor loading).

102

\. ~ .‘ ‘1. l’ .3 9!; ... Air» -Vd

A
u

.
§

u «
Il‘

A‘
u.

.1-
us

IN:
as

.‘.w
Q;

on.»
a c

N!
~ A

«.9
wk

Ga
an

Al...
‘1‘

Q

‘1‘

.\

1 . \u‘s

{v

 

Table 6.8. Error Rate of Heterogeneity (Samsung in the Pre-test)

 

 

1 _ _ - _ - - _ _ - _ .. .. _ - - - -
2 - _ - .. _ .. _ - - - .. _ .. - - .. _
3 _ _ _ _ _ _ .. - _ _ _ - - - .. _ _
4 _ - - _ _ - - - - - - .. _ _ - _ ..
5 1o 3 5 2 - — — - - - - - - - - _ -
6 1 7 13 12 - — — - - - - - - - _ - -
7 8 12 2 10 - - — - - - - - - - - _ _
8 1o 9 4 9 - - - - - - - - - - - - _
9 15 11 16 12 17 16 17 25 - - - - — - - - -
1o 15 15 2 2 13 10 12 5 - - - - - - - - -
11 15 16 2 4 13 1o 12 4 - - - - - — - - -
12 1 1 3 2 4 2 7 o o 2 3 - - - — - -
13 8 o 9 o 9 1 3 3 1o 6 2 - — - - - _
14 4 7 5 2 9 1 11 13 7 2 2 12 6 - - - -
15 1 o 5 3 6 1 3 4 7 1 o 9 7 - - - -
16 3 9 5 2 o 1 3 7 10 o 2 1 2 - - - -
17 7 1o 0 3 10 6 1 3 o 5 11 5 4 — — - -
18 3 2 2 9 8 5 12 7 4 3 6 4 6 - - - -
19 o 3 8 3 1 1 4 1 1 7 7 2 2 - - - -
20 6 8 3 8 5 6 1 o 1 5 1 2 3 - - - -
21 2 5 1o 15 3 6 1o 5 1 2 2 1 o - - - -

 

 

 

coqmunwaI-I
I
l
I
I

\O
I
I
I
I

10 - - - -
11 - - - -
12 - - - -
13 - - - -
14 - - — -
15 - - - -
16 - - - -
17 - - - -
18 - - - -
19 - - - -
20 - - - -
21 - - - -

 

- Error rate “10” in the table refers to “.10.” This rule applies to error rate of all error rate tables.
- 1, 2, 3 and 4: Items of Credibility of Samsung

- 5, 6, 7 and 8: Items of Attitude toward Samsung

- 9, 10, 11 and 12: Items of Purchase Intention toward Samsung Cellular Phone

- 13 and 14: Items of Familiarity with Samsung

- 15, 16, 17, 18, 19, 20 and 21: Environmental Attitude

103

Table 6.9. Error Rate of Heterogeneity (Hyundai in the Pre-test)

 

 

...:

\IQONONCDIhlenwO‘xIWF-‘U'lnb

mummwaI-I
I
I
I
I
I
I
I

HI I
I
I
I
I
I
I
I
I

I—II
(DxlI-I'WLAJUINU'IHHNWH

NHHHHI—‘l—‘I—‘HI—‘HW

oxomqmmprI-Io
HH H
I—‘H

wqumwmwOI-JU'INI—I

w
H

Hl—l
NHI—‘WNHQNI—‘NOJI—‘ONWOOD

#HHomHHNNmbwwaQr-‘m
wmqwqwzqwoqmmmomww

owxowI-IHwMI-Iwqu
quI—Imhmousoxopoo
OWQIhI-‘wI-‘QNHW
\Imooubmmooqow
H
\ll-‘U'IU'INNHUJHH
H
o
HNI—‘I—‘LDNU'ICA
l
I
I

,_.
1..
H
HH

H
H

 

 

(DQONU'IIbbJNH
I
I
I
I

\O
I
I
I
I

10 - - - -
11 - - - -
12 - - - —
13 - — - -
14 - - - -
15 - - - -
16 - - ~ -
17 - - - -
18 - — - —
19 - - - -
20 - - - -
21 - - - -

 

- 1, 2, 3 and 4: Items of Credibility of Hyundai

- 5, 6, 7 and 8: Items of Attitude toward Hyundai

- 9, 10, 11 and 12: Items of Purchase Intention toward Hyundai Cellular Phone
- l3 and 14: Items of Familiarity with Hyundai

- 15, 16, 17, 18, 19, 20 and 21: Environmental Attitude

104

Table 6.10. Error Rate of Heterogeneity (Nongsim in the Pre-test)

 

 

(DQCDUIDUJNH

NNHHHHI—‘HHHI—‘Hm
I—onoooqmmthI—Io

MHHWCDUIHOHwaO‘I—‘Okoubh

HH

mmwmhmmqr—IOHAI-Imoww

\INI-‘I-‘I-‘Iwa‘JNIbIbbJI-‘WCDUI

I—II—I
\Ipmowooqoqqmwmmaxm

 

 

mummthI-J

\O

10
11
12
13
14
15
16
17
18
19
20
21

 

I-‘

I-‘H
NIkaHIwaI-‘belUJIb-w

OQWQmmmII-SI—‘HI-‘LDO

NHWI—‘WI—‘NO‘INQNQN

H
mOHofi-QOOQHWOHID-

WHxlmHuhI-‘OU'IU'I

- 1, 2, 3 and 4: Items of Credibility of Nongsim

- 5, 6, 7 and 8: Items of Attitude toward Nongsim
- 9, 10, 11 and 12: Items of Purchase Intention toward Nongsim Instant Noodle
- 13 and 14: Items of Familiarity with Nongsim
- 15, 16, 17, 18, 19, 20 and 21: Environmental Attitude

105

O‘HOJQWU'II-‘NNOX

...;

I-‘NIthOI-‘(Dwxlh

H

...:
NQUIWI—‘WQW

mwNHxlNOCh
I
I
I
I

Table 6.11. Error Rate of Heterogeneity (Binggrae in the Pre-test)

 

 

1 - - _ _ .. .. _ _ - .. -
2 - _ - .. _ - - _ - - _ _ _ - - _ _
3 _ _ _ _ - _ - - - _ .. .. - _ .. - ..
4 - _ - _ .. .. .. _ - - _ - _ - .. _ -
5 12 5 6 3 - - - - - - - - - _ - _ _
6 1 2 2 4 — — - - - - - - - - _ - _
7 5 2 2 1 — — _ - - - - - - - - _ _
8 1 2 2 3 - — - - - - - - - - _ - _
9 11 9 o 3 10 1 3 7 - - — - - - - _ -
10 1 1 7 7 1 8 6 o - - - - - - _ - -
11 1 2 5 5 o 6 3 4 - — - - - - _ - _
12 4 o o 2 5 1 2 2 3 5 1 - - - - - -
13 3 1 2 3 3 2 8 5 2 2 3 - - - — — -
14 3 2 1 1 9 2 2 4 o 6 9 2 3 - - - -
15 5 2 1 6 6 o 1 1 2 3 10 13 11 - - - -
16 o 1 3 7 2 3 1 3 9 7 2 6 1o - - - -
17 1o 13 13 6 7 9 7 4 3 1 3 7 o - - - -
18 1 2 1 2 3 5 4 2 6 9 12 6 1 - - — —
19 4 2 6 8 4 7 4 5 o 4 o 3 4 - - - —
20 6 o 4 5 8 5 2 3 1 3 1 8 10 - - - —
21 4 7 5 o 2 2 4 2 3 2 4 3 1 - - - -

 

 

 

mummbWMI-I
I
I
I
I

 

- l, 2, 3 and 4: Items of Credibility of Binggrae

- 5, 6, 7 and 8: Items of Attitude toward Binggrae

- 9, 10, 11 and 12: Items of Purchase Intention toward Binggrae Instant Noodle
- l3 and 14: Items of Familiarity with Binggrae

- 15, 16, 17, 18, 19, 20 and 21: Environmental Attitude

106

,__- _. 1,»

1’

p~u

IF-

 

Table 6.12

Error Rate of Heterogeneity (Samsung in the Post-test)

 

 

1 _ - - .. _ _ - - _ .. - - .. _ -
2 - - _ - - _ _ _ - .. _ _ - _ ..
3 .. .. .. .. .. _ _ _ _ _ ._ .. _. _ _
4 - - - ._ - - - _ .. _ _ - _. _ ..
5 16 16 3 2 - - — _ - - - - - _ _
6 11 9 6 1o — - - — - - - - - _ _
7 15 15 5 13 - - — — _ _ _ - _ _ _
8 14 9 2 10 - - — — _ - - _ - _ _
9 5 8 18 19 13 15 11 16 - — - - - - _
1o 14 13 1 1 6 6 10 6 - - - - — - _
11 13 11 2 2 7 6 9 4 - - - - - - -
12 21 20 13 28 5 2 1 o 8 8 2 - - - -
13 22 21 11 27 5 1 1 1 8 7 o - - - -
14 5 1 5 o 5 1 5 3 o 3 3 o o - -
15 1 o 6 1 5 2 7 1 o 5 3 o 2 - -

 

- 1, 2, 3 and 4: Items of Credibility of Samsung
- 5, 6, 7 and 8: Items of Attitude toward Samsung
- 9, 10, 11 and 12: Items of Purchase Intention toward Samsung Cellular Phone
- 13 and 14: Items of Stimuli for Samsung
- 15 and 16: Information Credibility

107

Table 6.13

Error Rate of Heterogeneity (Hyundai in the Post-test)

 

 

1 _ .. .. _. .. - _ - _ .. .. .. - _
2 .. _ - _ - - _ - _ _ _ - _ _
3 - _ - - .. - - - _ _ - _ - _
4 _ - _ _ - - _ - - - - ._ - -
5 12 13 9 7 - - — - - - - - - -
6 9 8 5 12 - — — - - _ _ _ - _
7 12 8 4 11 - - - - - - - - - -
8 1o 7 1 8 - — — _ - - - _ - -
9 10 12 2 4 12 6 5 7 - — - - - _
10 1o 7 1 2 6 3 2 2 - - - — - _
11 9 7 1 5 10 5 3 2 - - - — - -
12 20 18 11 22 10 4 2 1 6 5 3 - - -
13 19 18 13 27 10 6 6 5 1 4 7 - - -
14 4 1 1 2 4 4 2 3 2 2 4 o 3 -
15 1 1 1 4 6 2 4 5 4 1 1 3 1 -

 

- l, 2, 3 and 4: Items of Credibility of Hyundai
- 5, 6, 7 and 8: Items of Attitude toward Hyundai
- 9, 10, 11 and 12: Items of Purchase Intention toward Hyundai Cellular Phone
- 13 and 14: Items of Stimuli for Hyundai
- 15 and 16: Information Credibility

108

Table 6.14

Error Rate of Heterogeneity (Nongsim in the Post-test)

 

 

H

CDQO‘IU'IIbWNH
1.1
OIbwthU'lI-‘Ql—‘wm

\O
1.1

10
11
12
13
14
15

H
mhxlONNQONNO‘

I—‘H
meoIh-aswmqw»

l—‘l—l
mwmmwwommpw

...:
mommI-IH»
oasI-awmhw
\IwI-IIbtnmp
moowHI—Iq
0.8.9.01
I—‘NwH
I—‘NHO

l

I

I

- 1, 2, 3 and 4: Items of Credibility of Nongsim

- 5, 6, 7 and 8: Items of Attitude toward Nongsim

- 9, 10, 11 and 12: Items of Purchase Intention toward Nongsim Instant Noodle
- 13 and 14: Items of Stimuli for Nongsim

- 15 and 16: Information Credibility

109

Table 6.15

Error Rate of Heterogeneity (Binggrae in the Post-test)

 

 

1 _ - _ - _ _ _ _ _ - .. _ _ - _
2 - .. _ _ - - _ _ _ _ _ - - - _
3 - - - - - - _ _ - - - _ _ - _
4 _ - _ .. _ - - .. - ._ - _ .. _ _
5 17 11 6 4 - - - — - - - - - _ _
6 7 6 2 6 - - - - — - - - - _ _
7 14 11 7' 11 - - - - - — - - - - _
8 10 6 4 9 - — - - - - - - - - -
9 2 1 2 9 4 1 5 S - — - — - - -
10 10 5 1 8 2 6 0 2 — - - — _ .. _
11 10 6 1 9 O 4 1 l - — — — - - ..
12 15 15 13 20 4 2 S 1 4 1 O - — — —
13 18 16 12 19 9 0 5 0 2 2 5 - - — -
14 5 S 1 1 2 2 O 2 2 2 O O 2 - -
15 1 0 5 6 4 1 2 4 1 2 3 1 0 - -

 

- 1, 2, 3 and 4: Items of Credibility of Binggrae

- 5, 6, 7 and 8: Items of Attitude toward Binggrae

- 9, 10, 11 and 12: Items of Purchase Intention toward Binggrae Instant Noodle
- 13 and 14: Items of Stimuli for Binggrae

- 15 and 16: Information Credibility

Error rate of Heterogeneity table 7.8 shows that item 9 (i.e. the first question for
purchase intention toward samsung cellular phone) and item 8 (i.e. the fourth question
for attitude of Samsung) in the pre-test generates relatively high error rate (.25) to
maintain heterogeneity among factors. However, except the case of Samsung in the pre-
test, items 8 and 9 in other cases of measurements in the both pre and post-test do not
generate high error: Hyundai (.11), Nongsim (.04), Binggrae (.07) in the pre-test and
Samsung (.16), Hyundai (.07), Nongsim (.7), Binggrae (.5) in the post-test. Therefore,
this study concluded that it was not necessary to exclude items 8 and 9 from the original

measurements in order to raise discrimination among factors.

110

 

I
‘4 . .
Ul.‘\

I I
L40.

{:1 I"

UTSS

Items 12 and 13 of the scale “stimuli” generated a high error rate for
discrimination among factors in Samsung, Hyundai and Binggrae of the post-test (refer to
table 7.12, 7.13 and 7.15). However, the high error rate of the “stimuli” scale is not
significant because the scale “stimuli” is measuring whether the experimental stimuli is
successful or not and it is not all related to any scales of dependent variables. Therefore,
the “stimuli” scale does not weaken the discrimination among factors of dependent
variables.

Internal consistencies of factors are represented by reliability coefficient “Alpha
(OI)”.Standardized Alpha (a) generated by CFA for each scale in the pre and post-test is

presented in the following table 6.16.

Table 6.16. Measurement Reliability — Standardized OI

 

 

 

 

 

 

Pre-Test
Credibility Attitude Intention Familiarity Environmental Attitude
Samsung .80 .90 .86 .71
Hyundai .83 .93 .90 .85
Nongsim .83 .91 .93 .62
Binggrae .88 .93 .94 .81
All subjects .68
Post-Test
Credibility Attitude Intention Stimuli Information Credibility
Samsung .86 .93 .92 .97
Hyundai .88 .94 .94 .95
Nongsim .91 .94 .97 .96
Binggrae .89 .91 .95 .97
All subjects .89

 

111

CF A revealed that all scales except two retained high reliability that is internally
consistent within the factor. The two scales are “environmental attitude” (on =.68) and

“familiarity” (a = .62, .71, .81 and .85). Even though the reliability of these scales is not
strong, the validity of the two scales is strong because they maintain high discrimination
among factors (refer to table 7.8, 7.9, 7.10 and 7.11). Considering the point that validity
is a relatively more important criterion for a healthy measurement than reliability, it is
concluded that these two scales are still usefill measurements for this research.

In summary, the original measurements used in the pre-test and post-test retains
high discrimination among factors (i.e., validity) and high internal consistency within
factor (i.e., reliability). The results of CFA suggested that the data were fitted with the
posited construct models for pre-test measurements and for post-test measurements.
Inspection of the factor loadings and errors produced from the discrepancy between the

obtained and predicted correlations resulted in no exclusion of items.

Statistics for Data Analysis

Changes of attitude toward four corporations, credibility of corporations, and
purchase intension toward products (i.e., three major variables) afier stimuli were
measured. The four corporations were Samsung Electronics Cellular Phone Co., Hyundai
Electronics Cellular Phone Co., Nongsim Ramyeun Co. and Binggrae Ramyeun Co.9 The

new cellular phone products of Samsung and Hyundai were unnamed, as were the new

 

9 Samsung Electronics Cellular Phone Co. is denoted by Samsung, Hyundai Electronics Cellular Phone Co.
is denoted by Hyundai, Nongsim Ramyeun Co. is denoted by Nongsim and Binggrae Ramyeun Co. is
denoted by Binggrae.

112

instant noodles products of Nongsim and Binggrae“). Stimuli were negative or positive
information of the four corporations’ environmental performances (i.e., CEPI: corporate
environmental performance information).

The Within Subjects Analysis was performed to investigate individual changes of
three major variables. A Paired Samples T-test” was performed to gain 1) raw mean
scores of individual changes of three major variables (i.e., gain score = score afier stimuli
— score before stimuli), 2) one-directional confidence interval for the gain scores at 95 %
level, and 3) one-tailed significance for the gain scores at a =.05. Three major variables
were measured on the 7-point scale. For a stimuli (i.e., treatment) effect size of three
major variables, stimuli (i.e., treatment) correlation and one-sided confidence interval of
stimuli correlation at 95 % level was calculated by the statistical computer program
“Within (Within Subjects Analysis, 1995 version)”.”

When a directional hypothesis is tested, the Inference Probability makes sense.
Inference probability (IP) for these research hypotheses were gained by the “Within.” IP
is the probability that an effect occurs in the hypothesized direction. In other words, IP is
the probability that the directional hypothesis is correct. Thus, if a directional hypothesis
is positive, [P is the probability of rho (i.e. correlation of population) being positive (i.e.,
P (p > 0)). If a directional hypothesis is negative, IP is the probability of rho being
negative (i.e., P (p < 0)). The inverse concept to IP is referred to as Reverse Probability
(RP). RP is the probability that the effect occurs in the opposite direction to hypothesis.

In order words, The RP is the probability that the directional hypothesis is incorrect.

 

'0 The unnamed new cellular phones of Samsung and Hyundai are denoted by Samsung cellular phone and
Hyundai cellular phone. The unnamed new instant noodles of Nongsim and Binggrae are denoted by
Nongsim instant noodle and Binggrae instant noodle.

” It was conducted by a statistical computer program “SPSS” version 10.0.

'2 “Within” was developed by John E. Hunter.

113

Thus, RP is 1 minus IP (RP = 1 — IP) because the probability ranges from 0 to 1. For
example, If [P is .85, RP is .15 (1.0 -.85). For an interpretation of IP, this study follows

the suggestions from Ralph Levinel3 :

Table 6.17. Interpretation of Inference Probability

 

Inference Probability (IP) Interpretation

 

.667 IP Probably in the direction predicted

.333 (1/3) [P < .667 (2/3) Too close to call* and
Wait for further studies to confirm the hypothesis

IP < .333 Probably not in the direction predicted

 

 

"‘ The sample size is not large enough to conclude whether stimuli effect is in the
direction predicted.l4

The stimuli (i.e., treatment) correlations were corrected for measurement error
through considering measurement reliabilities of Time 1 (Pre-test) and Time 2 (Post-

test). '5 This correction was conducted through the “Withinw.” Thus correlation reported

 

'3 Levine’ suggestions about an interpretation of IP is cited from his unpublished paper “Messing Again
With the Correlation: Inference Techniques And the Mysterious Inference Probability (Revised 7/24/98).
Ralph Levine was a professor in the department of Psychology and is a professor in the department of
Resource Development at Michigan State University.

'4 The idea of inference probability comes from unpublished papers by Dr. John E. Hunter and Ralph
Levine. John E. Hunter is a professor in the Department of Psychology, and Ralph Levine was a professor
in the department of Psychology and is a professor in the department of Resource Development at
Michigan State University. For more detail on inference probability, refer to Appendix D. Inference
Probability.

'5 Standardized reliability was used for correction.

'6 The “Within” program describes corrections for measurement as follows: “The variance in gain scores
may or may not be evidence of an interaction. Either all or a large portion of the apparent individual
differences in gain may be caused by error of measurement rather than variation in the stimuli effect.”

114

in this study is corrected stimuli (i.e., treatment) correlation coefficient. Basic
interpretational wording for the stimuli effect size in this paper follows the following
rules: 1) Very Small or Very Weak: 0 < r < 0.1, 2) Small or Weak: 0.1 S r < 0.2, 3)
Moderate: 0.2 S r < 0.3, 4) Large or Strong : 0.3 S r < 0.4 and 5) Very Large or Very
Strong: 0.4 S r.

Between groups analysis (i.e., group mean comparison) also was performed to
compare attitude, credibility, and purchase intention changes after stimuli at the group
level using Independent Samples T-test”. One group was exposed to positive CEPI and
another was negative CEPI in the post-test. An independent sample t-test was conducted
to test the statistical significance at a =.05 and two-tailed level for group mean

comparison of three major variables.

Attitude toward Corporation

Hypothesis for attitude toward corporation (Hypothesis One: H1) posited that
CEPI disclosures change consumer attitude toward corporations positively for non-
polluting corporations and negatively for polluting corporations.

Basic descriptive statistics related to the variable “attitude toward four
corporations” are summarized in the following table 6.17. Stimuli effects of attitude

toward four corporations are summarized in the table 6.18.

 

'7 It was conducted by a statistical computer program “SPSS” version 10.0.

115

Table 6.18. Descriptive Statistics of Attitude toward Four Corporations

 

 

GroupA

Samsung Pre-test Mean 5.36 5.31
SD .95 .99

Post-test Mean 5.60 4.28

SD 1.06 1.33

D Mean .24 -1.03

SD .92 1.23

N 154 151

Hyundai Pre-test Mean 3.90 3.94
SD 1.10 .98

Post-test Mean 3.35 4.53

SD 1.27 .98

D Mean -.55 .59

SD 1.04 1.07

N 154 152

Nongsim Pre-test Mean 5.59 5.59
SD .89 .79

Post-test Mean 5.70 4.08

SD 1.10 1.30

D Mean .11 -1.50

SD 1.06 1.37

N 151 152

Binggrae Pre-test Mean 4.24 4.13
SD .88 1.00

Post-test Mean 3.51 4.74

SD 1.21 .86

D Mean -.73 .61

SD 1.05 1.03

N 149 152

 

- SD = Standard Deviation

- D = Gain Score (Gain score of post-test — Gain score of pre-test)
- N = Sample size

116

Table 6.19

Attitude Changes toward Four corporations

 

 

 

 

 

Group D t Sig 95% CI ofD (1-tailed) r 95% CI ofr(1-tailed) IP
Lower Upper Lower Upper

Samsung A: + .24 3.28 .001 .12 7.00 .12 -.01 1 .94
Samsung 8: - -1.03 -10.28 .000 -7.00 -.86 -.42 -1 -.31 1.00
Hyundai 8: + .59 6.81 .000 .45 7.00 .30 .18 1 1.00
Hyundai A: - -.55 -6.57 .000 -7.00 -.41 -.23 -1 -.11 1.00
Nongsim A: + .11 1.29 .100 -.03 7.00 .06 -.08 1 .76
Nongsim B: - -1.50 -13.52 .000 -7.00 -1.32 -.59 -1 -.51 1.00
Binggrae B: + .61 7.35 .000 .48 7.00 .33 .21 1 1.00
Binggrae A: - -.73 -8.46 .000 -7.00 -.59 -.34 -1 -.22 1.00

 

- A = Group A that had type A of stimuli in the post-test

- B = Group B that had type B of stimuli in the post-test

- "+" = Positive information of corporate environmental performance

II II

= Negative information of corporate environmental performance

- D = mean of raw gain score (Gain score = gain score after stimuli - gain score before stimuli)
- r = Point estimate of stimuli correlation coefficient

- IP (Inference Probability) = Probability that effect occurs in the hypothesized direction.

- Sig = p—value

117

Attitude toward Siamsung18

The direction and degree (i.e., size) of attitude changes in raw score by negative

and positive CEPI stimuli are visualized in the following Figure 6.1.

Figure 6.1. Attitude Change toward Samsung

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Attitude Change (Samsung)
A 7.00
2
3 6.00 #4
" 6.00 R
8 4 00 \ +Gr0Up A
V ' *-
§ 3.00 Gm“? B
2
,5 2.00
< 1.00
Pre-test Post-test
+ Group A 5.36 5.60
+ Group B 5.31 4.28

 

 

 

 

 

 

 

 

- Group A was exposed to positive CEPI.
- Group B was exposed to negative CEPI.
- Data reflect raw scores

The individual attitude change toward Samsung caused by negative CEPI (i.e.,
gain score) was significant: Dw = -1.03, t (150) = -10.28, p < .001, and CI = -7 to -
.86.The degree of attitude change (i.e., stimuli effect size) from negative CEPI is very

strong: r = -.42, CI = -1 to -.31.The probability that attitude change caused by the

negative CEPI occurs in the positive direction is 100 percent: IP = 1.00.

 

‘8 Dw = raw mean gain score within subjects, t= t-value from t-test, p = p-value, r = stimuli (i.e., treatment)
correlation coefficient, CI = confidence interval, P1 = inference probability, and Db = difference between

groups

118

The attitude gain score for Samsung by positive CEPI was significant: Dw = .24, t
(153) = 3.28, p = .001, and CI = .12 to 7. The effect size of attitude by positive CEPI is
small: r = .13, CI = -.01 to 1.The probability that attitude change caused by the positive
CEPI occurs in the positive direction is 94 percent: IP = .94.

The group mean difference of attitude toward Samsung between two groups that
had opposite information of stimuli each other is also significant: Db= 1.27, t (277)
=10.22 and p < .001. The subjects responded to negative CEPI more strongly (r = -.42)
than positive CEPI (r = .12).

The data are strongly consistent with H 1. It was statistically significant that
subjects’ attitudes toward Samsung changed negatively when they were exposed to
negative CEPI and changed positively when they were exposed to positive CEPI. The
extent of the attitude change caused by the negative was very strong but the extent of the
attitude change caused by positive CEPI was small. The probability that consumers
change their attitude to the negative direction when they are exposed to the negative
CEPI is 100 percent and change their attitude in the positive direction when they are
exposed to the positive CEPI is 94 percent. As a result, it can be concluded that the H 1 is

strongly confirmed in the case of consumer attitude toward Samsung.

Attitude towzflHvundai

The direction and degree of attitude changes in raw score by negative and positive

CEPI stimuli are presented in the following Figure 6.2.

119

Figure 6.2. Attitude Change toward Hyundai

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Attitude Change (Hyundai)
7.00
E 6.00
8
; 5.00
E +Group A
g 4.00 < +Gmup B
Z: 3.00
8 2.00
1.00
Pre-test Post-test
+ Group A 3.90 3.35
—a— Group B 3.94 4.53

 

 

 

 

 

 

 

 

- Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The individual attitude change toward Hyundai caused by negative CEPI was
significantsz = -.55, t (153) = -6.57, p < .001, and CI = -7 to -.41. The degree of
attitude change (i.e., effect size) from negative CEPI is moderate: r = -.23, CI = -1 to -.11.
The probability that attitude change caused by the negative CEPI occurs in the negative
direction is 100 percent: IP = 1.00.

The attitude gain score for Hyundai by positive CEPI was significant: Dw = .59, t
(151) = 6.81, p < .001, and CI = .45 to 7. Effect size of attitude by positive CEPI was
substantial: r = .30, CI = .18 to 1. The probability that attitude change caused by the

positive CEPI occurs in the positive direction is 100 percent: [P = 1.00.

120

The group mean difference in attitude between the two groups that had opposite
information of stimuli is also significant: Db= 1.14, t (304) =9.46 and p < .001. In
contrast to the case of Samsung, the subjects were a little more responsive to positive
CEPI (r = .30) than to negative CEPI regarding Hyundai (r = -.23).

The data are strongly consistent with H 1. It was statistically significant that the
subjects’ attitudes toward Hyundai changed negatively when they were exposed to
negative CEPI and changed positively when subjects were exposed to positive CEPI. The
degree of attitude change (i.e., stimuli effect size) caused by the negative and positive
CEPI are moderate and substantial. The probability that consumers change their attitude
in the negative direction when they are exposed to the negative CEPI is 100 percent and
change their attitude in the positive direction when they are exposed to the positive CEPI

is also 100 percent. Therefore, it can be concluded that the Hypothesis One is strongly

confirmed in the case of consumer attitude toward Hyundai.

Attitude toward Nongsim

 

The direction and degree of attitude changes in raw score by negative and positive

CEPI stimuli are presented in the following Figure 6.3.

121

Figure 6.3. Attitude Change toward Nongsim

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Attitude Change (Nongsim)
2::
o , fl
8 6 00 \
E p 4.00 \ + Group A
v ' +
g 3.00 Group 8
§ 2.00
“ 1.00
Pre—test Post-test
+ Group A 5.59 5.70
+ Group B 5.59 4.08

 

 

 

 

 

 

 

 

- Group A was exposed to positive CEPI.
- Group B was exposed to negative CEPI.
- Data reflect raw scores

The individual attitude change toward Nongsim caused by negative CEPI was
significant: Dw = -1.50, t (151) = -13.52, p < .001, and CI = -7 to —1.32. The degree of
attitude change (i.e., stimuli effect size) from negative CEPI is very strong: r = -.59, CI =
-1 to -.51. The probability that attitude change caused by the negative CEPI occurs in the
negative direction is 100 percent: IP = 1.00.

The attitude gain score for Nongsim by positive CEPI was barely significant at the
level of or = 0.05 and one-tailed significance test: Dw = .11, t (150) = 1.29, p = .100, and
CI = -.03 to 7. The stimuli effect size of attitude by positive CEPI was also very small: r
= .06, CI = -.08 to 1. The probability that change in attitude caused by the positive CEPI

occurs in the positive direction is not strong: IP = .76.

122

The group mean difference in attitude toward Nongsim between the two groups
that had opposite information of stimuli is also significant: Db= 1.61, t (284) =11.48 and
p < .001. Consistent with the case of Samsung, subjects responded to negative CEPI
much more strongly (r = -.59) than positive CEPI (r = .06).

The data are consistent with H 1. It was statistically significant that subjects’
attitude toward Nongsim became negative when they were exposed to negative CEPI and
became positive when they were exposed to positive CEPI. The degree of attitude change
(i.e., effect size) caused by the negative CEPI is very strong and the probability that
consumers change their attitude in the negative direction when exposed to the negative
CEPI is 100 percent. The degree in credibility change caused by the positive CEPI is very
small. However, the probability that consumers change their perception of credibility in
the positive direction when they are exposed to the positive CEPI is strong (IP = .76),
meaning that the effect probably occurs in the direction predicted. Meanwhile, the result
from the two-group comparison revealed that raw mean gain scores of two groups are
significantly different. Therefore, it can be concluded that in the case of consumer
attitude for Nongsim, the H 1 is confirmed in the case of both negative CEPI disclosure

and positive CEPI disclosure.

Attitude toward Binggrae
The direction and degree of attitude changes in raw score by negative and positive

CEPI stimuli are presented in the following Figure 6.4.

123

Figure 6.4. Attitude Change toward Binggrae

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Attitude Change (Binggrae)
7.00
g 6.00
I» 5.00
2' ° \ -+- Group B
2 3.00
2
35 2.00
1.00
Pre-test Post-test
+ Group A 4.24 3.51
+ Group B 4.13 4.74

 

 

 

 

 

 

- Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The individual attitude change toward Binggrae caused by negative CEPI was
significant: Dw = -.73, t (148) = -8.46, p < .001, and CI = -7 to -.59. The degree of
attitude change by negative CEPI is strong: r = -.34, CI = -1 to -.22. The probability that
attitude change caused by the negative CEPI occurs in the negative direction is 100
percent: IP = 1.00.

The attitude gain score for Binggrae by positive CEPI was significant: Dw = .61, t
(151) = 7.35, p < .001, and CI = .48 to 7. The degree of attitude change from positive
CEPI is also strong: r = .33, CI = .21 to 1. The probability that attitude change caused by

the positive CEPI occurs in the positive direction is 100 percent: IP = 1.00.

124

The group mean difference in attitude toward Binggrae between the two groups
that were given Opposite information is also significant: Db= 1.61, t (299) =11.20 and p <
.001. Subjects responded to negative CEPI (r = -.34) and positive CEPI (r = .33) in the
almost equal intensity.

The data are strongly consistent with H 1. It was statistically significant that
subjects’ attitudes toward Binggrae became negative when they were exposed to negative
CEPI and became positive when they were exposed to positive CEPI. The degree of
attitude change caused by the both negative and positive CEPI is substantial. The
probability that consumers change their attitude in the negative direction when they are
exposed to the negative CEPI is 100 percent and change their attitude in the positive
direction when they are exposed to the positive CEPI is 100 percent. As a result, it can be

concluded that H1 is strongly confirmed in the case of attitudes toward Binggrae.

Credibility of Corporation

The hypothesis for credibility about corporation (Hypothesis Two: H2) posited
that CEPI disclosures change consumer corporate credibility positively for non-polluting
corporations and negatively for polluting corporations.

Basic descriptive statistics related to the variable “credibility of four corporations”
are summarized in the following Table 6.19. Stimuli (i.e., Treatment) effects of

credibility of four corporations are summarized in the Table 6.20.

125

Table 6.20. Descriptive Statistics of Credibility for Four Corporations

 

 

GroupA GroupB

Samsung Pre-test Mean 5.21 5.14
SD .80 .87

Post-test Mean 5.59 4.33

SD .95 1.12

D Mean .38 -.81

SD .82 1.10

N 154 151

Hyundai Pre-test Mean 3.99 3.99
SD .98 .88

Post-test Mean 3.41 4.61

SD 1.14 .84

D Mean -.57 .62

SD 1.00 .93

N 154 152

Nongsim Pre-test Mean 5.34 5.31
SD .86 .75

Post-test Mean 5.62 4.10

SD 1.05 1.19

D Mean .28 -1.21

SD 1.04 1.21

N 154 152
Binggrae Pre-test Mean 4.20 4.17
SD .84 .89

Post-test Mean 3.53 4.72

SD 1.18 .84

D Mean -.67 .55

SD 1.16 .90

N 153 152

 

- SD= Standard Deviation
- D= Gain Score (Gain score of post-test — Gain score of pre-test)

- N= Sample size

126

Table 6.21

Credibility Changes of Four Corporations

 

Group D T Sig 95% CI ofD(1-tailed) r 95% CI ofr(1-tailed) lP

 

 

 

Lower Upper Lower Upper
Samsung A: + .38 5.71 .000 .27 7.00 .23 .10 1 1.00
Samsung 8: - -.81 -9.03 .000 -7.00 -.66 -.41 -1 -.30 1.00
Hyundai 8: + .62 8.21 .000 .50 7.00 .38 .26 1 1.00
Hyundai A: - -.57 -7.13 .000 -7.00 -.44 -.28 -1 -.15 1.00
Nongsim A: + .28 3.30 .001 .14 7.00 .15 .02 1 .97
Nongsim B: - -1.21 -12.35 .000 -7.00 -1.05 -.55 -1 -.46 1.00
Binggrae B: + .55 7.58 .000 .43 7.00 .33 .21 1 1.00
Binggrae A: - -.67 -7.19 .000 -7.00 -.52 -.33 -1 -.21 1.00

 

- A = Group A that had type A of stimuli in the post-test

- B = Group B that had type B of stimuli in the post-test

- "+" = Positive information of corporate environmental performance

- "-" = Negative information of corporate environmental performance

- D = mean of raw gain score (Gain score = gain score afier stimuli - gain score before stimuli)
- r = Point estimate of stimuli correlation coefficient

- IP (Inference Probability) = Probability that effect occurs in the hypothesized direction.

- Sig = p-value

Credibility of Samsung
The direction and extent (i.e., size) of credibility changes in raw score by negative

and positive CEPI stimuli are presented in the following Figure 6.5.

127

Figure 6.5. Credibility Change for Samsung

 

Credibility Change (Samsung)

 

 

 

 

 

 

 

 

 

 

 

1.: 7.00
§ 6.00 W
3 5.00
2 \x + Group A
S 4.00 ‘7‘“ Gro B
g 3.00 ”p
g 2.00
0 1.00
Pre-test Post-test
""' Group A 5.21 5.59
+GroggB 5.14 4.33

 

 

 

 

 

 

 

 

- Group A was exposed to positive CEPI.
- Group B was exposed to negative CEPI.
- Data are raw scores

The individual change of perception of credibility about Samsung caused by
negative CEPI (i.e., gain score) was significant: Dw = -.81, t (150)= -9.03, p < .001, and
CI = -7 to -.66. The degree of credibility change (i.e., stimuli effect size) by negative
CEPI is strong: r = -.41, CI = -1 to -.30. The probability that credibility change caused by
the negative CEPI occurs in the negative direction is 100 percent: IP = 1.00.

The credibility gain score for Samsung by positive CEPI was significant: Dw =
.38, t (153) = 5.71, p < .001, and CI = .27 to 7. Effect size ofcredibility bypositive CEPI
is moderate: r = .23, CI = .10 to 1. The probability that the change in credibility caused

by the positive CEPI occurs in the positive direction is 100 percent: IP = 1.00.

128

The group mean difference in credibility about Samsung between the two groups
that were presented with opposite information is also significant: Db= 1.18, t (277)
=10.66 and p < .001. Subjects responded to negative CEPI more strongly (r = -.41) than
positive CEPI (r = .23).

The data are strongly consistent with H 2. It was statistically significant that
subjects changed negatively their perception of credibility about Samsung when they
were exposed to negative CEPI and changed positively their perception of credibility
when they were exposed to positive CEPI. The extent of credibility change caused by the
negative and positive CEPI was strong and moderate. The probability that consumers
change their perception of credibility to the negative direction when they are exposed to
the negative CEPI is 100 percent and change them in the positive direction when they are
exposed to the positive CEPI is 100 percent. Therefore, it can be concluded that H 2 is

strongly confirmed in the case of credibility for Samsung.

Credibility of Hyundai

The direction and extent of credibility changes in raw score to negative and

positive CEPI stimuli are presented in the following Figure 6.6.

129

Figure 6.6. Credibility Change for Hyundai

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Credibility Change (Hyundai)
1? 7.00
§ 6.00
z .
g 5 00 p____ + Group A
S 4.00 - ——————1 +6 8
g 3.00 “NP
3 2.00
o 1.00
Pre-test Post-test
+ GroupA 3.99 3.41
+ Group B 3.99 4.61

 

 

 

 

 

 

 

— Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The individual change of perception of credibility about Hyundai caused by
negative CEPI was significant: Dw = -.57, t (153)= -7.13, p < .001, and CI = -7 to -.44.
The degree of credibility change (i.e., effect size) by negative CEPI is moderate: r = -.28,
CI = -1 to -.15. The probability that credibility change caused by the negative CEPI
occurs in the negative direction is 100 percent: [P = 1.00.

Credibility gain score for Hyundai by positive CEPI was significant: Dw = .62, t
(151) = 8.21, p < .001 and CI = .50 to 7. Effect size of credibility by positive CEPI was
large: r = .38, CI = .26 to 1. The probability that the credibility change caused by the

positive CEPI occurs in the positive direction is 100 percent: IP = 1.00.

130

The group mean difference in credibility between the two groups that had
opposite information of stimuli is also significant: Db= 1.20, t (304) =10.81 and p < .001.
The difference with the case of Samsung is that subjects responded more strongly to
positive CEPI (r = .38) than to negative CEPI of Hyundai (r = -.28).

The data are strongly consistent with H 2. It was statistically significant that
subjects reduced their perception of credibility about Hyundai when they were exposed to
negative CEPI and increased their perception of credibility when they were exposed to
positive CEPI. The extent of credibility change caused by the negative and positive CEPI
was moderate or strong. The probability that consumers change their perception of
credibility in the negative direction when they are exposed to negative CEPI is 100
percent and change them in the positive direction when they are exposed to the positive
CEPI is also 100 percent. Therefore, it can be concluded that H 2 is strongly confirmed in

the case of consumer’s perception of credibility for Hyundai.
Credibility of Nongsim

The direction and extent of credibility changes in raw score to negative and

positive CEPI stimuli are visualized in the following Figure 6.7.

131

Figure 6.7. Credibility Change for Nongsim

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Credibility Change (Nongsim)
2:; 7.00
3 6.00 H
3 5.00 K +
g 4 00 \ Group A
' +-
g 3.00 Group B
’3 2.00
o 1.00
Pre-test Post-test
+ Group A 5.34 5.62
+Group B 5.31 4.10

 

 

 

 

 

 

 

 

- Group A was exposed to positive CEPI.
- Group B was exposed to negative CEPI.
- Data are raw scores

The individual perception change of credibility about Nongsim caused by
negative CEPI was significantsz = -1.21, t (151) = -12.35, p < .001, and CI = -7 to —
1.05. The degree of credibility change (i.e., effect size) from negative CEPI is very
strong: r = -.55, CI = -1 to -.46. The probability that credibility change caused by the
negative CEPI occurs in the negative direction is 100 percent: [P = 1.00.

The credibility gain score for Nongsim by positive CEPI was significant: Dw =
.28, t (153) = 3.30, p = .001, and CI = .14 to 7. The stimuli effect size of credibility by
positive CEPI was small: r = .15, CI = .02 to 1, but the probability that credibility change

caused by the positive CEPI occurs in the positive direction is very strong: [P = .97.

132

The group mean difference (i.e., two group comparison) of credibility about
Nongsim between two groups that had been presented with opposite information is
significant: Db= 1.49, t (296) =11.56 and p < .001. Consistent with the case of Samsung,
subjects responded to negative CEPI more strongly (r = -.55) than positive CEPI (r =
.15).

The data are strongly consistent with H 1. It was statistically significant that
subjects’ perception of credibility for Nongsim became negative when they were exposed
to negative CEPI and became positive when they were exposed to positive CEPI. The
degree of attitude change (i.e., effect size) caused by the negative CEPI is very strong and
the probability that consumers change their perception of credibility in the negative
direction when exposed to the negative CEPI is 100 percent. While, the degree of
credibility change caused by the positive CEPI is weak, but the probability that
consumers change their perception of credibility in the positive direction when they are
exposed to the positive CEPI is 97 percent. Therefore, it can be concluded that the H 2 is

strongly confirmed in the case of consumer perception of credibility for Nongsim.

Credibilifl of Binggrae
The direction and extent of credibility changes in raw score to negative and

positive CEPI stimuli are presented in the following Figure 6.8.

133

Figure 6.8. Credibility Change for Binggrae

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Credibility Change (Binggrae)
3.. 7.00
2
§ 6.00
3 5.00
g 4 00 K/ + Group A
.3." ' \K + Group B
S 3.00
E 2.00
o
1.00
Pre-test Post-test
+ Group A 4.20 3.53
+ Group B 4.17 4.72

 

 

 

 

 

 

 

 

- Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The individual perception change of credibility about Binggrae caused by
negative CEPI was significant: Dw = -.67, t (152) = -7.19, p < .001, and CI = -7 to -.52.
The degree of change in credibility (i.e., effect size) caused by negative CEPI was strong:
r = -.33, CI = -1 to -.21. The probability that change in credibility caused by the negative
CEPI occurs in the negative direction is 100 percent: [P = 1.00.

The credibility gain score for Binggrae resulting from positive CEPI was
significantsz = .55, t (151) = 7.58, p < .001, and CI = .43 to 7. Effect size of
credibility by positive CEPI was large: r = .33, CI = .21 to 1. The probability that the
change in credibility caused by the poSitive CEPI occurs in the positive direction is 100

percent: [P = 1.00.

134

The group mean difference in credibility about Binggrae between the two groups
that were given opposite information is also significant: Db= 1.23, t (286) =10.33 and p <
.001. Subjects responded to negative CEPI (r = -.33) and positive CEPI (r = .33) in the
equal intensity.

The data are strongly consistent with H 2. It was statistically significant that
subjects reduced their perception of credibility change about Binggrae when they were
exposed to negative CEPI and increased them when they were exposed to positive CEPI.
The degree of credibility change caused by the both negative and positive CEPI is strong.
The probability that consumers change their perception of credibility in the negative
direction when they are exposed to the negative CEPI is 100 percent and change their
perception of credibility in the positive direction when they are exposed to the positive
CEPI is also 100 percent. As a result, it can be concluded that H 2 is strongly confirmed

in the case of credibility for Binggrae.

Purchase Intention toward Products

The hypothesis for purchase intention toward products (Hypothesis Three: H 3)
posited that CEPI disclosures increase consumer purchase intention for the products of
non-polluting corporations and decrease consumer purchase intention for the products of
polluting corporations.

Basic descriptive statistics related to the variable “purchase intention toward
products of the four corporations” are summarized in the following Table 6.21. Stimuli

effects of purchase intention toward four products are summarized in the Table 6.22.

135

Table 6.22. Descriptive Statistics of Purchase Intention

 

 

GroupA

Samsung Pre-test Mean 4.68 4.73
SD 1.33 1.35

Post-test Mean 5.1 1 3.84

SD 1.23 1.76

D Mean .43 -.89

SD 1.26 1.56

N 151 152

Hyundai Pre-test Mean 3.04 3.04
SD 1.28 1.24

Post-test Mean 2.79 4.10

SD 1.32 1.39

D Mean -.25 1.06

SD 1.16 1.38

N 151 152

Nongsim Pre-test Mean 5.54 5.66
SD 1.19 1.06

Post-test Mean 5.69 4.00

SD 1.14 1.61

D Mean .15 -1.66

SD 1.19 1.86

N 153 151
Binggrae Pre-test Mean 4.19 4.29
SD 1.25 1.43

Post-test Mean 3.49 4.82

SD 1.44 1.14

D Mean -.70 .53

SD 1.38 1.47

N 153 151

 

- SD = Standard Deviation

- D = Gain Score (Gain score of post-test — Gain score of pre-test)
- N = Sample size

136

Individual Purchase Intention Changes toward Four Products

Table 6.23

 

 

 

 

Group D t Sig 95% CI of D (Hailed) r 95% CI of r (1 -tailed) IP
Lower Upper Lower Upper

Samsung A: + .43 4.17 .000 .26 7.00 .18 .05 1 .99
Samsung 8: - -.89 -7.01 .000 -7.00 -.68 -.28 -1 -.16 1.00
Hyundai 8: + 1.06 9.50.000 .83 7.00 .39 .27 1 1.00
Hyundai A: - -.25 -2.64 .004 -7.00 -.09 -.10 -1 .03 .89
Nongsim A: + .15 1.52 .066 -.01 7.00 .06 -.07 1 .79
Nongsim B: - -1.66 -10.98 .000 -7.00 -1.41 -.53 -1 -.44 1.00
Binggrae B: + .53 4.43 .000 .33 7.00 .21 .08 1 1.00
Mme A: - -.70 -6.33 .000 -7.00 -.52 -.26 -1 -.14 1.00

 

- A = Group A that had type A of stimuli in the post-test
- B = Group B that had type B of stimuli in the post-test

- "+" = Positive information of corporate environmental performance
- "-" = Negative information of corporate environmental performance

- D = mean of raw gain score (Gain score = gain score after stimuli - gain score before stimuli)
- r = Point estimate of stimuli correlation coefficient

-'IP (Inference Probability) = Probability that effect occurs in the hypothesized direction.
- Sig = p—value

Purchase Intention toward Samsung Cellular Phone

The direction and extent of change in purchase intention in raw score to negative

and positive CEPI stimuli are presented in the following Figure 6.9.

137

Figure 6.9. Purchase Intention Change toward Samsung Cellular Phone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Purchase Intention Change (Samsung)
.. 7.00
2
§ 6.00
z
E 5.00 <9
C
o —a— Group A
-.-. 4.00
E; N —7<— Group B
5 3.00
8
g 2.00
3
e 1.00
Pre-test Post-test
+ Group A 4.68 5.11
—7<— Group B 4.73 3.84

 

 

 

 

 

 

 

 

- Group A was exposed to positive CEPI.
— Group B was exposed to negative CEPI.
- Data reflect raw scores

The changes in individual purchase intention toward the Samsung cellular phone
related to negative CEPI (i.e., gain score) was significant: Dw = -.89, t (151) = -7.01, p <
.001, and CI = -7 to -.68. The degree of change in purchase intention resulting fi'om
negative CEPI (i.e., effect size) was moderate: r = -.28, CI = -1 to -.16. The probability
that changes in purchase intention related to negative CEPI occur in the negative
direction is 100 percent: IP = 1.00.

The gain score in purchase intention for Samsung cellular phone resulting from
positive CEPI was significant: Dw = .43, t (150) = 4.17, p < .001, and CI = .26 to 7.

Effect size of purchase intention by positive CEPI is small: r = .18, CI = .05 to 1.

138

However, the probability that purchase intention change caused by the positive CEPI
occurs in the positive direction is 99 percent: IP = .99.

The group mean difference in purchase intention toward Samsung cellular phone
between the two groups that were presented with opposite information is also significant:
Db= 1.31, t (289)=8.07 and p < .001. Subjects responded to negative CEPI more strongly
(r = -.28) than positive CEPI (r = .18).

The data are strongly consistent with H 3. It was statistically significant that the
subjects changed their purchase intention negatively toward Samsung cellular phone
when they were exposed to negative CEPI and changed their purchase intention
positively toward Samsung cellular phone when they were exposed to positive CEPI. The
degree of purchase intention change related to the negative information was moderate and
the probability that consumers would change their purchase intention in the negative
direction when they are exposed to the negative CEPI is 100 percent. The extent of
change in purchase intention resulting from positive CEPI is small, but the probability
that consumers change their purchase intention in the positive direction when they are
exposed to the positive CEPI is 99 percent. Therefore, it can be concluded that H 3 is
strongly confirmed in the case of consumer purchase intention toward Samsung cellular

phone.

Purchase Intention towafrd Hyundai Cellular Phone

The direction and extent of change in purchase intention in raw score to negative

and positive CEPI stimuli are visualized in the following Figure 6.10.

139

Figure 6.10. Purchase Intention Change toward Hyundai Cellular Phone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Purchase Intention (Hyundai)

9
§ 7.00
E, 6.00
2’ 5.00 G A
O
E 4.00 A —7<— roup
o + Group B
*_a 3.00 —
8 2.00
(U
f, 1.00
5 Pre-test Post-test
n.

—-,<— Group A 3.04 2.79

—a— Group B 3.04 4.10

 

 

 

 

 

 

 

 

- Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The change in individual purchase intention toward Hyundai cellular phone
related to negative CEPI was significant: Dw = -.25, t (150) = -2.64, p = .004, and CI = -
7 to -.09. The degree of purchase intention change (i.e., effect size) resulting from
negative CEPI is small: r = -.10, CI = -1 to -.03. The probability that changes in purchase
intention change resulting from the negative CEPI occur in the negative direction is 89
percent: IP = .89.

The gain score in purchase intention for Hyundai cellular phone resulting from
positive CEPI was significant: Dw = 1.06, t (151) = 9.50, p < .001, and CI = .88 to 7.

Effect size of purchase intention by positive CEPI is large: r = .39, CI = .27 to 1. The

140

probability that changes in purchase intention change resulting from positive CEPI occur
in the positive direction is 100 percent: [P = 1.00.

The group mean difference in purchase intention between two groups that were
presented with opposite information is also significant: Db= 1.31, t (301) = 8.96 and p <
.001. In contrast to the case of Samsung, the subjects responded a little more strongly to
positive CEPI (r = .39) than to negative CEPI for Hyundai (r = -.10).

The data is strongly consistent with H 3. It was statistically significant that the
respondents changed their purchase intention negatively toward Hyundai cellular phone
when they were exposed to negative CEPI and changed their purchase intention
positively toward Hyundai cellular phone when they were exposed to positive CEPI. The
extent of change in purchase intention related to negative and positive CEPI is
respectively small and large. However, the probability that consumers change their
purchase intention in the negative direction when they are exposed to the negative CEPI
is 89 percent (i.e., probably in the direction predicted) and change their purchase
intention in the positive direction when they are exposed to the positive CEPI is 100
percent.

Therefore, it can be concluded that in the case of consumer purchase intention
toward Hyundai cellular phone, H 3 is confirmed in the case of negative CEPI disclosure

and it is strongly confirmed in the case of positive CEPI disclosure.

Purchase Intention toward Nongsim Instant Noodle

The direction and degree of changes in purchase intention in raw score to negative

and positive CEPI stimuli are presented in the following Figure 6.11.

141

Figure 6.11. Purchase Intention Change toward Nongsim Instant Noodle

 

 

 

 

 

Purchase Intention Change (Nongsim)

 

 

 

—a— Group A
—7<— Group B

 

 

 

 

 

 

 

 

§ 7.00

8 6 00

a u

E 5 00 é\

§ \

O

:5” 3.00

3

g 2.00

2

a 1.00

Pre-test Post-test

—a— Group A 5.54 5.69
+ Group B 5.66 4.00

 

 

 

 

 

— Group A was exposed to positive CEPI.
- Group B was exposed to negative CEPI.

- Data reflect raw scores

The change in individual purchase intention toward Nongsim instant noodle
related to negative CEPI was significant: Dw = -1.67, t (150) = -10.98, p < .001, and CI
= -7 to ——1 .41. The extent of change in purchase intention (i.e., effect size) resulting from
negative CEPI is very strong: r = -.53, CI = -1 to -.44. The probability that change in

purchase intention related to negative CEPI occurs in the negative direction is 100

percent: IP = 1.00.

However, the gain score in the purchase intention for Nongsim instant noodle
related to positive CEPI was significant (in one-tailed significance test at or = 0.05 level):
Dw = .15, t (152) = 1.52, p = .066, and CI = -.01 to 7. The effect size ofpurchase

intention by positive CEPI is also very small: r = .06, CI = -.07 to 1. The probability that

142

 

 

change in purchase intention related to the positive CEPI occurs in the positive direction
is not strong: [P = .79.

The group mean difference in purchase intention toward Nongsim instant noodle
between the two groups that were presented with opposite information is significant: Db=
1.81, t (255) =10.08 and p < .001. Consistent with the case of Samsung, subjects
responded to negative CEPI much more strongly (r = -.53) than positive CEPI (r = .06).

The data is strongly consistent with H 3 in the case of negative CEPI stimuli but
are weakly consistent with H 3 in the case of positive CEPI stimuli. The subjects changed
their purchase intention negatively toward Nongsim instant noodle when they were
exposed to negative CEPI and changed positively their purchase intention toward
Nongsim instant noodle when they were exposed to positive CEPI was statistically
significant. The degree of purchase intention change resulting from negative CEPI is very
substantial and the probability that consumers would change their purchase intention in
the negative direction when they are exposed to the negative CEPI is 100 percent. The
degree of change in purchase intention related to the positive CEPI is very weak,
however the probability that consumers change their purchase intention in the positive
direction when they are exposed to the positive CEPI is still large ([P = .79), meaning
that effect occurs probably in the direction predicted. Meanwhile, the result from the two-
group comparison revealed that raw mean gain scores of two groups are significantly
different. Therefore, it can be concluded that in the case of consumer purchase intention
toward Nongsim instant noodle, H 3 is strongly confirmed in the case of negative CEPI

disclosure and is also confirmed in the case of positive CEPI disclosure.

143

Purchase Intention toward Binggrae Instant Noodle

The direction and extent of change in purchase intention in raw score related to

negative and positive CEPI stimuli are presented in the following figure 6.12.

Figure 6.12. Purchase intention Change toward Binggrae Instant Noodle

 

Purchase Intention Change (Binggrae)

 

7.00
6.00

5.00 (A
4.00 \

 

 

 

 

 

Purchase Intention (raw score)

 

 

 

 

 

 

 

 

 

 

3.00
2.00
1.00 ‘
Pre-test Post-test + Group A
+ Group B
+Group A 4.19 3.49
+ Group B 4.29 4.82

 

 

 

 

 

- Group A was exposed to negative CEPI.
- Group B was exposed to positive CEPI.
- Data reflect raw scores

The change in individual purchase intention toward Binggrae instant noodle
related to negative CEPI was significant: Dw = -.70, t (152) = -6.33, p < .001, and CI = -
7 to -.52. The degree of purchase intention change (i.e., effect size) by negative CEPI is
moderate: r = -.26, CI = -1 to -. 14. The probability that change in purchase intention

related to the negative CEPI occurs in the negative direction is 100 percent: [P = 1.00.

144

The gain score in the purchase intention for Binggrae instant noodle by positive
CEPI was significant: Dw = .53, t (150) = 4.43, p < .001, and CI = .33 to 7. Effect size of
purchase intention by positive CEPI was moderate: r = .21, CI = .08 to 1. The probability
that changes in purchase intention related to the positive CEPI occur in the positive
direction is 100 percent: [P = 1.00.

The group mean difference in purchase intention toward Binggrae instant noodle
between the two groups that were presented with opposite information is also significant:
Db= 1.23, t (302) =7.55 and p < .001. The subjects responded to negative CEPI (r = -.26)
and to positive CEPI (r = .21) in the almost equal intensity.

The data are strongly consistent with H 3. It was statistically significant that
subjects reduced their purchase intention toward Binggrae instant noodle when they were
exposed to negative CEPI and increased their purchase intention toward Binggrae instant
noodle when they were exposed to positive CEPI. The degree of purchase intention
change caused by the both negative and positive CEPI is moderate. The probability that
consumers would change their purchase intention in the negative direction when they are
exposed to negative CEPI is 100 percent and change their purchase intention in the
positive direction when they are exposed to positive CEPI is 100 percent. As a result, it
can be concluded that H 3 is strongly confirmed in the case of purchase intention toward

Binggrae.

Stimuli by Subject Interaction

In the absence of error of measurement, any variation in gain scores can be

attributed to the Stimuli (i.e., treatment) by Subject Interaction. If there is no Stimuli by

145

Subject Interaction, all individuals have the same change score (i.e., standard deviation of
gain score , SD Av = 0). The degree of Stimuli by Subject Interaction can be measured
by the followings:

1. Raw score SD “STG” (i.e., Standard deviation of gain score)

2. Standard score SD “5” (STG / Within group SD)

3. Self-impact correlation “ir.”

If there are interactions between stimuli and subjects, there would be differences
in the effect size that can be explained by different initial level. For example, subjects
who have strong attitude toward Samsung may be little sensitive to any information of
Samsung, but subject who have weak attitude toward Samsung may strongly response to
information about Samsung. This is measured by the extent of the correlation between
initial level and effect size. This correlation is called the self-impact correlation (i.e., ir).
The sizes of the Stimuli by Subject Interaction related to Attitude, Credibility and
Purchase Intention changes are summarized in the following Table 6.23. The Stimuli
(i.e., Treatment) by Subject Interaction reported in this Table 6.23 is calculated based on
the corrected basic statistics for measurement error. Statistical significance of the Stimuli
by Subject Interaction is tested at or = .05 and one-tailed level, assuming that distribution

of STG and ir are in a normal curve.

146

Table 6.24

Stimuli by Subject Interaction

 

 

 

Attitude Credibility Purchase Intention

s STG ir S STG ir s STG ir
Samsung A .852 .827 -.287 .815 .661 -.142 .864 1.006 -.528

it it *i it it N Sig it it it
Samsung B .994 1.112 -.259 1.015 .917 -.295 .951 1.434 -.186
Hyundai A .831 .958 -.262 .896 .895 -.278 .834 1.038 -.390
Hyundai B 1.046 .978 -.536 .976 .744 -.520 1 .004 1.264 -.389
Nongsim A 1.027 1.001 —.342 1.039 .948 -.350 .988 1.118 -.51 7
Nongsim B 1.251 1.275 -.341 1.185 1.080 -.273 1.366 1.811 -.499
Binggrae A .954 .977 -.188 1.096 1.072 -.282 .985 1.277 -.370
Binggrae B 1.048 .912 -.667 .955 .760 -.564 1.121 1.415 -.686

 

- s = Standard Score SD (Standard Deviation)

- STG = Raw Score SD

- it = Self-impact correlation

- *"' = Statistically Significant, alpha = .5, two-tailed test
- N sig = Not significant

Table 6.23 shows that except for a self-impact correlation (ir = -.142) related to
credibility change of Samsung Group A, all Stimuli by Subject Interaction of STG, s and
ir are statistically significant at or = .05, one-tailed level.'9 It means that masking

variables are affecting the extent of the effect of dependent variables: attitude, credibility

 

'9 The fact that standard deviation is not a zero and statistically significant, means that l) masking variables
that are unknown variables affecting the intensity of influence of CEPI on attitude, credibility and purchase
intention, 2) situational random factors (e.g., hearing a news about Samsung Electronics’ bad behavior just
before administration of the post-test), or 3) random fimctioning of human psychology (e.g., participants’
though or feeling during the experiment) were intervened in the experiment. Moderator variable is a special
case of masking variable.

147

rl

and purchase intention. It also implies that there are moderator variables that influence
the effect size of three dependent variables: attitude, credibility and purchase intention.

Therefore, it is relevant to investigate moderators for the three dependent variables.

Environmental Attitude as a Moderator20

It is hypothesized that a positive relationship exists between consumer
environmental attitude toward pollution and consumer purchase behavioral change
caused by CEPI disclosure. Thus, the fourth hypothesis (H 4) posited that a customer
who has strong environmental attitudes against pollution (EAP) changes the three
dependent variables more than a customer who has low EAP. In other words, a positive
relationship exists between EAP and the three dependent variables: attitude toward
corporation (AC), corporate credibility (CC), and purchase intention (PI). Data analysis

results related to H 4 are presented in the following Table 6.24.

 

2° Moderator is defined as a variable that intervenes between independent and dependent variable and
influences the degree of dependent variable. However, it is neither an independent variable nor a mediator
variable. Moderator variable is a special case of the masking variable. Masking variables are unknown
variables that affect the defendant variable. Meanwhile, Mediator variable is a middle variable located
between the first and the final variable in the causal diagram so that mediator is both independent and
dependent variable. For more details, see chapter VII, p. 221.

148

Table 6.25

Environmental Attitude toward Pollution and Effect Size of Three Variables

 

 

 

Environmental Attitude Credibility Purchase Intention
Attitude& r sil IP H4 r sig IP H4 r sig IP H4
Samsung A:+ .09 .126 .88 C -.03 .361 .36 F .04 .312 .69 C
Samsung 82- -.26 .001 1.00 C -.33 .000 1.00 C -.34 .000 1.00 C
Hyundai A:- -.09 .128 .87 C -.13 .061 .94 C -.24 .002 1.00 C
Hyundai B:+ .10 .110 .89 C .04 .310 .69 C .01 .471 .53 F
Nongsim A:+ .14 .045 .96 C .13 .054 .95 C .08 .169 .83 C
Nongsim B:- -.30 0001.00 C -.21 .006 1.00 C -.20 .006 1.00 C
Binggrae A:- -.07 .197 .80 C -.20 .007 .99 C -.20 .007 .99 C
Mme B:+ -.01 .467 .47 F .07 .185 .82 C .10 .119 .88 C

 

- A = Group A / B = Group B in the experiment.
- Positive CEPI was given to Samsung A, Hyundai B, Nongsim A and Binggrae B
- Negative CEPI was given to Samsung B, Hyundai A, Nongsim B and Binggrae A
- + / - = Direction of r required to confirm H 4 under CEPI stimuli given
- H 4 = positive correlation (+)
- "-" is resulted from H 4 ( + ) and negative CEPI (-)
- "+" is resulted from H 4 ( + ) and positive CEPI (+)
- C = Confirmed / F = need Further research / N = Not confirmed
- r = Pearson correlation coefficient
- [P = Inference Probability
- sig = p-value

Environmental Attitude and Attitude toward Corporation

Group B’s attitude for Binggrae revealed the opposite directional correlation to H
4: r = -.01, p = .467, and [P = .47. The observed opposite correlation between
environmental attitude and attitude about Binggrae is extremely weak (r = -.01) and also
not statistically significant (p = .467 at the one-tailed or = 0.05 level). Reverse probability
(RP = .53) is almost equivalent to inference probability ([P = .47). Thus, the observed

correlation could be considered as “no correlation.” However, inference probability (IP =

149

.47) falls between .333 and .667 ( 1/3 < [P < 2/3 ) so the sample size is not large enough
to draw a conclusion. For a sound conclusion, we would need further research to decide
“no correlation” related to H 4 in the case of positive CEPI stimuli about Binggrae
(Group B of Binggrae).

Four groups reveal that data are consistent with H 4: Group A of Samsung (r =
.09, p = .126, and [P = .88), Group A onyundai (r = -.09, p = .128 and [P = .87), Group
B onyundai (r = -.10, p = .110 and [P = .89) and Group A ofBinggrae (r= -.O7, p = .197
and [P = .80). In these four groups, the observed correlations were not statistically
significant at the one-tailed or = 0.05 level, but inference probability is moderately high. It
means that the probability that the correlation between environmental attitude and
attitude toward corporation is positive (i.e., in the direction hypothesized) is much higher
than the reverse probability that the correlation is negative. Therefore, it is concluded that
H 4 is confirmed in the cases of Group A of Samsung, Group A of Hyundai, Group B of
Hyundai and Group A of Binggrae.

The data from three groups are strongly consistent with H 4: Group B of Samsung
(r = -.26, p = .001 and IP = 1.00), Group A of Nongsim (r = .14, p = .045 and [P = .96)
and Group B of Nongsim (r = -.30, p < .001 and [P = 1.00). In these three groups, the
observed correlations were considerably significant at the one-tailed or = 0.05 level and
the inference probabilities are strongly high. Thus, the researcher concludes that H 4 is
strongly confirmed in the cases of Group B of Samsung, Group A of Nongsim and Group

B of Nongsim.

150

Figure 6.13. Inference Probability of H4 for Attitude toward Corporation

 

 

 

 

 

 

 

 

 

 

 

 

 

Environmental Attitude &
Attitude to Corporation
'5
In
.1:
E
a .50
o
c
2
.2
I:
"' '00 A B A B A 8 A 8
Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
[+Series1 .88 1.00 .87 .89 .96 1.00 .80 .47

 

 

 

Figure 6.13 summarizes that except for one case of positive CEPI stimuli about
Binggrae (Group B of Binggrae), H 4 is confirmed in all cases (i.e., 7 of 8 cases). For a
conclusion concerning Binggrae group B, further research is needed. Thus, the researcher

concludes that “environmental attitude toward pollution” is a moderator variable for the

variable “attitude toward corporation.”

Environmental Attitude and Corporate Credibility

Group A’s perception of credibility for Samsung revealed the opposite directional

correlation to H 4: r = -.03, p = .361, and [P = .36. For H 4 to be confirmed, the

correlation between environmental attitude and credibility of Samsung should be positive
because Group A was exposed to positive CEPI of Samsung and hypothesized direction

of H 4 is positive. However, the observed correlation is negative (r = -.03). Meanwhile,

151

 

 

the observed correlation is not statistically significant (p = .361 at the one-tailed or = 0.05
level). Thus, the observed negative correlation (r = -.03) could be generated from
sampling error (Sample size of Group A = 154). Inference probability (IP = .36) falls
between .333 and .667 ( 1/3 < [P < 2/3 ) so the sample size is not large enough to draw a
conclusion. In order to draw a sound conclusion, we would need further research to
decide disconfirmation of H 4 in the case of positive CEPI stimuli about Samsung (Group
A of Samsung). I

The two groups reveal that data are consistent with H 4: Group B of Hyundai (r =
-.04, p = .310 and [P = .69) and Group B ofBinggrae (r = -.07, p = .185 and IP = .82). In
the two groups, the observed correlations were not statistically significant at the one-
tailed or = 0.05 level. However, the inference probability is larger than .667. This means
that the probability that the correlation between environmental attitude and corporate
credibility is positive (i.e., in the direction hypothesized) is moderately larger than the
reverse probability that the correlation is negative. Therefore, the researcher concludes
that H 4 is confirmed in the cases of Group B of Hyundai and Group B of Binggrae.

Five groups revealed that data are strongly consistent with H 4: Group B of
Samsung (r = -.33, p < .001 and IP = 1.00), Group A onyundai (r = -.13, p = .061 and
IP = .94), Group A of Nongsim (r = .13, p = .054 and [P = .95), Group B of Nongsim (r =
-.21, p = .006 and IP = 1.00), and Group A of Binggrae (r = -.20, p = .007 and [P = .99).
In these five groups, the observed correlations were considerably significant at the one-
tailed or = 0.05 level and the inference probabilities are strongly high. Thus, this study
concludes that H 4 is strongly confirmed in the cases of Group B of Samsung, Group A

of Hyundai, Group A of Nongsim, Group B of Nongsim, and Group A of Binggrae.

152

Figure 6.14. Inference Probability of H4 for Corporate Credibility

 

Environmental Attitude &
Corporate Credibility

 

 

 

 

 

 

 

 

 

 

 

 

B

I!

.0

.9 50

8 /

c

2

:3 00

" ' A B A B A 8 A B

Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae

LSeries1 .36 1.00 .94 .69 .95 1.00 .99 .82

 

 

 

 

 

Figure 6.14 shows that except for one case of positive CEPI stimuli about
Samsung (Group A of Samsung), H4 is confirmed in all cases (i.e., 7 of 8 cases). For the
conclusion about Samsung Group A, further research is needed. Thus, this study
concludes that “environmental attitude toward pollution” is a moderator variable for the

variable “corporate credibility.”

Environme_ntzg Attitude and Purchgse Intention

Group B’s purchase intention for Hyundai’s cellular phone revealed “no”
correlation: r = .01, p = .471, and [P = .53. The observed correlation between
environmental attitude and purchase intention is very weak (r = .01) and also not

statistically significant (p = .471 at the one-tailed a = 0.05 level). The inference

153

probability ([P = .53) is almost equivalent to reverse probability (RP = 47). Thus, the
observed correlation could be considered as “no correlation.” However inference
probability (IP = .53) falls between .333 and .667 ( 1/3 < [P < 2/3 ) so the sample size is
not large enough to draw a conclusion. Further research is needed to decide “no
correlation” in relation to H 4 in the case of positive CEPI stimuli about Hyundai (Group
B of Hyundai).

Three groups reveal that data are consistent with H 4: Group A of Samsung (r =
.04, p = .312, and IP = .69), Group A ofNongsim (r = .08, p = .169 and IP = .83), and
Group B of Binggrae (r = .10, p = .119 and [P = .88). In these three groups, the observed
correlations were not statistically significant at the one-tailed or = 0.05 level, but
inference probability is moderately high. It means that the probability of a relationship in
the hypothesized direction is much higher than the reverse probability of a relationship in
the opposite direction of H 4. Therefore, it is concluded that H 4 is confirmed in the cases
of Group A of Samsung, Group A of Nongsim, and Group B of Binggrae.

Four groups revealed that data are strongly consistent with H 4: Group B of
Samsung (r = -.34, p < .001 and [P = 1.00), Group A of Hyundai (r = -.24, p = .002 and
[P = 1.00), Group B of Nongsim (r = -.20, p = .006 and [P = 1.00) and Group A of
Binggrae (r = -.20, p = .007 and IP = .99). In these four groups, the observed correlations
were considerably significant at the one-tailed or = 0.05 level and the inference
probabilities are very high. Thus, it is concluded that H 4 is strongly confirmed in the
cases of Group B of Samsung, Group A of Hyundai, Group B of Nongsim and Group A

of Binggrae.

154

Figure 6.15. Inference Probability of H4 for Purchase Intention

 

Environmental Attitude &
Purchase Intention

 

 

 

 

E

u

.D

2

n. .50

o

o

c

2

:3 00

" ' A B A B A 8 A B

Sansung Sansung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae

Series1 .69 1.00 1.00 .53 .83 1.00 .99 .88

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.15 shows that except for one case of positive CEPI stimuli about
Hyundai (Group B of Hyundai), H 4 is confirmed in all of the cases (i.e., 7 of 8 cases).
For a conclusion concerning the Hyundai group B, further research is needed. Thus, it is
concluded that “environmental attitude toward pollution” is a moderator variable for the

variable “purchase intention.”

Corporate Familiarity as a Moderator

Hypothesis Five (H 5) posited that consumers with high familiarity with a
corporation (CFC) change their consumer attitude (AC), corporate credibility (CC), and
purchase intention (PI) less than one with low familiarity. In order words, H 5 states that
a negative relationship exists between CFC and three variables of AC, CC, and PI. Data

analysis results related to H 5 is presented in the following table 6.25.

155

Table 6.26

Familiarity with Corporation and Effect Size of Three Variables

 

 

 

Familiarity & Attitude Credibility Purchase Intention
r siLlP H5 r sig IP H5 r sig—IP H5
Samsung A:- -.15 .029 .97 C -.10 .108 .90 C -.22 .003 1.00 C
Samsung B:+ -.20 .008 .01 N -.12 .071 .07 N -.07 .210 .21 N
Hyundai A:+ .07 .193 .81 C .01 .439 .56 F .08 .177 .82 C
Hyundai 82- .00 .482 .48 F -.08 .153 .85 C -.01 .448 .55 F
Nongsim A:- -.19 .010 .99 C -.07 .199 .80 C -.17 .017 .99 C
Nongsim B:+ -.05 .279 .28 N .00 .494 .50 F -.01 .453 .45 F
Binggrae A:+ -.04 .301 .30 N .01 .431 .57 F -.05 .288 .29 N
Mme B:- -.28 0001.00 C -.22 .003 1.00 C -.24 .001 1.00 C

 

- A = Group A / B = Group B in the experiment.
- Positive CEPI was given to Samsung A, Hyundai B, Nongsim A and Binggrae B
- Negative CEPI was given to Samsung B, Hyundai A, Nongsim B and Binggrae A
- + / - = Direction of r required to confirm H 5 under CEPI stimuli given
- H 5: negative correlation (-)
- "+" is resulted from H 5 ( - ) and negative CEPI (-)
- "-" is resulted from H 5 ( - ) and positive CEPI (+)
- C = Confirmed / F = need Further research / N = Not confirmed
- r = Pearson correlation coefficient
- IP = Inference Probability
- sig = p—value

Familiarity fll Attitude toward Corporation

Three groups’ attitudes revealed the considerably opposite directional correlation
to H 5: Group B of Samsung (r = -.20, p = .008, and IP = .01), Group B of Nongsim (r = -
.05, p = 279 and [P = .28) and Group A of Binggrae (r = -.04, p = .301 and IP = .30). In
all three cases, the observed correlation should be positive in order to confirm H 5
because negative CEPI was given to them. However, the directions of observed

correlation are negative and the correlations are statistically significant at the one—tailed or

156

= 0.05 level. Reverse probabilities are considerably higher than the inference probability:
Group B of Samsung (RP = .99), Group B of Nongsim (RP = .72) and Group A of
Binggrae (RP = .70). It means that the probability of response occurring in the opposite
direction of H 5 is significantly larger than the probability of its occurring in the same
direction of H 5. Therefore, this study concludes that H 5 is strongly disconfirmed in the
case of Group B of Samsung, Group B of Nongsim and Group A of Binggrae.

Data revealed that no correlations would exist between familiarity and attitude in
one case: Group B of Hyundai (r = -.00, p = .482 and IP = .48). For Group B of Hyundai,
the observed correlations were zero and not statistically significant at one-tailed or = 0.05
level. [P (.48) is almost equivalent to RP (1- .48 = .52). Meanwhile, the inference
probability (IP = .48) falls between .333 and .667 ( 1/3 < [P < 2/3 ) and the sample size
is not large enough to draw a conclusion. Further research is needed to conclude that “no
correlation” is related to H 5 in the cases of Group B of Hyundai.

A group revealed that data are consistent with H 5: Group A of Hyundai (r = .07,
p = .193 and [P = .81). In this group, the observed correlation should be positive in order
to confirm H 5 because negative CEPI was given to them. The result correlation observed
was positive. The observed correlations were not significant at the one-tailed or = 0.05
level, but the inference probabilities was moderately higher than RP (1 - .81 = .19). Thus,
it is concluded that H 5 is confirmed in the cases of Group A of Hyundai.

Three cases showed that data are strongly consistent with H 5: Group A of
Samsung (r = -.15, p = .029 and IP = .97), Group A ofNongsim (r = -.19, p = .010 and [P
= .99) and Group B of Binggrae (r = -.28, p < .001 and IP = 1.00). In these three cases,

the observed correlations are significant and the inference probabilities are very high.

157

Therefore, this study concludes that H 5 is confirmed in the cases of Group A of

Samsung, Group A of Nongsim and Group B of Binggrae.

Figure 6.16. Inference Probability of H5 for Attitude toward Corporation

 

 

 

 

 

Familiarity and Attitude

1.00
'3
«II
.n
E 50
3 U
I:
2
2
E

'00 A 8 A B A 8

Samsung Samsung AHyundaI B HyundaI Nongsim Nongsim Binggrae Binggrae
Series1 .97 .01 .81 .48 .99 .28 .30 1.00

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.16 shows that familiarity with corporations is a severe contingency2|
variable for attitude toward corporation, related to the CEPI disclosure. H 5 is confirmed
in the four cases, but disconfirrned for the three cases. For the conclusion of one case,
additional research is needed. Therefore, this study concludes that H 5 related to attitude

toward corporation is disconfinned and “familiarity” is not a moderator variable to the

variable “attitude toward corporation.”

 

2' Contingency variable is defined as a variable, of which effect occurs in both positive and negative
direction. Contingency variable is a special case of the moderator variable.

158

Familiarity and Comorgte Credibility

Group B’s perception of credibility for Samsung revealed the strongly opposite
directional correlation to H 5: r = -.12, p = .071, and [P = .07. For H 5 to be confirmed,
the correlation between familiarity and credibility for Samsung should be positive
because Group B was exposed to negative CEPI of Samsung and the hypothesized
direction of H S is negative. However, the observed correlation is negative (r = -.12) and
is statistically significant (p = .071 at the one-tailed or = 0.05 level). The inference
probability is .07 and the reverse probability (RP)22 is .93 (i.e., RP = 1-.07 = .93). The
probability that the correlation between familiarity and corporate credibility exists in the
opposite direction of H 5 is very larger than the probability that the correlation exists in
the same direction of H 5. Therefore, it is concluded that H 5 is strongly disconfirmed in
the case of negative CEPI stimuli about Samsung (Group B of Samsung).

Data revealed that there “no correlation” or “almost no correlations” would exist
between familiarity and attitude in the three cases: Group A of Hyundai (r = -.01, p =
.439 and IP = .56), Group B of Nongsim (r = .00, p = .494 and [P = .50) and Group A of
Binggrae (r = .01, p = .431 and IP = .57). For Group B of Nongsim and Group A of
Binggrae, the observed correlations were not statistically significant at one-tailed 0I =
0.05 level and IP was almost equivalent to RP. It means that the probability of the
correlation existing in the hypothesized direction is almost equal to the reverse
probability of its correlation existing in the opposite direction of H 5. However, the

inference probabilities fall between .333 and .667 ( 1/3 < [P < 2/3 ); thus, the sample

 

22 Reverse probability (RP) is the probability that effect occurs in the opposite direction of hypothesis.

159

size is not large enough to make conclusions. To draw a conclusion, we need further
research to decide that “no correlation” exists in relation to H 5 in the cases of Group A
of Hyundai, Group B of Nongsim and Group A of Binggrae.

The three groups revealed that data are moderately consistent with H 5: Group A
of Samsung (r = -.10, p = .108 and [P = .90), Group B onyundai (r = -.08, p = .153 and
[P = .85) and Group A of Nongsim (r = -.07, p = .199 and [P = .80). In these three groups,
the observed correlations were not significant at the one-tailed or = 0.05 level but the
inference probabilities are substantially higher than RP. Thus, this study concludes that H
5 is confirmed in the cases of Group A of Samsung, Group B of Hyundai and Group A of
Nongsim.

One case showed that the data are strongly consistent with H 5: Group B of
Binggrae (r = -.22, p = .003 and [P = 1.00. In this case, the correlation is significant and
the probability that the correlation between familiarity and corporate credibility exists in
the direction opposite to H 5 is .00 (RP =1- 1.00). Therefore, this study concludes that H

5 is strongly confirmed in the cases of Group B of Binggrae.

160

Figure 6.17. Inference Probability of H5 for Corporate Credibility

 

Familiarity and Corporate Credibility

 

 

 

 

1.
:E' 00
3
III
n
i 60 \
3 .
r:
2
.2
E
.00
A B A B A B A 8
Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
Series1 .90 .07 .56 .85 .80 .50 .57 1.00

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.17 shows that familiarity with a corporation is a severe contingency23
variable for corporate credibility, related to the CEPI disclosure. H 5 is confirmed in the
four cases, but disconfirmed in the case of Group B of Samsung. Further research is
needed to conclude that no correlation exists in the three cases. Therefore, this study
concludes that H 5 related to corporate credibility is disconfirmed and “familiarity” is not

a moderator variable to the variable “corporate credibility.”

Familigrity and Purchase Intention

Two groups’ perception of credibility revealed a considerably opposite directional
correlation to H 5: Group B of Samsung (r = -.O7, p = .210, and [P = .21) and Group A of
Binggrae (r = -.05, p = .288 and [P = .29). In these two cases, the observed correlations

are not statistically significant at the one-tailed or = 0.05 level but reverse probabilities are

 

23 Contingency variable is defined as a variable, of which effect occurs in both positive and negative
direction. Contingency variable is a special case of the moderator variable.

161

higher than inference probability: Group B of Samsung (RP = .79) and Group A of
Binggrae (RP = .71). It means that the probability that the correlation between corporate
familiarity and purchase intention exists in the opposite direction of H 5 is significantly
stronger than the probability that the correlation exists in the same direction of H 5.
Therefore, this study concludes that H 5 is strongly disconfirmed in the case of Group B
of Samsung and Group A of Binggrae.

Data revealed that almost no correlations exists between familiarity and attitude in
the two cases: Group B onyundai (r = -.01, p = .448 and [P = .55) and Group B of
Nongsim (r = -.01, p = .453 and [P = .45). In these two cases, the observed correlations
are almost zero and not statistically significant at one-tailed or = 0.05 level. [P is almost
equivalent to RP: Group B of Hyundai (RP = .45) and Group B of Nongsim (RP = .55).
However, the inference probabilities fall between .333 and .667 ( 1/3 < [P < 2/3 ) with
the result that the sample size is not large enough to draw a conclusion. As a conclusion,
further research is needed to decide “no correlation” related to H 5 in the cases of Group
B of Hyundai and Group B of Nongsim.

One group revealed that data are consistent with H 5: Group A of Hyundai (r = -
.08, p = .177 and [P = .82). In this group, the observed correlations were not significant at
the one-tailed or = 0.05 level but the inference probabilities are moderately higher than
RP (1 - .81 = .19). Thus, it is concluded that H 5 is confirmed in the cases of Group A of
Hyundai.

Three cases showed that data are strongly consistent with H 5: Group A of
Samsung (r = -.22, p = .003 and [P = 1.00), Group A of Nongsim (r = -.17, p = .017 and

[P = .99) and Group B of Binggrae (r = -.24, p = .001 and IP = 1.00). In these three cases,

162

the observed correlations are significant and the inference probabilities are very high.
Therefore, it is concluded that H 5 is confirmed in the cases of Group A of Samsung,

Group A of Nongsim and Group B of Binggrae.

Figure 6.18. Inference Probability of H5 for Purchase Intention

 

 

 

 

 

Familiarity and Purchase Intention
:- 1.00
3
III
.9
2 50
3 i V
c
2
8
- .00
A B A 8 A 8 A 8
Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
[Siries1 1.00 .21 .82 .55 .99 .45 .29 1.00

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.18 shows that familiarity with corporations is a severe contingency
variable for purchase intention, related to the CEPI disclosure. H 5 is confirmed in the
four cases, but for the two cases H 5 is disconfirmed. For the conclusion in the two cases,
further research is needed. Thus, this study concludes that H 5 related to purchase
intention is disconfirmed and “familiarity” is not a moderator variable to the variable

“purchase intention.”
Information Credibility as a Moderator

Hypothesis Six (H 6) posited that the greater the credibility of CEPI (CCI), the

greater the effect of the CEPI rating on the three dependent variables: consumer attitude

163

(AC), corporate credibility (CC), and purchase intention (PI). That is, a positive
relationship exists between CCI and the three dependent variables: AC, CC, and PI. Data

analysis results related to H 6 are presented in the following Table 6.26.

Table 6.27

Information Credibility and Effect Size of Three Variables

 

 

 

Information Attitude Credibility Purchase Intention
Credibility& r sig IP H6 r sig lP H6 r sig IP H6
Samsung A:+ .21 0051.00 C .28 .000 1.00 C .25 .001 1.00 C
Samsung 82- -.21 0041.00 C -.11 .094 .91 C -.11 .093 .91 C
Hyundai A:- -.24 .001 1.00 C -.31 .000 1.00 C -.14 .041 .96 C
Hyundai B:+ .11 .088 .92 C .16 .027 .98 C .14 .046 .96 C
Nongsim A:+ .15 .037 .97 C .18 .014 .99 C .07 .184 .82 C
Nongsim B:- -.09 .148 .85 C -.06 .239 .76 C -.21 .005 1.00 C
Binggrae A:.- -.26 .001 1.00 C -.27 .000 1.00 C -.05 .284 .72 C
Binggrae B:+ .27 0001.00 C .24 .001 1.00 C .12 .066 .94 C

 

- A = Group A / B = Group B in the experiment.
- Positive CEPI was given to Samsung A, Hyundai B, Nongsim A and Binggrae B
- Negative CEPI was given to Samsung B, Hyundai A, Nongsim B and Binggrae A
- + / - = Direction of r required to confirm H 6 under CEPI stimuli given
- H 6: positive correlation (+)
- "-" is resulted from H 6 ( + ) and negative CEPI (—)
- "+" is resulted from H 6 ( + ) and positive CEPI (+)
- C = Confirmed / F = need Further research / N = Not confirmed
- r = Pearson correlation coefficient
- IP = Inference Probability
- sig = p-value

Information Credibility and Attitude toward Corporation
Related to attitude change, Group B of Nongsim revealed that data are consistent
with H 6: r = -.09, p = .148, and [P = .85. In this case, the observed correlations were not

statistically significant at the one-tailed a = 0.05 level, but inference probability is larger

164

than RP (1 - .85 = .15). It means that the probability that the correlation between CEPI
credibility and attitude toward corporation is positive (i.e., the hypothesized direction of
H 6) is considerably larger than the reverse probability that the correlation is negative.
Therefore, it is concluded that H 6 is confirmed in the cases of Group B of Nongsim.
Except for Group B of Nongsim, seven of eight groups revealed that data are
strongly consistent of H 6: Group A of Samsung (r = .21, p = .005 and [P = 1.00), Group
B of Samsung (r = -.21, p = .004 and IP = 1.00), Group A of Hyundai (r = -.24, p = .001
and [P = 1.00), Group B of Hyundai (r = .11, p = .088 and [P = .92), Group A of
Nongsim (r = .15, p = .037 and [P = .97), Group A of Binggrae (r = -.26, p = .001 and [P
= 1.00), and Group B of Binggrae (r = .27, p < .001 and IP = 1.00). In these seven groups,
the observed correlations were highly significant at the one-tailed 01 = 0.05 level and the
inference probabilities are very high. Thus, this study concludes that H 6 is strongly
confirmed in the cases of Group A of Samsung, Group B of Samsung, Group A of
Hyundai, Group B of Hyundai, Group A of Nongsim, Group A of Binggrae and Group B

of Binggrae.

165

Figure 6.19. Inference Probability of H 6 for Attitude toward Corporation

 

Information Credibility 8. Attitude

 

 

 

 

g 1-00 C t W ¢
23
a
.0
E
w .50
o
c
2
a
E .00
A B A B A B A 8
Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
Eerie“ 1.00 1.00 1.00 .92 .97 .85 1.00 1.00

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.19 shows that data are strongly consistent with H 6 in 7 of 8 cases and
consistent with H 6 in one case. Therefore, this study concludes that H 6 about attitude
toward corporation is confirmed and “information credibility” is a moderator variable to

the variable “attitude toward corporation.”

Information Credibility and Corporate Credibility

Group B of Nongsim related to credibility change revealed that the data is
consistent with H 6: r = -.06, p = .239, and [P = .76. In this case, the observed
correlations were not statistically significant at the one-tailed CI = 0.05 level, but the
inference probability is larger than the reverse probability (RP: l - .76 = .24). It means
that the probability of positive correlation between CEPI credibility and corporate

credibility is considerably larger than the probability that the correlation is negative (i.e.,

166

the opposite direction of H 6). Therefore, it is concluded that H 6 is confirmed in the
cases of Group B of Nongsim.

Except for Group B of Nongsim, the seven of eight groups revealed that data are
strongly consistent with H 6: Group A of Samsung (r = .28, p < .001 and IP = 1.00),
Group B of Samsung (r = -.11, p = .094 and IP = .91), Group A onyundai (r = -.31, p <
.001 and [P = 1.00), Group B of Hyundai (r = .16, p = .027 and [P = .98), Group A of
Nongsim (r = .18, p = .014 and [P = .99), Group A of Binggrae (r = -.27, p < .001 and [P
= 1.00), and Group B of Binggrae (r = .24, p = .001 and [P = 1.00). In these seven groups,
the observed correlations were considerably significant at the one-tailed or = 0.05 level
and the inference probabilities are strongly high. Thus, this study concludes that H 6 is
strongly confirmed in the cases of Group A of Samsung, Group B of Samsung, Group A
of Hyundai, Group B of Hyundai, Group A of Nongsim, Group A of Binggrae and Group

B of Binggrae.

Figure 6.20. Inference Probability of H6 for Corporate Credibility

 

 

 

 

 

 

Information Credibility &
Corporate Credbility
E 1.00 W 4
3
«I
a
2
n.
a .50
u
I:
2
.9
I:
" '00 A B A B A B A 8
Samsung Samsung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
[Series1 1.00 .91 1.00 .98 .99 .76 1.00 1.00

 

 

 

 

 

 

 

 

 

 

 

 

 

167

Figure 6.20 shows that data are strongly consistent with H 6 in 7 of 8 cases and
consistent with H 6 in one case. Therefore, this study concludes that H 6 related to
corporate credibility is confirmed and “information credibility” is a moderator variable to

the variable “corporate credibility.”

Information Credibility and Purchase Intention

Related to purchase intention, data about two groups are consistent with H 6:
Group A of Nongsim (r = .07, p = .184, and IP = .82) and Group A of Binggrae (r = -.05,
p = .284 and IP = .72). In these two cases, the observed correlations were not statistically
significant at the one-tailed 0: = 0.05 level, but inference probability is larger than RP:
Group A of Nongsim (RP = 1 - .82 = .18) and Group A of Binggrae (RP = 1 - .72 = .28).
It means that the probabilities that the correlation between CEPI credibility and purchase
intention exists in the hypothesized direction are considerably larger than the reverse
probabilities that the correlation exists in the opposite direction of H 6. Therefore, it is
concluded that H 6 is confirmed in the cases of Group A of Nongsim and Group A of
Binggrae.

Except for the Group A of Nongsim and Group A of Binggrae, six of eight groups
revealed that data are strongly consistent with H 6: Group A of Samsung (r = .25, p =
.001 and IP = 1.00), Group B of Samsung (r = -.11, p = .093 and IP = .91), Group A of
Hyundai (r = -.14, p = .041 and IP = .96), Group B onyundai (r = .14, p = .046 and [P =

.96), Group B of Nongsim (r = -.21, p = .005 and IP = 1.00) and Group B of Binggrae (r

168

= .12, p = .066 and IP = .94). In these six groups, the observed correlations were strongly
significant at the one-tailed a = 0.05 level and the inference probabilities are very high.
Thus, this study concludes that H 6 is strongly confirmed in the cases of Group A of
Samsung, Group B of Samsung, Group A of Hyundai, Group B of Hyundai, Group B of

Nongsim and Group B of Binggrae.

Figure 6.21. Inference Probability of H 6 for Purchase Intention

 

Information Credibility &
Purchase Intention

 

 

 

 

3. 1.00 A
g Vv—VV
a
.n
2
n- .50
o
u
c
o
3
I.-
5 .00
A B A B A B A B
Sarrsung Sansung Hyundai Hyundai Nongsim Nongsim Binggrae Binggrae
[Series1 1.00 .91 .96 .96 .82 1.00 .72 .94

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.21 shows that data are strongly consistent with H 6 in 6 of 8 cases and
also consist of H 6 in 2 cases. Therefore, the researcher concludes that H 6 related to
purchase intention is disconfirmed and “information credibility” is a moderator variable

to the variable “purchase intention.”

169

CHAPTER VII

DISCUSSION AND INTERPRETATION

Chapter VI reported statistical data analysis and results. Results include both
effects of the corporate environmental performance information (CEPI) about the four
Korean corporations (i.e., Samsung, Hyundai, Nongsim and Binggrae) and effects of
three moderators (i.e., familiarity with corporations, environmental attitude and CEPI
credibility) on three dependent variables (i.e., attitude toward the four corporation,
credibility of the four corporation and purchase intention toward the products of the four
corporations).

This study’s experiment tests the six hypotheses regarding PID. Among them,
five hypotheses (H 1, 2, 3, 4, and 6) were confirmed, and one hypothesis (H 5) was
disconfirmed. Participants in this experimental test consisted of three hundred and six
Korean university students. The amount of results reported in the last chapter is vast, and
its contents are complex. In order to obtain more clear understandings about the test
results and to conduct rich discussion and appropriate interpretations for the test results, it
is needed to organize the results of data analysis in the way of comparing all results from
the four corporations. Thus, this chapter begins with summary and comparison of the

results of all hypotheses tests.

Summary of Test of Hypothesis One, Two and Three
As shown in Table 7.1, all three major hypotheses are confirmed. The three

hypotheses are as follows:

170

H 1: CEPI disclosures change consumer attitude toward corporations positively

for non-polluting corporations and negatively for polluting corporations.

H 2: CEPI disclosures change the consumers’ perception of corporate credibility
positively for non-polluting corporations and negatively for polluting

corporations.

H 3: CEPI disclosures increase consumer purchase intention for the products of
non-polluting corporations and decrease consumer purchase intention for the

products of polluting corporations.

Table 7.1

Summary of Test of Hypothesis One, Two and Three

 

 

Samsung Hyundai Nongsim Binggrae
H l: Attitude N.CEPI S. Confirmed 8. Confirmed 8. Confirmed S. Confirmed
P. CEPI 8. Confirmed 8. Confirmed Confirmed S. Confirmed
H2: Credibility N. CEPI 8. Confirmed S. Confirmed S. Confirmed S. Confirmed
P. CEPI 8. Confirmed 8. Confirmed S. Confirmed S. Confirmed
H3: Purchase N. CEPI 8. Confirmed 8. Confirmed 8. Confirmed 8. Confirmed
Intention P. CEPI 8. Confirmed S. Confirmed Confirmed S. Confirmed

 

- S. Confirmed = Strongly Confirmed

- N. CEPI = Negative CEPI
- P. CEPI = Positive CEPI

Table 7.1 shows that all of the Hypothesis One, Two and Three are confirmed. In

the case of attitude change of Nongsim by positive CEPI, stimuli correlation is very weak

171

(r = .06) and credibility change is barely significant (p = .100) at the level of a = .05 and
one-tailed significance test. However, inference probability is larger (IP = .76) than
reverse probability (R1 = .24). Thus, attitude change of Nongsim is interpreted as
probably in the direction hypothesized. In the case of purchase intention toward Nongsim
instant noodle by positive CEPI, stimuli correlation is very small (r = .06). However,
purchase intention change is significant (p = .066) at the level of or = .05 and one-tailed
significance test and inference probability is larger (IP = .79) than reverse probability (R1
= .21). Thus, the purchase intention change toward Nongsim instant noodle is interpreted
as probably in the direction hypothesized. Except these two cases, all cases retained
strong stimuli correlations, highly significant change of gain scores and high inference
probability.

In the following, results of data analysis about changes of attitude, credibility, and

purchase intention caused by CEPI disclosure are summarized in the charts.

CEPI Disclosure and Change of Attitude toward Corporflm: H 1
The Hypothesis One (H 1) is that CEPI disclosures change consumer attitude
toward corporations (AC) positively for non-polluting corporations and negatively for

polluting corporation.

172

Figure 7.1. Attitude Change and Effect Direction

 

D of Attitude

 

3.00
2.00
1.00
.00 “
-1.00 ‘
-2.00
-3.00

 

 

 

 

Raw Gain Score

 

 

1 2 3 4 5 6 7 8
ISeries1 .24 -1.03 .59 -.55 .11 -1.50 .61 -.73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- D = Mean of difference of raw score between time 1 (Pre-test) and time 2 (Post-test).
- Raw score scale is —7 to 7.
- 1 = the case of Positive CEPI of Samsung
2 = the case of Negative CEPI of Samsung
3 = the case of Positive CEPI of Hyundai
4 = the case of Negative CEPI of Hyundai
5 = the case of Positive CEPI of Nongsim
6 = the case of Negative CEPI of Nongsim
7 = the case of Positive CEPI of Binggrae
8 = the case of Negative CEPI of Binggrae
- These notes also apply to the next Figure 7.2, 7.3, 7.4, 7.5, 7.6, 7.7, 7.8 and 7.9.

Figure 7.1 summarizes the result that the subjects changed their level of attitude
of the four corporations in the negative direction when they were exposed to negative
CEPI of the four corporations, and that the subjects changed their attitude toward the four

corporations in the positive direction when they were exposed to positive CEPI of the

four corporations.

173

Figure 7.2

Stimuli Correlation of Attitude Change as Effect Size and Pattern

 

Effect Size (Attitude Change)

1.00
.80
.60
.40
.20
.00

-.20

-.40

—.60

-.80

-1.00

Stimuli r

1 2 3 4 5 6 7 8
ISeries1 .12 -.42 .30 -.23 .06 -.59 .33 -.34

 

 

 

 

Figure 7.2 shows that the extent of attitude changes in response to CEPI (stimuli
effect size) varies depending on corporations and on the positive or negative of CEPI.

No pattern by corporation was found in attitude changes. However, subjects
responded more sensitively to negative information than to positive information about
Samsung and Nongsim. For Hyundai, subjects more sensitively responded to positive

information than negative information.

174

Figure 7.3. Probability of Effect in the Direction Predicted (Inference Probability)

 

Inference Probability

 

Effect Direction (Attitude Change)

 

 

 

 

 

.50
.00
1 2 3 4 5 6 7 8
+Series1 .94 1.00 1.00 1.00 .76 1.00 1.00 1.00

 

 

 

 

 

 

 

 

 

 

 

 

Figure7.3 reveals two findings:

1.

The probabilities that the population (i.e., the Korean university students) of
subjects changes their attitude toward the four corporations in the negative
direction are very high (100 %) for all cases when they are exposed to
negative CEPI.

The probabilities that the population changes their attitude of the four
corporations in the positive direction are very high (100 %) for three cases,
high (76 %) for one case and moderate (76 %) for one case when they are

exposed to positive CEPI.

175

No pattern by negative and positive CEPI was found in attitude changes.
However, subjects’ showed a more discemable response to negative information than to

positive information.

CEPI Disclosure and Chage of Corporate Credibility: H 2
The Hypothesis Two (H 2) is that CEPI disclosures change consumers’ perception
of corporate credibility (CC) positively for non-polluting corporations and negatively for

polluting corporations.

Figure 7.4. Corporate Credibility Change and Effect Direction

 

 

 

 

 

 

 

 

 

 

 

D of Credibility
3.00
2 2.00
3 100
a, .
= - J
-- .oo~
8 I I I .—
5 -1.00
-2.00
-3.00
1 2 3 4 5 6 7 8
.Series1 .38 -.81 .62 -.57 .28 -1.21 .55 -.67

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 7.4 summarizes the result that subjects changed their perception of
credibility for four corporations to the negative direction when they were exposed to

negative CEPI of the four corporations, and subjects changed their perception of

176

credibility for four corporations in the positive direction when they were exposed to

positive CEPI of the four corporations.

Figure 7.5. Stimuli Correlation of Credibility Change as Effect Size and Pattern

 

Effect Size (Credibility Change)

1.00
.80
.60
.40
.20
.00

-.20
-.40
-.60
-.80
-1.00

Stimuli r

 

fl Series1

 

 

 

Figure 7.2 shows that the extent of credibility changes in response to CEPI
(stimuli effect size) varies depending on the corporations and on the positive or negative
of CEPI.

No pattern by corporation was found in credibility changes. However, subjects
responded more sensitively to negative information than to positive information about
Samsung and Nongsim. For Hyundai, subjects more sensitively responded to positive
information than to negative information. This tendency of sensitivity by corporation
related to credibility change is consistent with the tendency of sensitivity related to

attitude change.

177

Figure7.6. Probability of Effect in the Direction Predicted (Inference Probability)

 

Inference Probability

 

Effect Direction (Credibility Change)

 

 

 

 

1.oo——o—o-—o—+<F—o—+—o—
.50
.oo
1 2 3 4 5 6 7 8
+Series1 1.00 1.00 1.00 1.00 .97 1.00 1.00 1.00

 

 

 

 

 

 

 

 

 

 

 

 

Figure 7.3 reveals two findings:

1.

The probabilities that the population of subjects changes their attitude toward
the four corporations in the negative direction are very high (100 %) when
they are exposed to negative CEPI.

The probabilities that population changes their attitude toward the four
corporations in the positive direction are very high (100 % or 97%) when

they are exposed to positive CEPI.

No pattern or tendency by negative and positive CEPI was found in credibility changes.

CEPI Disclosure and Change of Purchase Intention toward Products: H 3.

178

The Hypothesis Three (H 3) is that CEPI disclosures increase consumer purchase
intention for the products (PI) of non-polluting corporations and decrease consumer

purchase intention for the products of polluting corporations.

Figure 7.7. Purchase Intention Change and Effect Direction

 

D of Purchase Intention

 

 

 

 

       

 

 

 

 

 

 

 

 

3.00
g 2.00
8
a, 1.00
-§ .001
o
g -1.00
G
0‘ -200
-3.00
1 2 3 4 5 6 7 8
ISeriesi .43 -.89 1.06 -.25 .15 -1.66 .53 -.7o

 

 

 

 

 

 

 

 

 

 

 

 

Figure 7.7 summarizes the result that subjects changed their purchase intention
toward the products of the four corporations in the negative direction when they were
exposed to negative CEPI about the four corporations and their purchase intention toward
the products of the four corporations in the positive direction when they were exposed to

positive CEPI about the four corporations.

179

Figure 7.8

Stimuli Correlation of Purchase Intention Change as Effect Size and Pattern

 

Effect Size
(Puchase Intention Change)

 

1.00
.80
.60

.40 I
.20 1

.00 j

-.20 I

-.40
-.60
-.80
-1.00

 

 

 

 

 

 

   

 

 

 

 

 

Treatment r

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lSeries1 .18 -.28 .39 -.1O .06 -.53 .21 -.26

 

 

 

 

Figure 7.8 shows that the extent of purchase intention changes in response to
CEPI (stimuli effect size) varies depending on the corporations and on the positive or
negative of CEPI.

No pattern by corporation was found in attitude changes. However, subjects
responded more sensitively to negative information than to positive information about
Samsung and Nongsim. For Hyundai, subjects more sensitively responded to positive
information than to negative information. This subject’s sensitivity by corporation related
to purchase intention is consistent with the subject’s sensitivity related to attitude and

credibility change.

180

Figure 7.9

Probability of Effect in the Direction Predicted (Inference Probability)

 

Effect Direction
(Purchase Intention Change)

1.00 W

.50

 

 

 

.00

Inference Probability

 

 

 

 

 

 

 

 

 

 

 

+Seriesl .99 1.00 1.00 .89 .79 1.00 1.00 1.00

 

 

 

 

Figure 7.9 reveals two findings:

1. The probabilities that the population of subjects changes their purchase
intention toward the products of the four corporations in the negative direction
are very high (100 %) for three cases and high (89 %) for one case when they
are exposed to negative CEPI.

2. The probabilities that population changes their purchase intention toward the
products of the four corporations in the positive direction are mostly very high
(100% for two cases and 99% for one case) and moderately high (79%) for

one case when they are exposed to positive CEPI.

No pattern by negative and positive CEPI was found in attitude changes.

However, subjects’ more discemable response to negative information than to positive

181

information is observed. This subject’s sensitivity related to purchase intention and type

of information is consistent with the subject’s sensitivity related to attitude change.

Summary of Test of Hypothesis Four, Five and Six1

Hypotheses Four, Five, and Six are as follows:

H 4: Consumers who have strong environmental attitudes about pollution (BAP)
will change the three dependent variables more than consumers who had
weak EAPZ. In other words, a positive relationship exists between EAP and
the three dependent variables: attitude toward corporation (AC), corporate

credibility (CC), and purchase intention (PI).

H 5: Consumers that are highly familiar with a corporation (CFC) change
consumer attitude, corporate credibility, and purchase intention less than one
with low familiarity. In other words, a negative relationship exists between
CFC and the three dependent variables: attitude toward corporation (AC),

corporate credibility (CC), and purchase intention (PI).

 

' Correlations among three moderator variables related H 1, 2, and 3 are reported in Appendix C.

2 In this research, environmental attitude specifies the environmental attitude about pollution because the
research is related to industrial pollution. Thus, high EAP is defined as the status of a strong concern,
awareness, and belief about pollution. Low EAP is defined as the status of a weak awareness, concern, and
belief about pollution.

182

H 6: The greater the credibility of CEPI, the greater the effect of the CEPI rating
on the dependent variables. That is, a positive relationship exists between

CCI and the three dependent variables: AC, CC, and PI.

In brief, H 4 and H 6 are confirmed, while H 5 is disconfirmed.

Environmental Attitude and Effect of CEPI: H 4.

The Hypothesis Four (H 4) is that customers who have strong environmental
attitudes toward pollution (EAP) will change the three dependent variables more than
customers who have weak EAP. In other words, a positive relationship exists between
EAP and the three dependent variables: attitude toward corporation (AC), corporate
credibility (CC), and purchase intention (PI). Table 7.2 and Figure 7.10 summarize the

results of H 4.

Table 7.2. Summary of Test of Hypothesis Four

 

Environmental Group: Attitude Credibility Purchase Intention

 

 

Attitude & CEPI H4 H4 H4
Samsung A:+ C F C
Samsung B: - C C C
Hyundai A: - C C C
Hyundai B:+ C C F
Nongsim A:+ C C C
Nongsim B: - C C C
Binggrae A: - C C C
Binggrae B:+ F C C

 

- A = Group A / B = Group B in the experiment.

183

- " - " = Negative CEPI

- " + "= Positive CEPI

- C = Confirmed

- F = Further Research is required

As Table 7.2 summarized, H 4 is confirmed in all cases except the following three
cases: 1) Binggrae positive CEPI and credibility change, 2) Samsung positive CEPI and
attitude change, and 3) Hyundai positive CEPI and purchase intention change. To draw a
conclusion about these three cases, further researches are required because inference
probabilities of these three cases fall between .333 and .667 (1/3 < IP < 2/3) and the

sample size is not large enough to draw a conclusion.

Figure7. l 0. Summary of Inference Probability related to H 4

 

Environmental Attitude and Three Variables

1.00

 

 

 

 

 

 

 

Inference Probability
01
o

 

 

 

.00 .
Q Q r}, o . <0 - 6‘ o
verb Q? i” ‘2’ e 9% \ 92%

 

 

 

- Atti = Attitude toward Corporation
- Cred = Credibility of Corporation
- P. Inten = Purchase Intention toward Product

184

— These notes also apply to Figure 7.11 and 7.12.

Figure 7.10 shows that the probability that the correlation between Environmental
Attitude toward Pollution (EAP) and l) Attitude toward Corporation (AC), 2) Credibility
of Corporation (CC) and 3) Purchase Intention toward Product (PI) is positive, are mostly

very high. Thus, it is found that EAP is a moderator variable3 for AC, CC and PI.

Familiarity with Corporation and Effect of CEPI: H 5

The Hypothesis Five (H 5) is that consumers that are highly familiar with a
corporation (CFC) will change consumer attitude, corporate credibility, and purchase
intention less than one with low familiarity. In other words, a negative relationship exists
between CFC and the three dependent variables: attitude toward corporation (AC),
corporate credibility (CC), and purchase intention (PI). Table 7.3 and Figure 7.11

summarize the results of H 5.

 

3 Moderator is defined as a variable that intervenes between independent and dependent variable and
influences the intensity of effect in the dependent variable, but it is not an independent variable and also not
a mediator variable. Masking variables are unknown variables that affect the defendant variable. Moderator
variable is a special case of the masking variables. Meanwhile, Mediator variable is a middle variable
located between the first and the final variable in the causal diagram so that mediator is both independent
and dependent variable. For more details, see Chapter VII, p. 221.

185

Table 7.3. Summary of Test of Hypothesis Five

 

Familiarity Group: Attitude Credibility Purchase Intention

 

 

CEPI H 5 H 5 H 5
Samsung A:+ C C C
Samsung B: - N N N
Hyundai A: - C F C
Hyundai B:+ F C F
Nongsim A:+ C C C
Nongsim B: - N F F
Binggrae A: - N F N
Binggrae B:+ C C C

 

- A = Group A / B = Group B in the experiment.
- " - " = Negative CEPI

- " + "= Positive CEPI

- C = Confirmed

- F = Further Research is required

- N = Not Confirmed

In summary, H 5 is disconfirmed. As Table 7.3 summarized, H 5 is confirmed in
some cases. Some cases require further research to decide confirmation of H 5 test. But H
5 is not confirmed in other cases in which effects occurred in the opposite direction

hypothesized.

186

Figure 7.11. Summary of Inference Probability related to H 5

 

Familiarity and Three Variables

 

 

 

 

 

 

 

.1?

IE

.3 .

g Cred

‘: ' ' i ' 'Atti

g " '0'- P. Inten
2

.2

E

 

 

 

 

 

 

 

Figure 7.11 shows that probability that the correlation between Familiarity with
Corporations (CFC) and 1) Attitude toward Corporation (AC), 2) Credibility of
Corporation (CC) and 3) Purchase Intention toward Product (PI) is negative, is very high
in some cases. However, the probability that the correlation between environmental
attitude toward pollution and l) Attitude toward Corporation, 2) Credibility of

Corporation and 3) Purchase Intention toward Product is positive (i.e., reverse direction

187

of H 5), are also very high. Thus, it is found that CFC is severe contingency variable“, not

a moderator variable for AC, CC and PI.

CEPI Credibility and Effect of CEPI: H 6.

The Hypothesis Six (H 6) is that: The greater the credibility of CEPI, the greater the
effect of the CEPI on the dependent variables. That is, a positive relationship exists
between CCI and the three dependent variables: AC, CC, and PI. As table 7.4

summarizes, H 6 is clearly confirmed in all cases.

Table 7.4. Summary of Test of Hypothesis Six

 

Information Group: Attitude Credibility Purchase Intention

 

 

Credibility CEPI H 6 H 6 H 6
Samsung A:+ C C C
Samsung B: - C C C
Hyundai A: - C C C
Hyundai B:+ C C C
Nongsim A:+ C C C
Nongsim B: - C C C
Binggrae A: - C C C
Binggrae B:+ C C C

 

- A = Group A / B = Group B in the experiment.
- " - " = Negative CEPI

- " + "= Positive CEPI

- C = Confirmed

 

‘ Contingency variable is defined as a variable, of which effect occurs in both positive and negative
direction. Contingency variable is a special case of moderator variable.

188

Figure 7.12. Summary of Inference Probability related to H 6

 

Information Credibility and Three Variables

 

 

 

 

 

 

 

 

 

 

 

E
'9;
g —I— Cred
E .50 ---A---Atti
£3 — -O- — P. Inten
O
b
.2
E
.00 . T
Q Q q} o '6‘ «9 0 0
0 0 8 e e (o <5
9 Q Q
6° «sf 40 at 0° 0° 0°» .09
{5” $90 9" ‘b if <29; \ e”

 

 

 

Figure 7.12 shows that the probability that the correlation between Credibility of
CEPI (CCI) and 1) Attitude toward Corporation (AC), 2) Credibility of Corporation (CC)
and 3) Purchase Intention toward Product (PI) is positive (H 6), are mostly very high.

Thus, it is found that CCI is a moderator variable for AC, CC and PI.

CEPI Disclosure and Consumer Attitude Change (H 1)

This study finds that CEPI damages or improves consumers’ attitude toward
corporations (i.e., favor to corporation) by the proof of the Hypothesis One (H 1) that
CEPI disclosures change consumer attitude toward corporations (AC) positively for non-
polluting corporations and negatively for polluting corporations. In other words, this
study proves two facts: 1) information regarding corporate environmental performance

influences consumers’ attitude toward corporation (AC), and 2) CEPI disclosure is one

189

cause of diverse causes for consumer attitude change. This finding about consumer
attitude change by CEPI disclosure can be interpreted that CEPI influences consumer
purchase behavior through attitude change.

Consumer attitude toward corporation is a consumer’s psychological tendency
that is expressed by evaluation of corporations (e. g., Samsung Electronics, Hyundai
Electronics, Nongsim, and Binngrae) with favor or disfavor. Therefore, the causal
relationship between CEPI disclosure and attitude change means that positive CEPI
generates consumers’ psychological tendency to favor non-polluting corporations and
negative CEPI generates consumers’ psychological tendency to disfavor polluting.

Attitude change by CEPI seems to be preceded by change of corporate image or
reputation. Most advocators for Public Information Disclosure as a Pollution Control
Tool (PID) believe that CEPI influences corporate image or reputation.5 Environmental
information about polluting corporations damages their image and reputation while
environmental information of non-polluting corporations improves their image and
reputation. Wheeler and Afsah, (1996) stated, “Public knowledge of environmental
performance has important implications for reputaionally sensitive companies” (p. 2).
Thus, it is inferred that CEPI generates bad or good image or reputation about
corporations, and these bad or good corporate reputations or images generate consumers’
psychological tendency to favor or disfavor corporations. However, this inference about
causal order among three variables (i.e., CEPI disclosure -) image or reputation change

-) attitude change) scientifically is not proven .

 

5 About relationship between CEPI disclosure and reputational incentives to polluters, refer to Chapter III,
p. 62.

190

This research provides evidence supporting the assertion that environmental
information especially influence attitude. It is generally accepted that information
influences attitude, but there could be controversy to the assertion that environmental
information influences attitude. A great deal of anecdotal and scientific evidence suggests
that political information influences constituents’ attitude. Information about a certain
presidential candidate (i.e., sexual scandal) produced negative voter attitude toward that
candidate. Advertising information also influences attitudes of audiences. Social
scientists often divide attitude into the three categories: cognition, affection, and
behavior. The cognitive category encompasses thoughts that people have about the
attitude object (e. g., Samsung Electronics). The affective category consists of feelings or
emotions that people have in relation to the attitude object. The behavioral category
contains people’s actions with respect to the attitude object (Eagly & Chaiken, 1993). Yi
(1990) conducted an experiment to investigate cognitive and affective priming of the
context for print advertisements. The cognitive context primed certain attributes of a
product and the affective context triggered emotional reactions among the audience. He
found that the both cognitive and affective context influenced consumers’ evaluations of
the advertised brand, but in a different process. That is, cognitive priming effects
operated mainly through attitude toward the brand, whereas affective priming effects
worked primarily via attitude toward the advertisement. That is, Yi (1990) found that
advertise information has influence on forming consumers’ attitudes, even though the
process of attitude change by advertised information is diverse (e. g., cognitive process,

affective process, or behavioral process). Yi’s finding is consistent of the finding of this

191

research, even thought the findings are about different information: advertising
information and environmental information.

Meanwhile, subjects’ attitude changes are processed through different
psychological routes. Cognitive Response Theory claims that cognitive responses
mediate the effect of persuasive messages on attitude change. Messages that evoke
predominantly favorable recipient-generated thoughts (e. g., valence or subject
ageement) should be persuasive, whereas those that evoke mostly unfavorable thoughts
(e. g., counterargument) should be unpersuasive (Eagle & Chaiken, 1993).

Petty and Cacioppo (1981 a,1986 a, and 1986 b) provides another explanation of
the process of attitude change using the term of “elaboration likelihood” that refers to the
extent to which people think about issue-relevant arguments contained in persuasive
message. In the Elaboration Likelihood Model (ELM) of Petty and Cacioppo, when the
elaboration likelihood is high, the probability is high that recipients will follow the
central route (e. g., systematic process or argument based thinking) to persuasion; and
when the elaboration likelihood is low, the probability is high that recipients will follow
the peripheral route to persuasion.

Therefore, when subjects spend much time thinking about the CEPI presented
(e. g. Samsung Electronics is a severely polluting corporation), it is probable that the
process of attitude change seems likely to follow the central route. If subjects spend little
time thinking about the CEPI presented, the process of attitude change follows the
peripheral route. When subjects spend moderate time for thinking, the cognitive process

of attitude change would follow both central and peripheral route (i.e., concurrent or

192

interactive process of both). However, ELM did not make it clear for the case of
moderate elaboration.

Also, it is not clear about what routes (i.e., central or peripheral routes) subjects
adopted as the process of attitude change? According to the ELM, attitude changes that
result mostly from the central route would show greater temporal persistence, greater
prediction of behavior, and greater resistance to counter-persuasion than attitude changes
that result mostly from the peripheral route. However, this argument has not been proven
yet. So the remaining question for further study is whether the attitude change through
the central route really has greater temporal resistance or not.

Attitudes directly influence behavior or influence behavior through intentions6.
According to Eagly and Chaiken (1993), “An attitude toward a specific behavior directed
toward a given target in a given context at a given time should predict the specific
behavior quite well because this attitude exactly corresponds to the specific behavior” (p.
167). Even though it is not attitudes toward a specific behavior7, attitude toward targets
(i.e., corporations) also could strongly influence the behavior of purchasing products of
corresponding corporations. Winters (1989) suggests that consumers' attitude toward a
corporation has a direct effect on brand purchasing. Therefore, it is assumed that attitude
toward corporations (i.e., attitude toward target), even thought it is not attitude toward
behavior, also influences purchase behavior. In this respect, it is assumed that a positive

relationship exists between attitude toward corporations and purchase behavior (or

intention) toward products of the corresponding corporations. Based on this assumption,

 

6 The theory of reasoned action asserts that intentions have a direct effect on behavior and attitudes have a
direct effect only on behavioral intentions, not on behavior. About the reasoned action, refer to F ishbein
(1967 and 1980) and Ajzen(1988 and 1991).

Attitude toward behavior (e.g., buying Samsung Cellular phone) is a distinctly different class of attitude
than the attitude toward targets (e.g., Samsung Co.).

193

the finding related to consumer attitude change by CEPI disclosure is interpreted that
CEPI influences consumer purchase behavior through attitude change. That is, positive
CEPI influences positively consumer purchase behavior and negative CEPI influence

negatively consumer purchase behavior.

CEPI Disclosure and Consumer’s Perception of Corporate Credibility (H 2)

This study finds that CEPI damages or improves corporate credibility by the
confirmation of the Hypothesis Two that CEPI disclosures change consumer’s perception
of credibility of corporation (CC) positively for non-polluting corporations and
negatively for polluting corporations. In other words, this study finds two facts: 1)
information of corporate environmental performance influences consumer’s CC, and 2)
CEPI disclosure is one cause of diverse causes for consumer’s perceptional change about
credibility. The finding about credibility change by CEPI can be interpreted that CEPI
influences on consumer purchase behavior.

This study provides evidence supporting the PID advocates’ assertion that CEPI
damages or improves corporate image or reputation, and consequently, creates
repuational incentives for corporate pollution performance. The finding that CEPI
damages or improves corporate credibility is compatible with the assertion of PID
advocates that CEPI creates reputational incentives for pollution reduction because
environmental information of polluting corporations damages their image and reputation
and environmental information of non-polluting corporations improves their image and
reputation. The terms “reputation” or “image” are conceptually very similar to the term

of credibility. “Corporate credibility” is defined as the perceived expertise, truthfulness,

194

and/or honesty of the firm (Mackenzie & Lutz, 1989). Image is defined as the concept of
someone or something that is held or projected by the public and reputation is the general
estimation by which one is held by the public (N ewell, 1993). Hence, the term
"credibility" seems to be definitionally more specific than either of the terms "image" or
"reputation," and image and reputation, on the other hand, seem to be much broader in
scope, and encompass many other dimensions including credibility (N ewell, 1993). Thus,
change of corporate image or reputation can be considered as change of corporate
credibility. Therefore, PID advocators’ assertion that CEPI damages or improves
corporate image and reputation is simultaneously proved by this study’s confirmation of
Hypothesis Two that CEPI disclosures change consumer’s perception of credibility about
corporation positively for non-polluting corporations and negatively for polluting
corporations.

Meanwhile, corporate credibility seems to influence attitude toward corporation
and purchase behavior (or intention) of the products of the corresponding corporations.
Newell (1993) found that corporate credibility had a positive effect on attitude toward
brands (i.e., corporations). Fombrun (1996) asserted that high corporate credibility is
important in producing positive attitude changes toward the advertisement and toward the
brand as well as in influencing purchase intentions. Lafferty & Goldsmith’s (1999)
experiment also proved that corporate credibility or reputation influences consumers’
attitudes toward brand and purchase intentions. In this respect, the finding about
credibility change by CEPI can be interpreted like that CEPI influence consumer
purchase behavior. That is, positive CEPI influence positively consumer purchase

behavior and negative CEPI influence negatively consumer purchase behavior.

195

CEPI Disclosure and Consumer Purchase Intention Change (H 3)

This study finds that CEPI influences consumer purchase intention (PI) by the
confirmation of the Hypothesis Three that CEPI disclosures increase consumer purchase
intention for the products of non-polluting corporations and decrease consumer purchase
intention for the products of polluting corporations. This study also proves that CEPI
disclosure is one of many causes to change purchase intention.

The finding about PI implies that CEPI influences consumer purchase behavior
because purchase intention is considered to be identical to the purchase behavior in this
study. Intention is a psychological construct distinct from attitude. Intention represents
the person’s motivation or conscious plan to exert effort to carry out a behavior and
purchase intention has an effect on the purchase behavior (Eagly & Chaiken, 1993). As
mentioned in Chapter IV, purchase behavior can be well predicted from purchase
intention. Thus the finding about PI can be interpreted that CEPI influences consumer
purchase behavior. That is, positive CEPI influences consumer purchase behavior for the
products of non-polluting corporations and negative CEPI influences negatively

consumer purchase behavior for the products of polluting corporations.

Environmental Attitude and Sensitivity to CEPI (H 4)

This study finds that a consumer with high environmental attitude is more
sensitive to information of corporate environmental performance information than a
consumer with low environmental attitude. This study proved the Hypothesis Four that a

customer who has high environmental attitudes toward pollution (EAP) changes the three

196

dependent variables more than a customer who has low EAPg. In other words, a positive
relationship exists between EAP and the three dependent variables: attitude toward
corporation (AC), corporate credibility (CC), and purchase intention (PI).

This finding means that the extent of changes in consumers’ purchase behavior by
CEPI disclosure varies depending on the intensity of consumers’ environmental attitude
toward pollution because the extent of changes in consumers’ AC, CC, and PI varies
depending on the intensity of consumers’ environmental attitude.

The finding about the positive correlation between purchase behavior and
environmental attitude is consistent with Chan’s (1996) finding that respondents who
were more concerned about environmental issues tended to purchase more
environmentally fi‘iendly goods. The finding about the correlation between purchase
behavior and environmental attitude also supports the Roper Organization Inc.’s (1990)
finding that consumers with higher level of environmental attitude have stronger
tendency to “avoid buying products from companies not environmentally responsible,”
compared to consumer of total public. The finding about environmental attitude and
purchase behavior is helpful in explaining why the movements for green consumerism or
boycotts have been initiated mostly by environmentalists who apparently have high
environmental attitude. People who have high level of environmental attitude are mostly
much concerned with environmental issues or information. People with high

environmental attitude are very much in accordance with their attitude (Roper

 

8 In this research, environmental attitude specifies the environmental attitude toward pollution (EAP)
because the research is related to industry pollution. Thus, high EAP is defined as the status of a strong
concern, awareness, and belief about pollution. Low EAP is defined as the status of a weak awareness,
concern, and belief about pollution.

197

Organization Inc., 1990). Response to CEPI starts from concern about environmental
information.

The finding about the positive correlation between purchase behavior and
environmental attitude leads to the following interpretation that the public’s
environmental attitude toward pollution influences the effectiveness of the public
information disclosure as a pollution control tool. However, according to Roper
Organization Inc. (1990), the real story is more complex one than research interpretation.
They asserted that in general, there is a large gap between attitude and behavior about the
environment. In the real world, consumers purchase behavior is some times not consistent
and sometimes consistent with environmental attitude in terms of willing to pay more for
green products. That is, some consumers with high environmental attitude are not willing
to pay more for green products. Thus, this study limits the interpretation of the
confirmation of Hypothesis Four as to that environmental attitude is a factor influencing

effectiveness of public information disclosure as a pollution control tool.

Corporate Familiarity and Sensitivity to CEPI (H 5)

The researcher hypothesized that subjects who had prior knowledge about the
corporation (i.e., corporate familiarity) resist changing their attitude more than subjects
who have little knowledge about the corporation, based on the familiarity theory that the
less information possessed by an individual, the greater the change induced by any new
piece of information (Eagle and Chaiken, 1993). For instance, the less familiar a

consumer is with Samsung Electronics, the more unstable and easy to change the attitude

198

toward Samsung Electronics. A familiarity about Samsung Electronics is defined as a
proximate of the amount of stored information or knowledge about Samsung Electronics.

However, this study disconfirmed the Hypothesis Five that consumers with high
familiarity with a corporation (CFC) change consumer attitude, corporate credibility, and
purchase intention less than one with low familiarity. In other words, this study could not
find any relationship between CFC and the three dependent variables: attitude toward
corporation (AC), corporate credibility (CC), and purchase intention (PI). For example,
subjects exposed to positive Samsung CEPI showed negative relationship between
corporate familiarity and three variables (i.e., confirmation of H 5) but subjects exposed
to negative Samsung CEPI exhibited positive relationship between them (i.e.,
disconfirmation of H 5). It means that subjects with high familiarity of Samsung
Electronics showed both large and small change of AC, CC and PI and subjects with low
familiarity of Samsung Electronics also showed both large and small change of AC, CC
and PI. Thus, this finding implies that corporate familiarity is not a moderator variable
(i.e., severe contingency variableg) so that corporate familiarity is not a variable
influencing effectiveness of public information disclosure as a pollution control tool.

It is uncertain why the hypothesis about familiarity (H 5) was rejected in Korea. It
could be argued that because Korean peoples’ awareness and concern about
environmental problems are sufficiently strong10 that they may override their corporate
familiarity and its affect on corporate attitude. If this argument were true, all subjects

should show large degrees of attitude change. However, subjects also showed only small

 

9 Contingency variable is defined as a variable, of which effect occurs in both positive and negative
direction. Contingency variable is a special case of the moderator variable.

'0 The observed environmental attitude toward pollution of Korean university students was moderately
strong (M = 3.8, SD = 0.43 in S-point scale).

199

degrees of attitude change. Thus, this argument would not be true. For a better
explanation about the rejection of H 5, further study is needed.

Because corporate familiarity is not a moderator variable (i.e., little relationship
between corporate familiarity and consumers’ change of attitude, credibility, and
purchase intention), PID could be workable for even less known brands or corporations.
In other words, whether the corporation is well known or not (i.e., large or small
corporation), CEPI disclosure would influence consumers’ attitude, perception of
corporate credibility, and purchase intention. Therefore, the target of PID program need
not be limited to famous and large corporations.

This finding about familiarity is inconsistent with the social psychological theory
of familiarity: The less information possessed by an individual, the greater the change
induced by any new element of information. The familiarity theory says that the amount
of stored information or knowledge that is available and accessible to people moderates
attitude-behavior correspondence (Eagle & Chaiken, 1993). Lavine, Huff, Wagner and
Sweeney (1998) also assert that one with more information has a stronger and a more
stable attitude than one with less information. For example, many consumers do not
change their purchase behavior even though they are exposed to information of corporate
environmental performance because they already have strong familiarity (or preferences)
for specific products or brands. That is, the familiarity theory implies that consumers with
much knowledge or information about corporation are not sensitive to CEPI for changing
their attitude or credibility toward corporation and consumers with much knowledge or
information about a certain product is not sensitive to CEPI for changing their purchase

intention toward a corresponding product. However, data revealed that subjects with low

200

familiarity were sensitive or not sensitive to CEPI and subjects with high familiarity were
sensitive or not sensitive to CEPI. Familiarity theory asserts a relationship between
consumers’ familiarity with corporations and consumers’ purchase attitude and/or
behavior changes. However, this study did not find any relationship between corporate
familiarity and change of attitude, credibility and purchase intention. Thus, it is apparent
that familiarity theory does not work well for the case of CEPI disclosure and purchase
behavior change, even though the clear explanation about why familiarity theory does not
work well in the case of CEPI disclosure and purchase behavior change is not found.
Instead of familiarity theory, discrepancy theory provides some inference about
CBPI disclosure and purchase behavior change. Discrepancy in this theory means a
positional discrepancy between of message or information newly received and message
or information initially possessed. Discrepancy theory says that the greater the positional
discrepancy of a message, the more attitude change the message will produce (Kaplowitz
& Fink, 1997). A long time ago, Brock and Becker (1965) demonstrated that regardless
of overall prestige, a communicator was more persuasive when arguing against rather
than in accord with his self-interest. Hunter, Danes and Cohen (1984) 11 asserted that the
greater the discrepancy between the message sent to a person and the person’s attitude,
the greater the amount of change one would observe, keeping other factors constant.
Eagly and Chaiken (1993) also asserts that unexpected messages are more persuasive
than expected ones. This positional discrepancy theory is similar to cognitive dissonance
theory. Kaplowitz and Fink (1997), and Aronson, Turner and Carlsmith (1963) proposed

that the greater the discrepancy of the message fi'om the receiver, the greater the

 

” They describe their assertion as “information processing” theory. Also, Hunter, Levine and Sayres (1984)
dealt with a model that was based on the discrepancy theory in situation where one attitude also affected
other attitudes.

201

cognitive dissonance. The dissonance created by a discrepant message may be resolved
by changing one’s opinion or by derogating the message source (taking away the
message), denying the information or message itself. As discrepancy increases, source
derogation, rather than opinion change is employed to resolve one’s dissonance. The
theory of discrepancy and dissonance are also consistent with the inoculation theory.
McGuire (1964) reasoned that attitudes and beliefs are vulnerable to persuasive attack by
opposing arguments, and reasoned that protection against such attack may be achieved by
exposing people to weakened forms of the attacking message. A relatively weak attack
should encourage people to develop defenses against such attacks while not changing
their attitudes or beliefs. Application of discrepancy theory to this study is that the CEPI
of experimental stimuli was a high discrepant information to subjects who strongly
changed their attitude but it was a low discrepant information to subjects who changed
little their attitude. For example, when subjects previously possessed information that
Samsung Electronics was a non-polluting corporation but they were given to the CEPI
that Samsung Electronics was a severely polluting corporation, the CEPI is highly
discrepant information to them so that subject in this case would largely change their
attitude. As a reverse example, when subjects previously thought that Samsung
Electronics was a non-polluting corporation but they were given to the CEPI that
Samsung Electronics was a great corporation in terms of corporate environmental
performance, the CEPI is little discrepant information so that subjects would not change
their attitude. In belief, the discrepancy theory (or the dissonance or inoculation theory)
implies that consumers with more positional discrepancy between new information of

corporate environmental performance gained and the prior environmental information

202

possessed change their attitude more. Thus, based on this discrepancy theory, it can be
inferred that a positive relationship exists between CEPI discrepancy and AC, CC and PI
change.

However, this study cannot investigate the inference about CEPI discrepancy
because it does not contain sufficient data. Familiarity observed in this study is about the
general information about corporation, not about specific information about corporate
enviromnental performance. Regretfully, our data cannot answer whether subjects
possessed much prior information or little prior information of Samsung Electronics,
Hyundai Electronics, Nongsim and Binggrae because the experimental questionnaire is
designed to measure the amount of general knowledge about the four corporations.

Another inference based on the discrepancy theory is that Korean subjects have
little environmental performance information of the Korean four corporations: Samsung
Electronics, Hyundai Electronics, Nongsim, and Binggrae. This inference is based on
three points. The first point; it is assumed that subjects have almost an equal amount of
stored CEPI of Samsung, Hyundai, Nongsim, or Binggrae regardless of whether group A
or B in the pre-test that is before subjects gain the new information of the four
corporations. Subjects have almost equal corporate familiarity of the four corporations
regardless of whether they were in Group A or B12 so that it is assumed that subjects have
almost an equal amount of stored information about the four corporations. The second
point is that Group A subjects and Group B subjects had the same quality of information:
positive or negative CEPI. For example, subjects of both Group A and B have almost

same information that Hyundai is a polluting corporation or non-polluting corporations.

 

’2 Mean familiarity difference between group A and B is not statistically significant at or = .05 and two-
tailed test: Samsung (t = .234, p = .815), Hyundai (t = .722, p = .471), Nongsim (t = 2.531, p = .012) and
Binggrae (t = 1.838, p = .067) .

203

The third point; it is assumed that subjects have little amount of CEPI of the four
corporations. The first and second point implies that both Group A and Group B subjects
might have little amount of CEPI, or the same and much amount of CEPI. If both groups
had the greater amount of CEPI, one group would clearly change AC, CC or PI because
of high discrepancy while another group would change little them because of little
discrepancy. For example, in the case that 1) both Group A and Group B have
information about Samsung that Samsung is a non-polluting corporation, and that 2)
Group A subjects are given to the positive CEPI that Samsung is a non-polluting
corporation and 3) Group B subject are given to the negative CEPI that Samsung is a
severely polluting corporation, Group B subjects will experience high discrepancy
between stored Samsung CEPI (non-polluting corporation) and new Samsung CEPI
(polluting corporation), consequently, would change their AC, CC and PI. However,
Group A subjects would not change their AC, CC and PI because they experience little
discrepancy between stored Samsung CEPI (non-polluting corporation) and new
Samsung CEPI (non-polluting corporation).

Meanwhile, data analysis showed that regardless of Group A or B, both groups
clearly changed their AC, CC and PI for the four corporations. It means that subjects may
had little amount of CEPI of the four corporations. For example, subjects have little
information about Samsung’s environmental performance with the result that any new
information (whether it is positive or negative) is highly discrepant. Consequently,
subjects were sensitive to both positive and negative CEPI of Samsung. But if subjects
have a great deal of the same information about Samsung environmental performance, it

is impossible for both groups to clearly change their AC, CC and PI because both groups

204

have the same information (i.e., one of positive or negative information). The inference
that Korean subjects might have little information of the four corporations’
environmental performances is understandable because Korean government has not
disclosed any environmental information about individual corporations to the Korean
public.

The last inference based on the discrepancy theory is that the environmental
reputations of Samsung, Nongsim, and Binggrae are positive and of Hyundai is both
positive and negative. A pattern is found from the test of Hypothesis Five (Refer to Table

7.3).

Table 7.313

Summary of Test of Hypothesis Five

 

Familiarity Group: Attitude Credibility Purchase Intention

 

 

CEPI H 5 H 5 H 5
Samsung A:+ C C C
Samsung B: - N N N
Hyundai A: - C F C
Hyundai B:+ F C F
Nongsim A:+ C C C
Nongsim B: - N F F
Binggrae A: - N F N
Binggrae B:+ C C C

 

- A = Group A / B = Group B in the experiment.
- " - " = Negative CEPI

- " + "= Positive CEPI

- C = Confirmed

- F = Further Research is required

 

’3 For reader’s convenience, the same Table 7.3 located in the previous page is brought again.

205

- N = Not Confirmed

Table 7.3 reveals a pattern that for Samsung, Nongsim and Binggrae, H 5 is
confirmed in the case of positive CEPI and it is disconfirmed in the case negative CEPI.
However, for Hyundai, H 5 is confirmed in the case of both positive and negative. For
example, when subjects were exposed to the positive Samsung CEPI that Samsung is a
non-polluting corporation, subjects with much knowledge or information about Samsung
did not change their AC, CC and PI much while subjects with little knowledge or
information about Samsung changed their AC, CC and PI a great deal.

This pattern can be explained by the discrepancy theory. For example, assuming
that Samsung’s environmental reputation is basically positive so that subjects believed
that Samsung is a non-polluting corporation, when subjects were exposed to positive
CEPI stimuli, there was no discrepancy between subjects’ stored knowledge (i.e., non-
polluting corporation) and new information of stimuli. Consequently, subjects did not
change their AC, CC and PI much because there was no discrepancy in environmental
information. When the subjects were exposed to negative CEPI stimuli, there was high
discrepancy so that they changed their AC, CC and PI. The case of Binggrae and
Nongsim can be explained by the same reasoning with the case of Samsung. The case of
Hyundai is a mixed case: the same case of Samsung in the change of CC and PI but the
reverse case with Samsung in AC change. Thus, it is inferred that Hyundai’s

environmental reputation is both positive and negative.

206

Therefore, the environmental reputations of Samsung, Nongsim, and Binggrae
would be more positive than Hyundai. This point can be supported by data of attitude and

credibility toward the four corporations.

Table 7.5. Attitude and Credibility toward Four Corporations before Stimuli

 

Samsung Hyundai Nongsim Binggrae

 

 

Mean

Attitude 5.33 3.92 5.59 4.18
Credibility 5.18 3.99 5.33 4.19
Std. of Deviation

Attitude 0.97 1.04 0.84 0.95
Credibility 0.83 0.93 0.81 0.86

 

Figure 7.13. Attitude and Credibility for Four Corporations

 

Attitude and Credibility

 

IAttitude
D Credibility

 

 

Mean
O—SNw-bUIODV

 

 

Samsung Hyundai Nongsim Binggrae

 

 

 

207

Table 7.5 and Figure 7.13 show that the credibility and attitude of Samsung,
Nongsim and Binggrae is higher than Hyundai’s. Corporate environmental reputation or
image influences corporate attitude and credibility.” That is, there is a positive
relationship between corporate environmental reputation and corporate credibility and
attitude. Thus, higher credibility and attitude means higher environmental reputation.
Hyundai’s credibility and attitude is relatively lower than other three corporations. Thus,
it is inferred that environmental reputation or image of Samsung, Nongsim, and Binggrae

are more positive than Hyundai’s.

CEPI Credibility and Effectiveness of CEPI Disclosure (H 6)

This study finds that CEPI credibility influences effectiveness of CEPI disclosure
on consumer purchase behavior because the credibility of CEPI moderates the change of
consumer attitude, corporate credibility, and purchase intention. Test confirmed the
Hypothesis Six that the greater the credibility of CEPI (CCI), the greater the effect of the
CEPI on the dependent variables. That is, there is a positive relationship between CCI
and the three dependent variables: AC, CC, and PI.

This finding supports the theory of credibility that source credibility moderates
the intensity of attitude change (Eagle & Chaiken, 1993). Aronson, Turner and Carlsmith
(1963) found that for messages from a high credibility source, there was a substantial
increase in opinion change but when the source had only a moderate level of credibility,
opinion change decreased. Credibility of information source is identical to credibility of

information because information credibility is depending on credibility of information

 

'4 Regarding this assertion, refer to discussion about credibility and attitude in Chapter VII: p. 190 and p.
194.

208

source such as the US. EPA. Thus, this study measured information credibility, instead
of information source credibility. In general, the credibility theory is highly acceptable
because it is a kind of truth (i.e., truism). The US. EPA or the Korean Ministry of
Environment (KME) is a source of environmental information. When the US. EPA or
KME15 is highly credible to their public, the public will believe information of corporate
environmental performance provided by them. People do not accept knowledge that is
not credible. This is the reason why BAPEDAL (Indonesia’s National Pollution Control
Agency) and NIPR (the New Ideas in Pollution Regulation: a group of the World Bank)
who initiated the program PROFER, made considerable efforts to increase the credibility
of information of Indonesian corporate environmental performance. To increase the
credibility of CEPI, they based the grading system upon multiple sources of data,
conducted independent inspections, developed a user fi'iendly computer program for
analyzing the data, and designed a multiple step process for reviewing proposed grades
before disclosing CEPI to the public. Therefore, credibility of CEPI or credibility of
source of CEPI will be a critical element for effectiveness of PID (Public Information
Disclosure for Pollution Control).

In summary, this study observed credibility of information rather than credibility
of information source and data confirmed the Hypothesis Six. Thus, this study found that
credibility of CEPI directly influence effectiveness of CEPI disclosure on consumer

purchase behavior change.

 

’5 Credibility of information provided by KME was rated as 4.58 based on 7-point scale (M = 4.58 and SD
= 1.10). The 4.58 means slightly credible.

209

CEPI Disclosure and Consumer Purchase Behavior Change

This study concludes that corporate environmental performance information
(CEPI) influences consumer purchase behavior, based on the points that attitude and
credibility of corporations and purchase intention toward corresponding corporate
products predicts consumer purchase behavior toward the products. Thus, it is predicted
that 1) positive CEPI increases consumer purchase behavior and 2) negative CEPI
decreases consumer purchase behavior. These conclusions are proven by the
confirmation of the Hypothesis One, Two and Three: 1) positive CEPI changes positively
consumer attitude and credibility of corporations and purchase intention toward corporate
products and 2) negative CEPI changes negatively consumer attitude and credibility of
corporations and purchase intention toward corporate products. Attitude and credibility of
corporations directly influence consumers purchase behavior and purchase intention is
considered to be comparable to purchase behavior because behavior can be well predicted
from intention. Based on this relationship between attitude, credibility, purchase
intention, and purchase behavior, this study concludes that CEPI is a cause influencing
consumers to change their purchase behavior toward products of polluting or non-
polluting corporations.

This study’s findings are in accordance with Wheeler (1997)’s assertion that
environmental reputation does matter to corporations whose expected costs or revenues
are affected by judgments of environmental performance by customers, suppliers, and
stockholders. This study’s findings also supports Lanoie, Laplante and Roy (1997)’s

assertion that collective information, which compares firms with bad performance to

210

those with good performance, is more likely to allow consumers to switch away from the
products produced by the firms with bad environmental performance.

This study also found that the extent of purchase behavior change is influenced by
consumer’s environmental attitude and credibility of information, but not by consumer’s
corporate familiarity because the extent of attitude, credibility toward the corporations,
and of purchase intention toward products are influenced by consumer’s environmental

attitude and credibility of information, but not by consumer’s corporate familiarity.

Effect Pattern by Tm of Corporation and Information

No pattern was founded fiom negative and positive information disclosure as it
relates to the magnitude of change in the three dependent variables of attitude, credibility,
and purchase intention. However, it was found that subjects showed a more discemable
response to negative CEPI than to positive CEPI. This can be understood by realizing that
peoples have a behavioral tendency to severely punish polluting behavior because of their
perception of a direct health impact from pollution.

No pattern by type of corporation (food or electronics) was found in the three
variables. Two corporations (Samsung and Hyundai) are companies that produce
electronics equipment (e.g., cellular phones); other two corporations (Nongsim and
Binggrae) are corporations that produce food (e. g., instant noodle). Thus, it is possible to
infer that consumers are more sensitive to CEPI of food corporations than to CEPI of
electronics corporations, assuming that peoples are very sensitive to environmental
materials affecting on human health. However, data about the four corporations

(Samsung, Hyundai, Nongsim, and Binggrae) did not support this inference.

211

Meanwhile, it was found that subjects responded more strongly to negative
information about Samsung and Nongsim, but they responded more strongly to positive
information about Hyundai for all three variables. This can be explained by the
discrepancy theory that the greater the positional discrepancy of a message, the more
change in attitude the message will produce (Kaplowitz & Fink, 1997). Initial levels of
attitude and credibility for Samsung and Nongsim were relatively higher than for
Hyundai”. Negative CEPI of Samsung and Nongsim seems to produce greater positional
discrepancy due to subject’s relatively positive attitude and perception of credibility for
Samsung and Nongsim. Positive CEPI of Hyundai also seems to produce greater
positional discrepancy due to relatively negative attitude and credibility toward Hyundai.

However, the conclusions about the effect pattern by the nature of corporation and
information need to be delayed. The inferences about the effect pattern by nature of
corporation and information were investigated by only four cases of corporations. The
four cases are too small as a sample size to draw a conclusion. Therefore, further research
with a larger sample is required to draw a conclusion about effect pattern by nature of

corporation and information.

Theory Test and Application of Results

Calder, Phillips and Tybout (1981) identified two types of application of
consumer behavioral research: effects application research and theory application
research. The goal of effect research is to obtain findings that can be generalized directly

to a real world situation of interest. The goal of theory research is to obtain scientific

 

'6 Average intensity of attitude and credibility measured before stimuli (i.e., in pre-test) are as follows:
Samsung — 5..33 (attitude), 5.18 (credibility), Nongsim - 5.59, 5.33, and Hyundai — 3.92, 3.99 based on the
7-point scale. Refer to Table 7.5 and Figure 7.13.

212

theory that can be generalized through the design of theory-based interventions that are
available in the real world. Thus, effect research adopts a research method of random
sampling to represent statistically the real world population so that samples are
heterogeneous because of heterogeneity of population. Theory research prefers
experimental designs and homogenous subjects because they allow the strongest theory
test. For theory test, it is not necessary that a test setting should be an accurate
representation of the real world situation but it is important that the test setting allow
observing a particular event (e.g., causational relationships among factors involved in the
operationalization of theory) at research issue.

A goal of this study was to test a theory of PID (Public Information Disclosure for
Pollution Control) that CEPI disclosure generates reputational incentives for
corporations, and as a result, leads to their voluntary reduction of pollution. In other
words, a purpose of this study was to investigate whether CEPI disclosure causes
consumers to change their purchase behavior away from or toward the products of
polluting or non-polluting corporation. To observe the causal relationship between CEPI
disclosure and purchase behavior change, this study employed an experimental test with a
homogenous subject of Korean university students.

The population to which findings of this study can be applied, could be 1) all
Korean university students, 2) all university students of the world (e. g., American
university students), 3) all Korean consumers, or 4) all consumers of the world (e. g.,
American consumers). However, random sampling was not employed with the result that
participants of Korean university students (Group A: 154 and Group B: 152) of this

experiment could not be representatives of the real world populations. Thus, knowledge

213

from the results of this study cannot be applied to a general population in a same way
applicable to effect research such as a survey research sample.

This study observed subjects’ response to CEPI of only four corporations -
Samsung and Hyundai (electronic corporations), Nongsim and Binggrae (food
corporations), and observed purchase behavior change toward only two kinds of products
- cellular phone and instant noodle. If CEPI of other corporations (i.e., Daewoo Co., Kia
Car Co., etc.) and other products (i.e., golf clubs, TV, shoes, clothing, etc.) were
presented to subjects, it is not certain that subjects would response with the same results
as were the cases in this study. Thus, the study’s findings cannot be generalized to all of
corporations or all of products.

One example can help us to understand how to apply the results of the experiment
to a general population. Suppose that a new pain relief medicine such as Tylenol was
tested with one hundred and fifty Korean university students and clear effect of pain
relief was observed in the direction predicted. Of course, individual participants showed
the wide variety range of effect size (i.e., the intensity of pain relief). Can the results of
this experiment be generalized to all university student of Korea, America and the world,
and all people of Korea, America and the world? The answer is ‘Yes’ because human
body of Korean students responded to a new Tylenol with clear effect of pain relief. All
individuals have a different body conditions and groups of people in terms of nationality,
race, gender, age, etc. have a different body conditions so the individual or group’s
responses to a new Tylenol will be diverse in them of effect size (e. g., Tylenol works
well for some body but is not effective for every body). However, all people regardless

age (i.e., university student or not) and nationality are the same human beings in terms of

214

the human body. Thus, if the new Tylenol works well for Korean university students, we
can conclude that the new Tylenol effects well for all people regardless of age and
country. Only for more reliable and valid knowledge about efficacy of the new Tylenol,
the new Tylenol needs to be tested with a variety group of peoples (i.e., American) who
have a distinctly different body condition.

Using the same logic of the example of the new Tylenol, this study generalizes its
findings to populations. All individuals are different in terms of purchase behavior
regardless of age and country but they are also the same consumer in terms of human
system. A group of Korean university students is homogenous with other groups (e. g.
adults consumer, American consumer) because they are all human beings, but it is
heterogeneity with them in terms of consuming pattern, preference, or sensitivity to
environmental information. Thus, effectiveness of CEPI disclosure may differ among
individuals and consumer groups. But as consumers, they are also the same as human
being so that it is assumed that they have common psychological tendencies (i.e., attitude,
credibility and purchase intention) related to their purchase behavior. Therefore, there
would be no problem to generalize the findings of this study to the population (i.e.,
Korean and American university student, Korean and American consumers, and all world
consumers), basically acknowledging a wide variety of individual differences and

consumer groups’ differences.

Different Consumer Group and Generalization of the Study Results

Roper Organization Inc. (1990) classified consumers into five groups by the

criteria of pro-environmental activities such as willingness to pay more for green

215

products or supporting environmental regulations, based on data from 1989 survey: 1)

The True-Blue Greens (ll %), 2) The Greenback Greens (11 %), 3) The Sprouts (26 %),

4) The Grousers (24 %) and 5) The Basic Browns (28 %). The detailed identification of

the five consumer groups is as follows:

1.

The True-Blue Greens are the committed environmentalists. They believe that
individual actions can make a difference in protecting the environment. Their
behavior is very much in accordance with their attitudes. They are well
educated and have the highest median household income of any of the five
groups.

The Greenback Greens are a little less committed environmentalists that the
True-Blue Greens. They believe strongly in the primacy of protecting the
environment over economic development. They attribute their not doing more
for the environment to their busy lifestyles. They have the highest proportion
of working people (70 % are employed) as well as more full-time workers
than others (55%).

The Sprouts are the swing group in a movement toward a green consumer
society. They have adopted some pro—environmental behavior but they are far
behind the Greenback Greens in their willingness to make personal economic
sacrifices for the environment. They are quite well educated and have higher
than average household incomes.

The Grousers are usual consumers who are indifferent to environmental

protection and are pursuing convenience. They show a low level of pro-

216

environmental behavior and a low level of interest in paying higher prices for
green products. They are relatively poorly educated and have below-average
household incomes.

5. The Basic Browns are indifferent to environmental protection and believe that
there is nothing an individual can do about most environmental problems,
consequently, they are not carrying out any pro-environmental activities. This
group is the least exposed to information about environment and they are
mostly socially and economically disadvantaged group of the five (i.e., the

lowest household income and the least educated).

Referring to Roper Organization Inc.’s study, let us generalize the Inference
Probability (IP) of this study. Data from three hundred and six Korean university students
revealed that IP was 1.00 for attitude change toward Samsung. The first population for
generalization of IP is a group of all Korean university students. It is assumed that sample
students and population students are generally the same consumer group in terms of
consumption patterns, preferences, environmental attitude, etc. Thus, if other Korean
university students participate in an experiment of exactly the same condition and the
same time (the experiment was conducted May and June, 2001), it is inferred that we
could get almost the same results (i.e., IP = 1.00 or a little less). Thus, interpretation of IP
for attitude change is that the probability that Korean university students sensitively
respond to CEPI of the four corporations, and consequently, change their attitude
negatively for polluting corporations and positively for non-polluting corporations, would

be almost 100%.

217

However, for groups of all Korean consumers, all American university students
or consumers, the generalization should be carefully stated. It is because there is a wide
variety of consumer groups, and each consumer group has a different consumption
pattern, preference, and different level of environmental attitude so that they have
different sensitivities to CEPI and will show different intensity of response to positive
and negative CEPI. Thus, even if an experiment were conducted for different consumer
groups, the results of IP for attitude change by CEPI would be not the same (e. g., [P =
1.00 for some group, [P = .95 for some group, IP = .90 for some groups). However, it is
also possible to infer that there will be little opposite results to this study’s because they
are almost the same consumer in terms of living in the modern consumption society and
they are facing almost the same nature of domestic or global environmental problems.
Thus, Hypothesis One (i.e., attitudes change in response to CEPI) would be confirmed for
other consumer groups (i.e., Korean consumer and American university consumer), even
though the intensity of effect by CEPI disclosure would be different for each group. The
generalization of IP (Inference Probability) about the variables of Credibility, Purchase
Intention, Environmental Attitude and Information Credibility can be done in the same
way with the generalization of IP for the variable of Attitude.

In this respect, the application of findings of this research is that CEPI disclosure
can influence Korean university students to change their attitude toward corporations and
perception of credibility of corporations, and to change purchase intention toward
products. As a result, one can expect them to change their purchase behavior of products

of non-polluting or polluting corporations.

218

One thing required for more valid and reliable generalization of this study’s

findings is firrther experimental tests with diverse groups of consumers.

The Real World and Interpretation of Consumer’s Sensitivity to CEPI

 

In the real world, consumers’ purchase behavior is more complex than the theory
of PID suggests because multiple and diverse factors are involved in the process of
purchase decision. Consumers decide on their product purchase based on quality, price,
previous experience, individual preference, brand reputation, etc. In general, there is a
large gap between consumers’ attitude and behavior about the environment. According to
the poll conducted by the Pew Center for the People and the Press at the end of 1997, 73
percent of American respondents said they would pay five cents more for a gallon of
gasoline if the higher price would "significantly reduce global warming.” When the
suggested price increase was raised to 25 cents per gallon, 60 percent of American
respondents said they would be willing to pay (Gerstenzang, 1997). Rosendahl (1990)
also found similar results from 1989 survey. That is, 77 percent of American respondents
said that a company’s environmental reputation affected what they buy, 89 percent said
they were concerned about the environmental impact of products purchased, and 78
percent said they were willing to pay more for recyclable or biodegradable packaging.
However, a survey by the Roper Organization Inc. (1990) showed a large gap between
consumers’ attitude and behavior about the environment. For example, only 16 percent of
the respondents selected “avoiding buying products from companies they consider not
environmentally responsible” compared to 54 percent of a consumer group of The True-

Blue Greens who are the committed environmentalists. A great deal of anecdotal

219

evidence exists to show that the majority of consumers do not avoid buying the products
from some high polluting corporations, and consumers’ avoidance of buying products of
polluting corporation is mostly temporary.

”17 that is

According to the theory of reasoned action, the “subjective norm
influenced by the personal preferences and opinions of significant others (e. g.,
consumers’ parents, kids, lovers, friends, teachers, boss, etc.), directly influence
consumers’ final purchase decision (Fishbein, 1967; Fishbein, 1980; Ajzen, 1988; Ajzen
1991). Subjects of this study also showed a diverse pattern of change in attitude toward
the four corporations and credibility of the four corporations and purchase intention
toward the four corporate products. Some subjects were sensitive to CEPI information,
but others were not. Some subjects responded to CEPI in the reverse direction predicted.
Therefore, it is not valid to interpret the findings of this study that all subjects or all
consumers will change their purchase behavior in the direction hypothesized, in response
to CEPI. That is, individual differences are present.

This research found high probability that a majority of subjects change their
purchase behavior in the direction predicted, responding to CEPI of the four corporations.
Thus, it is appropriate to interpret the finding that significant number of consumers tends
to respond to CEPI and to change their purchase behavior and CEPI is a factor
influencing consumer purchase behavior. More precise interpretation of this finding is
that CEPI has become a critical factor influencing consumers ' purchase decision when
all other critical factors for purchase behavior (e. g. price and quality of products, brand

reputations and preferences, etc.) are the same.

 

'7 Subjective norm is a person’s belief about whether significant others think that he or she should engage
in the behavior. Significant others are individuals whose preferences about a person’s behavior in this
domain are important to him or her (Eagle & Chaiken, 1993, p. 169).

220

Lessons from Moderator Study for PID Policy Design

This study investigated three moderator variables: environmental attitude toward
pollution, familiarity with corporation, and information credibility. The moderator
variable is a special case of the masking variables that are unknown variables affecting
the dependent variable. Knowledge of moderator variables for CEPI and purchase
behavior is practically valuable for pollution policy designers because the primary goal of
pollution policy designers is to develop a cost-effective pollution control tool. Moderator
study for effectiveness of CEPI disclosure will provide ideas and knowledge for pollution

policy makers for more effective and more inexpensive pollution control programs of

PID.

Study for Moderator of Effectiveness of PID

Moderator is defined as a variable that intervenes between independent and
dependent variable and influences the degree of dependent variable. For example, CEPI
causes consumer attitude change. Highly credible CEPI changes consumer attitude a
great deal toward the corporation and little credible CEPI changes their attitude toward
corporations very little. Thus, credibility of CEPI influences the intensity of attitude
change. However, it is neither an independent variable nor an independent variable for

attitude. The independent variable is “CEPI.” Refer to Figure 7.14.

221

Figure 7.14. Moderator Variable

 

High Credibility of CEPI * Much Change

 

 

Low Credibility of CEPI P Little Change

The mediator variable is a middle variable located between the first and the last
variable in the causal diagram so that the mediator is both independent and dependent
variable. For example, CEPI changes consumer purchase behavior. Change of purchase
behavior changes corporate environmental performance - pollution reduction. So
purchase behavior is located between CEPI and pollution reduction because it is an
independent variable in relation to corporate pollution reduction and a dependent variable
for CEPI. CEPI is not a direct causation for corporate pollution reduction. Direct
causation for pollution reduction is a purchase behavior. Thus, purchase behavior is

defined as a mediator variable. Refer to Figure 7.15.

Figure 7.15. Mediator Variable
Purchase Behavior

CEPI Pollution Reduction

222

This study found that environmental attitude and information credibility are
moderator variables while corporate familiarity is not. However, data revealed a
possibility that there are many moderator variables. Table 623'8 shows that all Stimuli by
Subject Interaction of raw score of standard deviation and standard score standard
deviation are statistically significant at the level of one-tailed and or = .05. The fact that
standard deviation is not a zero and statistically significant, means that l) masking
variables that are unknown variables affecting the intensity of influence of CEPI on
purchase behavior, 2) situational random factors (e. g., hearing news about Samsung
Electronics’ bad behavior just before administration of post-test), or 3) random
functioning of human psychology (e. g., participants’ varying thought or emotion during
the experiment). It was very difficult to control randomly functioning variables of the
situational factors and the human psychology in the experiment because of the feature of
their randomness. However, unknown masking variables can be investigated if the theory

about masking variables are developed.

Lessons from Moderator Study for PID Policy Desigp

This study found that information credibility and environmental attitude were
critical elements for the effectiveness of PID because people who had a high level of
environmental attitude responded more sensitively to CEPI by changing their purchase
behavior than people who had a low level of environmental attitude, and people

responded more sensitively to highly credible CEPI by changing their purchase behavior

 

'8 Go back at p. for Table 6.23.

223

than to little credible CEPI. Meanwhile, most advocates for PID emphasize the
readability of CEPI because the public is able to read information in the easily understood
format.

CEPI in the PID system works in the following sequence: 1) peoples’ concem
about information, 2) their acceptance of information, and 3) their response to
information. Thus, environmental attitude, information credibility, and readability are

involved in these processes as follows:

1. One’s awareness of environmental issues is involved in the process of concern
about environmental information. People with high environmental awareness,
as committed environmentalists, possess strong concerns about environmental
information, compared to people with low environmental awareness (i.e.,
environmental attitude).

2. Information credibility is essential to the process of accepting information.
Informed people accept only credible information and reject information that
is not credible.

3. Another element involved in acceptance of information is the presentation of
information. The public may not accept or understand information in the
format of statistical data, scientific methods, academic writings, etc. because
they may not comprehend the information. PID is a pollution control strategy
that provides information about corporate environmental performance to the
common people. Thus, CEPI should be presented in a format that is easily

read and understood.

224

The effectiveness of PID related to products market varies, depending on the
following elements: 1) the public’s environmental awareness and attitude, 2) the
credibility and the presentation of information, 3) corporations and 4) their products.
When the public is highly concerned about environmental issues and environmental
information, PID is expected to be highly effective. Even though people are attentive to
CEPI presented by government, if it is not understandable and credible, information will
not be employed in the cognitive processes of many people. The degree of change of
consumers’ purchase behavior in response to positive or negative CEPI also varies,
depending on the corporations and their products because of individual differences in
their preference to a certain corporation or products.

The public’s environmental awareness and attitude might be beyond a concern of
policy designers because changing public attitudes takes long time and could be costly.
The elements of the products and the corporation also would be out of a concern of policy
makers because environmental regulatory authority cannot discriminate the disclosure of
CEPI in terms of policy implementation. When the CEPI of some corporations is
disclosed and that of other corporations is not disclosed, credibility of CEPI will weaken
and government will face strong objections from corporations whose negative CEPI is
made public. Basically the proposed PID is to disclose fairly the environmental
information of all corporations within its societal boundary.

However, the quality of information is a controllable element for policy designers.
Thus, this study emphasizes the point that quality of information is a critical element for

high effectiveness of PID. The designers of PROPER (Program for Pollution Control,

225

Evaluation and Rating: a PID program developed in Indonesia in 1995) had already
perceived the importance of credibility and readability of information so that they put
much effort in order to increase credibility and readability of environmental information
about Indonesian corporations. As PID advocates indicated, this study also points out that
in order to achieve a strong effectiveness of PID for pollution control, policy designers of
PID need to exercise a high level of care to assure a high level of credible and readable
information.

As a last finding from the moderator study, the relationship between corporate
familiarity and purchase behavior change, points out that the target of PID program is not
necessarily limited to well known and large corporations because PID could be workable
for even less known brands or corporations.

This finding refutes the previously argued limitation19 that PID will not be
effective for pollution reduction of the reputationallly insensitive corporations. Some less
known small corporations do not care about maintaining their reputation, even though
most large corporations seem to be very sensitive to their image or reputation. Small
corporations seem to have less sense of social responsibility for a clean environment. PID
would not be effective for these small corporations because they are indifferent to
corporate images or reputations.20 However, according to the finding about the
relationship between the degree of effectiveness of CEPI disclosure and corporate
familiarity, PID could be workable for even these small corporations, based on the
assumption that less known corporation are comparable to small corporations indifferent

to their reputations. When reputationally insensitive small corporations experience a

 

'9 Refer to Ch. 11 Limitations of PID, p. 35.
2" Refer to Ch. 11 Limitations of PID, p. 35.

226

reduced market share and decreased profits caused by CEPI disclosure, there would be no
way but for even less sensitive small corporations to reduce amount of pollution
emission. Thus, it is inferred that CEPI disclosure encourages even less known and
reputationally insensitive corporations to become sensitive to their environmental
reputation.

A question remains about the numerous small corporations that have no public
image or recognition because they are simply suppliers to the large corporations. In this
case, CEPI disclosure could also influence the small suppliers to reduce their pollution.
The large corporations are buyers that are in fact a type of consumer. Some large buyers
started to equate small supplier corporations’ environmentally risky operation with
increased cost of business (Hoffman, 2000 a). For example, Levi Strauss & Company
developed strict sourcing guidelines for selecting contractors that are different than the
traditional guidelines about price, quality, and delivery time (Hoffman, 2000 b). Thus, the
large corporations could consider their suppliers’ environmental performance as criteria
for whether they are good or bad business partners. If negative CEPI of small suppliers
generate negative attitude among buyer corporations toward small suppliers, suppliers
will have incentives to reduce their pollution in terms of their long-term profit
maximization. However, it is inferred that the efficacy of CEPI disclosure for the small
suppliers would not be strong at this time because only small numbers of large buyers are

starting to consider the environmental performance of their suppliers (Hoffman, 2000a).

227

Influence of PID on Long Term Business Strategy

This study demonstrates that consumers change their attitude, perception of
corporate credibility and purchase intention in a negative direction against polluting
corporations and in a positive direction for non-polluting corporations. Although this
study is based on a sample of Korean university students, it is believed that the results
might also hold true for the general public.

The changed attitude, corporate credibility and purchase intention of consumers
could influence corporations to adopt environmentally safe practices and to change their
traditional business norms and strategies, which are based on profit maximizing and
unlimited resource use. The potential of decrease of stock value, market shares and the
potential of increase of managemental business costs for polluting corporations”, which
are generated by negative CEPI disclosure, forms a new business environment. That new
environment will definitely disadvantage the polluting corporations and advantage the
non-polluting corporations. For polluting corporations to survive in a new business
environment, they need to adopt environmentally safe practices and to Change their
traditional business norms and strategies. Thus, PID could contribute to constructing a
new business environment disadvantaging polluting corporations and to pressing them to
adopt environmentally friendly business norms and strategies.

This study acknowledges that the contribution of CEPI disclosure to the. new
business environment would not be large in the short term. Transitioning from the
traditional business environment to the new environment, disadvantaging polluting

corporations, is already beginning in the real world (Hoffinan, 2000a). It is also inferred

 

2' Refer to Ch. 111, p. 65-67.

228

that the transition would proceed very slowly because change in individual attitude takes
time. The time required to retain environmental information varies depending individual,
but it tend to be temporary. That means, effectiveness of CEPI disclosure could also be
temporary. Thus, CEPI disclosure would be only one event contributing to changing
polluting corporations’ traditional business norms and strategies toward more socially

sustainable business norms and strategies.

Policy Applications of Study Findings: Implementation of PID and Suggestions for
Enhancing Effectiveness of PID

CEPI appears to be a critical factor influencing consumers’ purchase decision in
our modern society when all other critical factors affecting purchase behavior (e. g. price
and quality of products, brand reputations and preferences, etc.) are the same. CEPI
disclosure would be effective for voluntary pollution reduction of corporations. Based on
this potential effectiveness of PID, this study strongly suggests that governments could
employ PID as a supplementary or complementary tool to CAC (command-and-control)
for pollution control tool.

There is, however, a high probability of opposition or resistance from industry to
government’s adoption of PID because polluting corporations have strong incentives not
to disclose their negative information of environmental performance. It is known that
industry or businesses traditionally lobbied to lower national environmental standards and
to reduce environmental law enforcement.

Evidence from this study can be used to deter anti-PID adoption activities by

industry. This evidence can be used as a scientific foundation for persuading and

229

empowering politicians, government officials and pollution policy makers or designers to
employ PID as a pollution control tool.

PID programs could be employed at diverse levels of government at the local or
small city level. Most PID programs were employed in the national level of government.
The Philippines launched EcoWatch in 1997, Mexico is developing PEPI (Public
Environmental Performance Indicators), and Colombia also started a public disclosure
program to complement its pollution charge system.22 Therefore, it is believed that PID
could also be employed for pollution control at the state level. Even lower level of
government such as local or small city could employ PID programs for local pollution
control because when the government is capable of generating credible and easily-
understood information, PH) programs would be persuasive and workable. Even though
local media systems or community level NGOs may not be capable of disseminating the
disclosed information, the local government can directly disseminate CEPI to the public.
Therefore, application of PID is not necessarily limited to national levels of government.
However, for determining at which level of government PID will work more cost-
effectively for pollution control, more scientific research is required.

In PID programs, corporations have a responsibility to provide true information
about their environmental performances or they need to review the accuracy of
government information about their environmental performance (i.e., checking the record
on their compliance with national environmental standards). Corporate participation in
PID (i.e., providing their environmental performance information to government or

reviewing government’s evaluation) will be done on a voluntary basis23 or it could be

 

22 The results from those PROPER-type prograrrrs have not appeared yet (Wheeler, 1999).
23 In the case of PROPER, information was collected on the semi-voluntary basis. BAPEDAL

230

enforced by law (e.g., Toxic Release Inventory in the US). Non-polluting corporations
have incentives for voluntarily participating in PID program because CEPI disclosure
will benefit to them. The advantages gained by non-polluting corporations from a PID
program could encourage polluting corporations to participate. But, if it is apparent that
polluting corporations do not want to participate in a PID program, enforcement of a
corporation’s participation in PID may be required in some cases.

To force corporations to provide information about their pollution emission to the
public or governments, legislation may be necessary. In the case of TRI (Toxic Release
Inventory), the US. Congress passed in 1986 the Emergency Planning and Community
Right to Know Act (EPCRA) that includes the TRI designed to provide information to
the public on the release of toxic substances into the environment. EPCRA required
manufacturing establishments with certain threshold sizes of chemical emissions to
disclose publicly the quantity and type of toxic chemicals released into the environment.
Just in case that government cannot obtain necessary data for CEPI disclosure and the
PID program due to strong resistance from industry, the enabling legislation for PID
program will be needed.

Primary roles of Government in PID program are 1) Checking the truth of the
information provided by corporations and 2) transferring raw CEPI into easily readable
information like PROPER’s color-grading system, and 3) effectively disseminating the

CEPI. This study already strongly emphasized the need for maintaining high credibility

 

(Indonesia’s National Pollution Control Agency) surveyed initially 350 facilities participating in
PROKASIH, a semi-voluntary program for controlling the discharge of industrial pollution in waterways
initiated in 1989, formally labeled the Clean River Program. Through the survey, BAPEDAL judged that
176 facilities had sufficient data to be graded. BAPEDAL invited facilities not included in PROKASIH to
volunteer to be graded. As a result, the initial number of participants in the grading system was 187
facilities.

231

of environmental protection authorities and also advised the authorities to exercise a high
level of care to assure a high level of credible information, because the more credible is
CEPI is, the more effective PID is. Even though the environmental authorities retain a
high level of credibility, to enhance effectiveness of PID they need to make an effort to
enhance the credibility of CEPI disclosed.

Another important responsibility of government is to decide on an appropriate
corporation size. In other words, at what point will small corporations will be excluded
from CEPI disclosure. When all small corporations are included in CEPI disclosure, too
much information will confuse the individual consumers. Consumers are not capable of
memorizing a lot of information at one time. As a result, the effectiveness of information
disclosure could be severely weakened. Government’s capacity for dealing with
information is also limited. Thus, an appropriate corporation size for an effective PID
program will vary depending on social conditions because each society has a different
industrial context and governmental capacity. To solve this issue, more specific
investigations are needed in particular social contexts.

Government needs to make an effort to involve primary stakeholders of PID in
the conduct of PID programs. Internal stakeholders of PID program are government and
corporations. When government involves the related corporations from the beginning of
the process (i.e., before announcement of CEPI), a more effectiveness of PID can be
expected. Thus, it is recommended that government encourage corporate participation in
the whole process of PH) program development and implementation. In the case of
Indonesia, PROPER teams involved corporations in the process of collecting data and

assessing the accuracy and level of their environmental performance so that corporations

232

had enough time to think about the potential impacts of CEPI disclosure on their
businesses. As a result, significant improvement in compliance status was achieved
before the announcement of CEPI.

Important external stakeholders are NGOs, communities and mass media (i.e., the
press). In the case of PROPER, in order to ensure that press reports on the rating system
were accurate, BAPEDAL arranged for officers involved in the project to visit the offices
of major local newspapers, to explain the system, and to demonstrate the computer
program. Afsah, Laplante and Wheeler (1997) asserted that PROPER embodied a power
strategy for strengthening relationships with NGOs, communities, and the press. These
three groups were empowered to pressure polluting corporations to reduce pollution
following release of corporate environmental performance information (CEPI). NGOs
and communities can review data of corporate pollution, monitor progress of the PID
program and evaluate the success of PID program. Representatives of government,
corporations, NGOs, communities and the press can work together in an advisory board
of PID program. Thus, for added success of the PID program, it is suggested that
government make these stakeholders cooperators in the PID program and work together
with them throughout the whole process of PID program.

It is inferred that some polluting corporations will not reduce their pollution even
though disclosure of their CEPI damages their businesses. These corporations are likely
to be small, less known and reputationally insensitive. However, this study provides
evidence that PID may also be effective for pollution reduction of the reputationallly
insensitive corporations. That is, according to the finding about the relationship between

the degree of CEPI disclosure and corporate familiarity, PID could be workable for even

233

these small corporations indifferent to their reputations. When CEPI disclosure is
periodically (e. g., annually) conducted, negative impacts on polluting corporations’
business will be accumulated. Then, even corporations that do not reduce their pollution
emission in the first or second years of CEPI disclosure, will likely reduce their amount
of pollution in later years.

Corporate compliance rate with national environmental standards can be used to
evaluate success of the PID program. Government can measure corporate compliance
rates before and alter CEPI disclosure. For example, PROPER teams reported increases
of color-grades of corporations after disclosing the names and grades of facilities as an
evidence for success of PROPER.

Cost-savings can also be expected in the administration of a PID program,
compared with CAC (command-and-control). The supporting data or evidence for cost-
effectiveness of PID is not available. However, it is inferred that PID would be a less
expensive and more easily implemented pollution control approach, compared to both
CAC and MBI (market-based instruments).24 Costs for designing, implementing, and
monitoring PID would be smaller than CAC and MBI. Data for CEPI is mostly already
available from the database of pollution regulatory authority. The cost for information
collection, aggregation and dissemination is also decreasing (Tietenberg & Wheeler,
1998). All social drivers (e. g., mass media, grassroots organizations, NGOs, financial
markets, products markets, courts, etc.) are also freely available for PID implementation.

In order to develop more sophisticated targeting strategies, PID program
designers need to collect data about, or know the extent of environmental attitude of,

diverse consumer groups. This study found that a positive relationship exists between the

 

2’ Refer to Advantages of PID in Ch. II, p. 37.

234

extent of effectiveness of CEPI disclosure and environmental attitude of consumers.
Korean undergraduate students show a slightly high degree of environmental attitude (M
= 3.82, SD = .43 in the 5-point scale). Different groups would have different degrees of
environmental attitude. According to the positive correlation related to environmental
attitudes, effectiveness of CEPI disclosure would not be large for groups with weak
environmental attitudes, but it would be high for groups with strong degrees of
environmental attitudes. From this perspective, it is strongly suggested that governments
make a long-term plans to improve the public’s environmental attitude, if it is practically
and politically possible.

This study also suggests some additional as ways for enhancing the effectiveness
of PID program. Government need to conduct more aggressive information dissemination
to the public such as 1) performing public advertisement about PID programs on the
national level of newspaper, TV media and cable channels, using government funds, and
2) developing a more readable and humorous format. Rather than only relying on mass
media or social NGO’s information dissemination, government needs to develop its own
system for information dissemination, especially in situations where systems of mass
media and NGOs are not well established for dissemination of information.

And finally, this study suggests that PID program designers develop ideas for
sustaining the effectiveness of CEPI disclosure for a longer period of time. Several
anecdotal cases showed that public environmental activities such as consumers’
avoidance of buying products from polluting corporation are mostly temporary.
Therefore, as an idea for sustaining the efficacy of CEPI disclosure for an extended

period of time, this study suggests that government must disclose CEPI periodically,

235

perhaps annually, because consumers’ repeated exposure CEPI could help them to
remember the information. Also, an easily readable and humorous information format

(e.g., graphical or visual symbols, and images or color grades like PROPER’s) will help

lengthen the public’s memory.

236

CHAPTER VIII

SUMMARY, CONCLUSIONS AND LIMITATIONS

Summary

The primary goal of this study is to investigate the efficacy of PID (Public
Information Disclosure for Pollution Control), especially in Korea. Effectiveness of PID
is investigated through testing a theory of PID that disclosure of CEPI (Corporate
Environmental Performance Information) will influence consumers to change their
purchase behavior in the negative direction for polluting corporations and in the positive
direction for non-polluting corporations. It is based on the assumption that purchase
behavior change generates market incentive or pressure on corporations to reduce
pollution voluntarily.

PID can be an information-based strategy to encourage industries to limit their
pollution emissions. It is an Information Oriented Approach to pollution control. The idea
of Information Oriented Approach to pollution control is very simple. That is, an
environmental regulatory authority such as the US. EPA (Environmental Protection
Agency) or KME (Korean Ministry of Environment) provides reliable and easily
understood information about corporate environmental performance to the public, the
social actors driving corporate environmental practices. Then, social actors such as
environmental or social NGOs, the media, the courts, community or grassroots
organizations, and the product or the financial market begins to press corporations to
adopt environmentally safe practices and to Change their traditional business strategies of

maximizing profit without regard to environmental impact.

237

Pre and post experimental tests were conducted to investigate the first claim of
PID that consumers respond to CEPI and change their purchase behavior regarding
products of polluting and non-polluting corporations. This study strongly confirmed H1
that CEPI disclosures Change consumer attitude toward corporations (AC) positively for
non-polluting corporations and negatively for polluting corporations, H 2 that CEPI
disclosures change the consumers’ perception of corporate credibility (CC) positively for
non-polluting corporations and negatively for polluting corporations, H 3 that CEPI
disclosures increase consumer purchase intention (PI) for the products of non-polluting
corporations and decrease consumer purchase intention for the products of polluting
corporations, H 4 that customers who have strong environmental attitudes about pollution
(EAP) will change the three dependent variables of AC, CC, and PI more than customers
who have weak EAP], and H 6 that The greater the credibility of CEPI (CCI), the greater
the effect of the CEPI rating on the three dependent variables. Data did not clearly
confirmed H 5 that consumers with high familiarity about a corporation (CFC) Change

AC, CC and PI less than one with low familiarity.

Conclusion

This study found that CEPI disclosure changes 1) consumers’ attitude toward the
corporation, 2) perception of credibility of the corporation and 3) purchase intention
toward products of the corporation in the positive direction for non-polluting corporations

and in the negative directions for polluting corporations. These findings show that

 

' In this research, environmental attitude specifies the environmental attitude toward pollution (EAP)
because the research is related to industry pollution. Thus, high EAP is defined as the status of a strong
concern, awareness, and belief about pollution. Low EAP is defined as the status of a weak concern,
awareness and belief about pollution.

238

modern consumers tend to respond to CEPI, and consequently, change their p

behavior. Thus, CEPI can become a critical factor that influences the consume

 

purchasing decision on the condition that all other critical factors for making p‘irrcrrase'”" fl
decisions - price and quality of products, their brand reputations and preferences - are
constant.

This study also found 4) a positive correlation between consumer’s environmental
attitude about pollution and the effect of CEPI and 5) a strong positive correlation
between information credibility and the effect of CEPI. These findings mean that
consumers’ environmental attitudes and CEPI credibility are critical elements that
influence the degree of the effectiveness of PID.

This study did not find 6) a correlation between familiarity with corporation and
the effect of CEPI. Little relationship between corporate familiarity and purchase
behavior change means that the target of PID program is not necessary to limit to famous
and large corporations because PID could be workable for even less known brands or
corporations.

Based on an assertion that purchase behavior can be well predicted from attitude,
credibility, and purchase intention, this study predicts that there is a high probability that
CEPI disclosure influences consumers to change their purchase behavior in the negative
direction for polluting corporations and in the positive direction for non-polluting
corporations.

CEPI changes consumers’ purchase behavior negatively against the polluting
corporations, and thus causes a decline in the demand for their products. Decreased

product demand results in a loss of market share and, consequently, a reduction of profit.

239

Corporations are profit-maximizing entities that make decisions based on self-interest.
Corporate managers’ concern with corporate pollution is deprived in part from concern
about profits and costs. Thus, a reduction in profit by CEPI disclosure becomes a market
concern. In other words, the loss of market motivates corporations to reduce their total
volume of pollution in order to maximize their total profits. As a result, consumers’
purchase behavior change generates market incentive or pressure for corporations to
voluntarily reduce pollution. Based on these findings about purchase behavior change by
CEPI disclosure, this study predicts a high probability that Public Information Disclosure
(PID) can be an eflective approach for pollution control. Although this study is based on
a survey of Korean university student, it is believed that the results might also hold true

for the general public.

Limitation and Further Research

This study reveals important findings regarding the influence of Public
Information Disclosure (PID) on consumer purchasing behavior and its use for pollution
control. However, there are several limitations that warrant further discussion and

implications for future studies.

Policy Implications of Public Information Disclosure

In order to implement a PID (Public Information Disclosure for pollution control)
policy or program, societies or countries need to examine the practical effectiveness of
the PID approach. The purpose of this study was limited to testing a theory that

disclosure of CEPI (Corporate Environmental Performance Information) influences

240

consumers’ purchase behavior and that such purchase behavioral change could influence
corporate polluting behavior. This study proved the causal relationship between CEPI
disclosure and purchase behavior change and investigated theoretical effectiveness of
PID.

However, this study did not investigate the actual effectiveness of PID policy and
it did not focus on policy applications of PID. That is, this study is not making the
argument as to how well PID would operate in Korea. Even though the theory of PID is
valid, this does not ensure PID will be effective in the real world. The effectiveness of
PID will vary in different societies because each nation has a distinctive social
environment affecting the use and effectiveness of PID. In order to predict the actual
effectiveness of PID applications, additional survey research would be required in

particular social and cultural settings.

More Tests for More Reliable PID Theory

The PID theory test of this study is limited to Korean society, two Korean
universities, four Korean corporations and their products. The sensitivity of consumers to
CEPI depends on 1) consumers’ environmental concern with or attitude toward
environmental conditions, 2) the credibility and readability of CEPI, 3) relevant
corporations and 4) relevant products. The effectiveness of PID related to the product
market depends on consumers’ sensitivity to CEPI and social environment surrounding
implementing PID policy. A wide variety of consumer groups and each consumer group
have different consumption patterns, preferences and different levels of environmental

attitude and hence different sensitivities to CEPI. They will, thus, show different

241

intensities of response to positive and negative CEPI. Thus, if an experiment were
conducted under the same experimental conditions under which this study was conducted
but with different Korean consumer groups or with consumer groups of different
countries, the results might be different from the results of this study.

Thus, in order to gain more reliable knowledge about PID theory, an additional
experimental testing for PID theory is required under the conditions of diverse groups of

consumers, diverse corporations, and diverse products, and in diverse societies.

Study for Moderator of Effectiveness of PID

This study investigated three moderator variables: environmental attitude toward
pollution, familiarity toward corporation, and information credibility. The moderator
variable is a special case of the masking variables that are unknown variable affecting the
dependent variables. It is defined as a variable that intervenes between the independent
and dependent variable and influences the intensity of effect in the dependent variable.
Knowledge of moderator variables for CEPI and purchase behavior is practically
valuable for pollution policy designers. The primary goal of pollution policy designers is
to develop a cost-effective pollution control tool. Studies for moderator variables for
CEPI and purchase behavior will provide ideas and knowledge for pollution policy
makers for more effective and more inexpensive pollution control programs of PID.
Therefore, additional study about moderator variables for effectiveness of PID is

recommended.

242

APPENDICES

243

Appendix A. QUESTIONNAIRE FOR EXPERIMENTAL TESTS

Pre-Test Questionnaire

* Please write down your student identification number here (

* Circle one of the numbers to describe your thoughts or feelings about the following

corporations:

1. Indicate your thoughts or feelings about Samsung Electronics Cellular Phone Co.

Overall low-
Quality products

Not at all good

At manufacturing 1

Not at all
Dependable

Low respect for
The customer

Overall hi gh-
quality products

Very good at
manufacturing

Very
Dependable

High respect for
the customer

2. Indicate your thoughts or feelings about Hyundai Electronics Cellular Phone Co.

Overall low-
Quality products

Not at all good

At manufacturing 1

Not at all
Dependable

Low respect for
The customer

244

Overall high-
quality products

Very good at
manufacturing

Very
Dependable

High respect for
the customer

3. Indicate your thoughts or feelings about Nongsim Ramyeun Co.

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products
Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable 1 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

4. Indicate your thoughts or feelings about Binggrae Ramyeun Co.

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products
Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable 1 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

5. Indicate your thoughts or feelings about Samsung Electronics Cellular Phone Co.

Unreputable 1 2 3 4 5 6 7 Reputable
Untrustworthy 1 2 3 4 S 6 7 Trustworthy
Negative 1 2 3 4 5 6 7 Positive
Dislike 1 2 3 4 5 6 7 Like

245

6. Indicate your thoughts or feelings about Hyundai Electronics Cellular Phone Co.

Unreputable

Untrustworthy 1

Negative

Dislike

7. Indicate your thoughts or feelings about Nongsim Ramyeun Co.

Unreputable

Untrustworthy 1

Negative

Dislike

8. Indicate your thoughts or feelings about Binggrae Ramyeun Co.

Unreputable

Untrustworthy 1

Negative

Dislike

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

246

5

5

5

5

5

5

6

6

6

6

6

6

7

7

Reputable
Trustworthy
Positive

Like

Reputable
Trustworthy
Positive

Like

Reputable
Trustworthy
Positive

Like

* The following 4 corporations will distribute new products in the market. What is the
likelihood of your purchasing of those products?

1. Indicate the likelihood of your purchasing of a new cellular phone of Samsung
Electronics

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable l 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

2. Indicate the likelihood of your purchasing of a new cellular phone of Hyundai
Electronics.

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable l 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

3. Indicate the likelihood of your purchasing of a new instant noodle of Nongsim.

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable 1 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

4. Indicate the likelihood of your purchasing of a new instant noodle of Binggrae.

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable l 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

247

* As you read the following statements, please mark which ones apply to you.

1. Rate your knowledge of Samsung Electronics Cellular Phone Co., as compared to the
average undergraduate student.

I AM LESS I AM MORE
Knowledgeable l 2 3 4 5 6 7 knowledgeable

2. Circle one of the numbers below to describe your familiarity with Samsung
Electronics.

Not at all Extremely
Familiar 1 2 3 4 5 6 7 Familiar

3. Rate your knowledge of Hyundai Electronics Cellular Phone Co., as compared to the
average undergraduate student.

I AM LESS I AM MORE
Knowledgeable 1 2 3 4 5 6 7 knowledgeable

4. Circle one of the numbers below to describe your familiarity with Hyundai
Electronics Cellular Phone Co.

Not at all Extremely
Familiar 1 2 3 4 5 6 7 Familiar

5. Rate your knowledge of Nongsim Ramyeun Co., as compared to the average
undergraduate student.

I AM LESS I AM MORE
Knowledgeable l 2 3 4 5 6 7 knowledgeable

6. Circle one of the numbers below to describe your familiarity with Nongsim Ramyeun
Co.

Not at all Extremely
Familiar 1 2 3 4 5 6 7 Familiar

7. Rate your knowledge of Binggrae Ramyeun Co., as compared to the average
undergraduate student.

I AM LESS I AM MORE
Knowledgeable 1 2 3 4 5 6 7 knowledgeable

248

 

 

 

8. Circle one of the numbers below to describe your familiarig with Binggrae Ramyeun
Co.

Not at all Extremely
Familiar 1 2 3 4 5 6 7 Familiar

 

* As you read the following statements, please mark which ones apply to you.

1. I would be willing to walk a short distance rather than to ride a bus or a taxi in
order to reduce air pollution.

(l) Strongly disagree (2) Disagree (3) Neutral (4) Agree (5) Strongly Agree

1

2. I usually read newspaper articles or listen to TV news concerning our
environment. (Indicate your approximate frequency)

 

 

 

 

Not at all 1 2 3 4 5 Often

3. I do not purchase products that are known to cause pollution.

(1) Strongly disagree (2) Disagree (3) Neutral (4) Agree (5) Strongly Agree
4. Companies that produce excessive pollution should be strongly punished.

(1) Strongly disagree (2) Disagree (3) Neutral (4) Agree (5) Strongly Agree

5. Indicate your thoughts or feelings about the level of the pollution problem in
Korea.

 

No problem 1 ------------- 2 3 4 -------------- 5 Severe problem

 

6. Environmental problems should be given greater priority than economic growth.

(1) Strongly disagree (2) Disagree (3) Neutral (4) Agree (5) Strongly Agree

249

7. Government should do more to protect environment such as stronger
environmental regulation and law enforcement.

(l) Strongly disagree (2) Disagree (3) Neutral (4) Agree (5) Strongly Agree
8. If you were able to contribute 50,000 won to good causes, how much would you
donate to help the environment?

(1) 0 - 10,000 won (2) 10,000 —— 20,000 won (3) 20,000 — 30,000 won (4)
30,000 — 40,000 (5) 40,000 - 50,000 won

* These are the last questions.
1 What is your gender? Female Male

2. How old are you?

 

3. What is your major?

 

4. Where did you grow up?
(1) big city (2) middle or small city (3) Rural (4) others ( )

5. What is your family income per month? ( )

* Please write down the name of your university and date.

( ) University (mm / dd: )

250

Post-Test Questionnaire

* Please write down the type (A or B) of the print material that you received just before
(Type: )

* Please write down your student identification number here ( )

* Circle one of the numbers to describe your thoughts or feelings about the following
corporations:

1. Indicate your thoughts or feelings about Samsung Electronics Cellular Phone Co.

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products
Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable 1 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

2. Indicate your thoughts or feelings about Hyundai Electronics Cellular Phone Co.

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products
Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable l 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

3. Indicate your thoughts or feelings about Nongsim Ramyeun Co.

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products

251

Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable l 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

 

4. Indicate your thoughts or feelings about Binggrae Ramyeun Co.

 

Overall low- Overall high-
Quality products 1 2 3 4 5 6 7 quality products
Not at all good Very good at

At manufacturing 1 2 3 4 5 6 7 manufacturing
Not at all Very
Dependable l 2 3 4 5 6 7 Dependable
Low respect for High respect for
The customer 1 2 3 4 5 6 7 the customer

5. Indicate your thoughts or feelings about Samsung Electronics Cellular Phone Co.

Unreputable 1 2 3 4 5 6 7 Reputable
Untrustworthy 1 2 3 4 5 6 7 Trustworthy
Negative 1 2 3 4 5 6 7 Positive
Dislike l 2 3 4 5 6 7 Like

6. Indicate your thoughts or feelings about Hyundai Electronics Cellular Phone Co.

Unreputable 1 2 3 4 5 6 7 Reputable
Untrustworthy 1 2 3 4 5 6 7 Trustworthy
Negative 1 2 3 4 5 6 7 Positive
Dislike 1 2 3 4 5 6 7 Like

252

 

7. Indicate your thoughts or feelings about Nongsim Ramyeun Co.

 

Unreputable 1 2 3 4 5 6 7 Reputable

Untrustworthy l 2 3 4 5 6 7 Trustworthy

Negative 1 2 3 4 5 6 7 Positive

Dislike 1 2 3 4 5 6 7 Like F

8. Indicate your thoughts or feelings about Binggrae Ramyeun Co.

 

Unreputable 1 2 3 4 S 6 7 Reputable
Untrustworthy 1 2 3 4 5 6 7 Trustworthy
Negative 1 2 3 4 5 ' 6 7 Positive
Dislike 1 2 3 4 5 6 7 Like

* The following 4 corporations will distribute new products in the market. What is the
likelihood of you purchasing those products:

1. Indicate the likelihood of you purchasing a new cellular phone of Samsung

Electronics

Very unlikely l 2 3 4 5 6 7 Very likely
Improbable 1 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

2. Indicate the likelihood of you purchasing a new cellular phone of Hyundai

Electronics.

Very unlikely l 2 3 4 5 6 7 Very likely
Improbable 1 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

253

3. Indicate the likelihood of you purchasing a the new instant noodle of Nongsim.

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable 1 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

4. Indicate the likelihood of you purchasing a new instant noodle of Binggrae.

Very unlikely 1 2 3 4 5 6 7 Very likely
Improbable 1 2 3 4 5 6 7 Probable
Impossible 1 2 3 4 5 6 7 Possible

* Rate the extent to which the following corporations comply to the national pollution
standards.

1. Samsung Electronics Cellular Phone Co. has complied with national environmental
standards

Not at all 1 2 3 4 5 6 7 Very well

2. The environmental performance of Samsung Electronics Cellular Phone Co. has
created on the health of human being and nature

Very severe l 2 3 4 5 6 7 Very Positive
Adverse effects Effects
3. Hyundai Electronics Cellular Phone Co. has complied with national environmental

standards

Not at all 1 2 3 4 5 6 7 Very well

4. The environmental performance of Hyundai Electronics Cellular Phone Co. has
created on the health of human being and nature

Very severe 1 2 3 4 5 6 ‘7 Very Positive
Adverse effects Effects

254

 

5. Nongsim Ramyeun Co. has complied with national environmental standards
Not at all 1 2 3 4 5 6 7 Very well

6. The environmental performance of Samsung Nongsim Ramyeun Co. has created on
the health of human being and nature

Very severe l 2 3 4 5 6 7 Very Positive
Adverse effects Effects

7. Binggrae Ramyeun Co. has complied with national environmental standards
Not at all 1 2 3 4 5 6 7 Very well

8. The environmental performance of Binggrae Ramyeun Co. has created on the health
of human being and nature

Very severe 1 2 3 4 5 6 7 Very Positive
Adverse effects Effects

* As you read the following statements, please mark which ones apply to you.

1. Would you rate the credibility of environmental information provided by the Korean
Ministry of Environment?

Very incredible I 2 3 4 5 6 7 Very credible
2. Would you rate the expertise of environmental information provided by the Korean
Ministry of Environment?

No expertise 1 2 3 4 5 6 7 High expertise

* Please write down the name of your university and date.

( ) University (mm / dd: )

255

 

 

Appendix B. EXPERIMENTAL STIMULI: TYPE A AND B1

Stimuli Type A
Type A

...il ‘Hil‘ “Hf". ‘ "‘iii'llliiim‘ IIIIlelI Im‘IIK'i NI IiiIiIii» Ill ii“
In

”I I" n- “'1. :IIIIL [Y , "III [I ix“ RIM i“ Iii\ “IIIIMI IIIii IilhI I, i i
Iii“ mi”. ”1 lll'fluli I ll Ii "I" II“. " I ‘ Iii‘ “if“
iill‘l‘tliI . 1‘ illi. iii” ‘ii'l‘l \Ui ' “Ila “iii“ ii will i, .i

I “\Hll‘ic IIll IIIIIHIIIII MIWIII‘umIIL. IIII

‘d—ZF’I'

I
IV‘IMI’IW ilw"i‘i""{ilii"' LII V
lliilhiii‘l”“”ul"‘ “““itli‘l‘”.Ni)liiill l H ....'.'.t.
iIIIIIII‘ ‘.l‘i llinl ill1 M iii“ ll'mnlllii1 1'13““
. n tiill‘I‘W ”I‘ll“, . IIIIIIIIH in “Hut"; ‘WIHII

il III I1“”“Iii" II“ III‘ .

.1 Ill _
his»“WWII“lit“‘m'llll‘lull iiiiiill MI“ I “m" 1iiIluuliii ll" ” .
“ii iiiiiil “u” iiul i"" an} “iii ‘ t'ln I‘I‘IIIIIIi "
II III W II militia... ”ugh“:

‘ It

Ililli IIIII “II“ I “hii'ii‘i

. l 5 . ',
"I“. HInil WI 3 I
III 3‘

IIIII ;‘ i"

l
"I “.1! I I
Iliiifiiiim IiIiiiIlll

,,
1.3 :3
.

.‘I

 

 

' This English version of experimental stimuli A and B is translated from Korean version of stimuli Type A
and B, which was used in the Korean experiment.

256

(Please remember that what you read is Type A.)

The Environmental Policy Bureau of The Korean MinistrfxI of Environment (KME) has
recently employed a “Color-Grade System for Environmental Evaluation.” The grading
system is to rate the environmental performance of Korean industrial corporations based
on a 3 color-grading system. The 3 colors used for grading the level of corporations’

compliance to the national pollution standards and their meanings are as follows:

Red (severely Polluting Corporations): Factories or business

corporations that exert no environmental management efforts and whose business
activities pollute the nation’s air, water and soils, thus cause serious damages to the

nature and human beings, and environmental degradation.

 

Blue (Complying Corporations): Factories or business activities that

exert minimum efforts to meet the national standards of air, water, solid and toxic wastes.

Green (Green Corporations): Factories or business activities that use best

 

available cleaning technology; promote zero discharge of pollutants; and conduct

environmental impact management efforts with very satisfactory results.

257

 

* The Department of Environmental Evaluation in the Environmental Policy Bureau of
KME color graded the following four corporations (on January, 2001).

1.

POO

Cellular Phone Factory of Hyundai Electronics: Red (severely

polluting corporation)

corporation)

. Cellular Phone Factory of Samsung Electronics: Green (green

. Ramyeun Factory of Binggrae: Red (severely polluting corporation)

Ramyeun Factory of Nongsim: Green (green corporation)

 

Environmental Evaluation:
RED (Severely Polluting
Corporation)

Environmental Evaluation:

GREEN (Green Corporation)

 

Hyundai Electronics Cellular
Phone Co.
Binggrae Ramyeun Co.

Samsung Electronics Cellular
Phone Co.

Nongsim Ramyeun Co.

 

 

Hyundai Electronics Cellular
Phone Co. and Binggrae Ramyeun Co.
have increasingly emitted contaminated
water, polluted air, toxic chemicals and
solid wastes for the past 3 years.

When compared to 3 years ago,
there is a significant increase of the
probability of disease outbreak among
the residents living near the two
corporations. There are greater than
three times of increase of the number of
residents inflicted with various cancers,
such as lung or stomach cancers,
greater than 10 times increase in
respiratory diseases. greater than 7
times increase in skin diseases, and
more than 12 times of increase in other
unidentified diseases. It has been
proved that the main cause for the
increase of disease occurrence is the
pollution by Hyundai Electronics
Cellular Phone Co. and Binggrae
Ramyeun Co.

 

Samsung Electronics Cellular
Phone Co. and Nongsim Ramyeun Co.
have used the best available cleaning
technology; promoted zero discharge of
pollutants, and conducted
environmental impact management
efforts with very satisfactory results.

These corporations have been
disseminating their new environmental
technology and knowledge of corporate
environmental management. They have
also supported a variety of programs
and events promoting environmental
education and protection as well as the
recovery of ecological system in their
community.

 

(Data Source: Department of Environmental Evaluation in the Environmental Policy Bureau of KME,
1998-2000 years)

258

 

* Dgpartment of Environmental Evaluation in the Environmental Policy Bureau of KME

released a “Corporate Environmental Index,” an environmental corporate compliance
index to the national environmental standards The Corporate Environmental Index for
four corporations from 1998 to 2000 is as follows:

1. Hyundai Electronics Cellular Phone Co.: -159

2. Samsung Electronics Cellular Phone Co.: 109

3. Binggrae Ramyeun Co.: -l36

4. Nongsim Ramyeun Co.: 125

* Corporate Environmental Index (CE!)

0 CEI “O”: meets the national environmental standards of air, water, solid and
toxic waste pollution.

o CEI “ - 100”: violates the national environmental standards by emitting twice the
approved amount by air, water, solid and toxic waste pollution standards

0 CEI “ 100”: violates the national environmental standard by emitting a half of the
approved amount by air, water, solid and toxic waste pollution standards

Corporate Environmental Index

 

‘ISamsung
o - . . l' .
INonQSIm

Samsung Nongsim I l

 

Environmental Compliance to National
Standard

 

 

 

.50 -
-100 ~
-150 s -136
-159
-200
Company

(" For the purpose of microfilming this paper, the labels "Hyundai and "Binggrae" were removed from the
chart above so that the chart is different from original one used in the experiment.)

259

Stimuli Type B
Type B

. .1 ~‘ 1‘
“(I . ”.1 Jill‘

'* WW" W W. W W
" h "n ““ Killiullli " ”will W: Li? 1.“ .

”l"? (1‘01i llmfiii‘ I l“ ’i
«mm. limb-m. Nikki [Nil “film‘” It} in :21» .i “1"“

fi‘iifl ‘9'” M ii...” “film;
ll ii'iilv ll'
\ MM
11‘le .nl,

ll
M
l i ”ll“ “(Mia All“:
Mil ii «iii, iii "M
‘30.“: Will “\H'il‘fl‘im
(11“ Will \(u 1“ mm

‘ as
l
nil Hll“"l 1ill

1 “wimuuw

)1 l i I i i l “ 'ilu.
{:31qu ”Hi“ (Hill "’5 anqfl‘ gin“ ils‘ ‘S‘ilil lgil‘ll‘H :Iiiiini)“ “W. “in" I!“ l' ' \ 1&8!
I 1111“”, mill ll lil‘" “"
llL W i” ”\ 1“”
ll

till“ “i M “ii“ ‘ l (“aim“) (“I‘m

lllliim Mun-m. "mm.”-

.ll"

1‘, ill ”iii“ 1 1‘
ll\ WW 1% l)“ l
\ \i “ii hm l . . [L Infill " l WW 1.,

l
(”M s. " i
.. , . . . 1i .. '-s‘::.‘.ltl‘13-.....-I'f...
it""iglil‘liinlilqiil‘l 111‘). 1 - l
lWl'K’lW} Willi ' ”2'1
l .

I” ”l
“1:5“ “'3:

I‘l‘
r‘lllu'

zilllllll-

IWWI "sill!

 

260

(Please remember that what you read is Type B.)

The Environmental Policy Bureau of The Korean Ministry of Environment (KME) has
recently employed a “Color-Grade System for Environmental Evaluation.” The grading
system is to rate the environmental performance of Korean industrial corporations based
on a 3 color-grading system. The 3 colors used for grading the level of corporations’

compliance to the national pollution standards and their meanings are as follows:

Red (severely Polluting Corporations): Factories or business

 

corporations that exert no environmental management efforts and whose business
activities pollute the nation’s air, water and soils, thus cause serious damages to the

nature and human beings, and environmental degradation.

 

Blue (complying Corporations): Factories or business activities that

exert minimum efforts to meet the national standards of air, water, solid and toxic wastes.

Green (Green Corporations): Factories or business activities that use best

 

available cleaning technology; promote zero discharge of pollutants; and conduct

environmental impact management efforts with very satisfactory results.

261

* The Department of Environmental Evaluation in the Environmental Policy Bureau of
KME color graded the following four corporations (on January, 2001).

5. Cellular Phone Factory of Samsung Electronics: Red (severely
polluting corporation)

6. Cellular Phone Factory of Hyundai Electronics: Green (green
corporation)

7. Ramyeun Factory of Nongsim: Red (severely polluting corporation)

8. Ramyeun Factory of Binggrae: Green (green corporation)

 

 

Environmental Evaluation: Environmental Evaluation:
RED (Severely Polluting GREEN (Green Corporation)
Corporation)

Samsung Electronics Cellular Hyundai Electronics Cellular
Phone Co. Phone Co.

Nongsim Ramyeun Co. Binggrae Ramyeun Co.

 

Samsung Electronics Cellular Hyundai Electronics Cellular
Phone Co. and Nongsim Ramyeun Co. Phone Co. and Binggrae Ramyeun Co.
have increasingly emitted contaminated have used the best available cleaning
water, polluted air, toxic chemicals and technology; promoted zero discharge of

solid wastes for the past 3 years. pollutants, and conducted
environmental impact management
When compared to 3 years ago, efforts with very satisfactory results.
there is a significant increase of the
probability of disease outbreak among These corporations have been
the residents living near the two disseminating their new environmental
corporations. There are greater than technology and knowledge of corporate

three times of increase of the number of environmental management. They have
residents inflicted with various cancers, also supported a variety of programs

such as lung or stomach cancers, and events promoting environmental
greater than 10 times increase in education and protection as well as the
respiratory diseases, greater than 7 recovery of ecological system in their
times increase in skin diseases, and community.

more than 12 times of increase in other
unidentified diseases. it has been
proved that the main cause for the
increase of disease occurrence is the
pollution by Samsung Electronics
Cellular Phone Co. and Nongsim
Ramyeun Co.

 

 

 

 

(Data Source: Department of Environmental Evaluation in the Environmental Policy Bureau of KME,
1998-2000 years)

262

* Department of Environmental Evmuon in the EnvironmentaLPolicv Bureau of KME
released a “Corporate Environmental Index,” an environmental corporate compliance
index to the national environmental standards The Corporate Environmental Index for
four corporations from 1998 to 2000 is as follows:

5. Samsung Electronics Cellular Phone Co.: -159
6. Hyundai Electronics Cellular Phone Co.: 109
7. Nongsim Ramyeun Co.: -136
8. Binggrae Ramyeun Co.: 125
* Corporgte Environmental Index (CEI)
o CEI “O”: meets the national environmental standards of air, water, solid and
toxic waste pollution.

o CEI “ - 100”: violates the national environmental standards by emitting twice the
approved amount by air, water, solid and toxic waste pollution standards

- CEI “ 100”: violates the national environmental standard by emitting a half of the
approved amount by air, water, solid and toxic waste pollution standards

Corporate Environmental Index

 

 

 

 

 

 

 

 

 

 

 

‘50 125
109

g 100 «
.9
g
9 50 e
8 lHyundai
.5 E o ‘ r I I
“g g Hyundai Binggrae I Binggrae
5 g -50 — I
‘3
g -100 ~
8
g -150 ~ -136
“J -159

-200

Company

(* For the purpose of microfilming this paper, the labels "Samsung" and "Nongsim" were removed from
the chart above so that the chart is different from original one used in the experiment.)

263

 

 

Appendix C

CORRELATION AMONG FAMILIARITY, ENVIRONMENTAL ATTITUDE
AND INFORMATION CREDIBILITY

 

 

 

 

 

 

 

 

 

SF HF NF BF ENATT INFOCR

SAMFAMI Pearson 1.000 .470 .321 .233 .146 .005
Correlation

Sig. (2- . .000 .000 .000 .011 .931
tailed)

N 304 304 304 304 304 304

HYUNFAMI Pearson .470 1.000 .150 .244 .048 .100
Correlation

Sig. (2- .000 . .009 .000 .406 .082
tailed)

N 304 304 304 304 304 304

NONGFAMI Pearson .321 .150 1.000 .323 .142 .100
Correlation

Sig. (2- .000 .009 . .000 .013 .081
tailed)

N 304 304 306 306 306 306

BINGFAMI Pearson .233 .244 .323 1.000 .040 .080
Correlation

Sig. (2- .000 .000 .000 . .481 .165
tailed)

N 304 304 306 306 306 306

ENVIATTI Pearson .146 .048 .142 .040 1.000 .022
Correlation

Sig. (2- .01 1 .406 .013 .481 . .698
tailed)

N 304 304 306 306 306 306

INFOCRED Pearson .005 .100 .100 .080 .022 1.000
Correlation

Sig. (2- .931 .082 .081 .165 .698

tailed)

N 304 304 306 306 306 306

 

- ** Correlation is significant at the 0.01 level (2-tailed).
- * Correlation is significant at the 0.05 level (2-tailed).
- SAMFAMI (SF) = Familiarity with Samsung

- HYUNFAMI (HF) = Familiarity with Hyundai

- NONGFAMI (NF) = Familiarity with Nongsim

- BINGFAMI (BF) = Familiarity with Binggrae

- ENVIATTI (ENATT) = Environmental Attitude

- INFOCRED = Information Credibility

- These correlations are generated by SPSS. These are not the corrected correlation.

264

 

Appendix D. Inference Probability2

The inference probability index only makes sense when you have a directional
hypothesis. For example, if you have hypothesized a positive correlation between
learning job training skills self esteem, then you can calculate and interpret the inference
probability index

 

Definition -------- > For a
predicted positive correlation, the
inference probability is the
probability of rho being positive, i.e.,
P(p > 0). Asking the value of the
inference probability is a very useful
question to ask, for if you
hypothesized and expected a positive
correlation, then the estimate of the
inference probability gives you an idea of how probable your research hypothesis is. For
example, suppose that IP was .79. That means that it is very likely that rho is positive,
the chance of your directional hypothesis coming true is estimated to be about 79%.

In fancy mathematical notion, the
inference probability is

P=Pm0<pSLw

 

Definition -------- > Reverse probability. The reverse probability is the
probability that the directional hypothesis is incorrect, i.e., that the population correlation,
rho, is not positive. The reverse probability is defined as

RP=1-IP

Figure 7-3 How is the IP calculated? You may be awed by
the fancy definition of the inference probability.
However, from a practical point of view,
calculating the [P is a cup of tea and a piece of
cake. At this point I am assuming that you really
know how to get areas under the normal curve. If
it is a bit fuzzy and you do not feel as confident
about calculating areas, go back to the homework
. problems in Chapter 2 (associated with my notes)
Sample correlat'on’ or Shavelson’s Chapter 5 to see how to do those
69-. r xy = -45 calculations.

 

 

0.0 .4

 

 

2 Appendix D. is directly quoted from an unpublished paper (i.e., notes of course Psychology 815) of Dr.
Ralph Levine who was a professor in the Department of Psychology and is a professor in the Department of
Resource Development at Michigan State University. The researcher appreciates his allowance for quoting
parts of his unpublished manuscript.

265

The inference probability is the probability that rho is positive, if your directional
hypothesis is positive or the inference probability is the probability that rho is negative
when positing a negative value of rho. From the point of view of areas under the curve,
most of the time the IP will be at least .50. To see this, look at the Figure (7-3)

Suppose your sample correlation were .45. In the figure you can see that .45 is
placed in the center of the normal distribution. We are looking for the probability of
being positive. The positive values start from a correlation of 0.0 and then go to the end
of the curve which would be a correlation of 1.0. The areas under the curve are the
probabilities. These areas are B, which extends from the .45 to 1.0 and A which goes
from r = .00 to r = .45. Area A is 50% of the distribution. All we need to calculate is the f
area under the curve from 0.0 to the sample correlation, which in this sample is .45. I

In order to use the Standard normal curve table in the back of Shavelson or to
convert the raw correlation .45 to a z score, we need two things, namely the mean (which
is .45) and a standard deviation. The stande deviation is estimated by using a formula
like Equation which is the formula for the standard error of the correlation coefficient.
Once the mean and the standard error are known, we can find the z score for a correlation is
of 0.0 and use the standard normal table to find the probability of area A. Then all we
have to do is to add the probabilities of areas A and B together to obtain an estimate of
the inference probability.

 

Concretely, for the case where we predict a positive correlation do the following
steps:

1. Place the normal curve around your sample correlation by letting your
sample correlation be the mean of the distribution.

2. Calculate the standard error
3. Calculate the z score that corresponds to a correlation of 0.0

4. Find the area under the curve from the z score found in step 3.
This Area A

5. Since the sample correlation is in the center of the distribution all
you have to do is add .50, the probability of falling in Area B, to
the probability of falling in Area A. That sum is the IP!

[Interpretation of the IP when the sample correlation is 0.0. If the sample correlation is

exactly 0.0, then Area B equals .50, and Area A = 0.0. The [P in this case would be
exactly 50%.]

266

a Mom :23

LL

are as saw are
H

a EOIXI.
*léloll Lila 0:131

h

.9.
c1

(

q

as

Jr

Appendix E

Korean Version of QUESTIONNAIRE FOR THE EXPERIMENTAL TEST
gill} £22194 5453 OIEHOII J(“MEI El§0l| %3

I:
11"

2

7|9=1§OII Elli-ii Ciel
allilli E3H -’i‘-1‘=M|Si. Oii‘iE-P-I ’EUZIOI 2

O

2. ennui Ir—aasm 3114191 area via-Alas] tile eels—24 slalom :21

5| 9.125 %ZIE’J 4 3.1% 9151MB Ell-4111.
—% ‘O'E TF5ll-’i‘-’él’\lS’.?

>:< 01213.3 83111ng 7:101 avg/~12.
1. aeaxl aaaafl $142} 3

Pre—Test Questionnaire

HENRI %E—J
EIAIOI Cl

 

>265)

.xrozfl
SEEN
WEN.
€an

are

.flfl
arm

267

2

1

flilfilll as
elxloltl
¢H|Xl§ 5.33561:—
§|Al7l OILICI

3. %49 749131419} 3363?} El%*l%l°ll Ellis} 043151-91 43210114 £74123- ‘o’E 3H

3643 419.?

%Hilkifll Ec'.=’.‘=l§|/ll.‘ll
aaae EHEE
Cilfiili HEQOIEI 1 2 3 4 5 6 7 Ellilli JE’EOIEI
113153 £61218
Q Bill %Elfil 1 2 3 4 5 6 7 Q F‘JECI
NEIBIXI %E Halal EIEI
§|AIOIEI 1 2 3 4 5 6 7 §IAIOICI

fittlleE Eéfil-t- _ fitHIXIS Eéfil-E
5WD) olLICI 1 2 3 4 5 6 7 §|A)0|L'.l-

4. ‘fl—‘lifil 3193314191 133% Pl%*l§l°ll EH3} alibi-94 4321014 521%‘0’3 311

2923134194 gnalalxlgl
aaae aaae
Eiliili XiEEOIf—I 1 2 3 4 5 6 7 EIIXIIE JEEOIEI
7431553 @8158
E Bill %Eltl 1 2 3 4 5 6 7 it PJECI
L'ElaiXI %E L'EI‘ are
§|A10|tl 1 2 3 4 5 6 7 §|A)0|L'.l
553mg EEGIE .’tfllxli ééfil-t-
amp) 0114:) 1 2 3 4 5 6 7 4| IOIEI
5.4149411} $565M ilxloil 3172151419 ANION 1:29—8- 0131221111771
Eatol LIED} 1 2 3 4 5 6 7 glam gt}
HEIE 4‘— SJICI 1 2 3 4 5 6 7 Hill: 4‘- 916i
egaoltl 1 2 3 4 5 6 7 agaioltl
gmyq 1 2 3 4 5 6 7 gmgq

6.2353411} 1.1413317] 31401 £16.} 64121 421014 1:64:— ole-slam}

germ Lien 1 2 3 4 5 6 7 gatol if-Cl-
flilg 4‘- ‘RJICI 1 2 3 4 5 6 7 {ERIE 4‘- 916i
%EHMH 1 2 3 4 5 6 7 ggamq
NNEH 1 2 3 4 5 6 7 imam

268

7. s41 alt-1:144] £118.} 34121 4171014 e744:— 0161—3141144?

glam 14311:) 1 2 3 4 5 6 7 galol Zg-tl
NEH-El -1‘- 91121 1 2 3 4 5 6 7 Nil? 4‘- $21121
EEEMG 1 2 3 4 5 6 7 ggfimq
NNEG 1 2 3 4 5 6 7 §m§q

8. "9:13! alflillloll Ella} 5314191 8210M 126-8- 011113141147}?

 

glam 149.411 1 2 3 4 5 6 7 galol gt)
Nile-ii -’t— 21121 1 2 3 4 5 6 7 {LEIQ 4‘- 2,1111
$§§10|Cl 1 2 3 4 5 6 7 2513mm
amen 1 2 3 4 5 6 7 gmgq

>-'< 11% 7|fi§0| $1.1 E§l7|91 @315 {NIEE Aléloll EL éMElF—l. 0|? {Hill
fig :r‘DH‘éi 7132111 944171 211-Ell. :1 ESE OIEIIOII IIIME' EQOII 133315”
E 51 %Sél/KIE. Dial-“.1194 éUZlOI EE- 5—6 aom 135.3 53% EPJXI aW313“
9:125 $7.191 4 11% 511614134 ELICI.

1. 421% 51141751317] 4111%--%...

EEGIJQXI 9.11:1 1 2 3 4 5 6

Q

512.1313 3:161

4°45le
911% 31 Lilli 1 2 3 4 5 6 7 #1212 Ed @111
31%| 7l-EQOI 91111 1 2 3 4 5 6 7 519:1 7i-‘5—SOI 91111

2. i{lull Eli/$141317] fill-16%...
3119:1611 gm alt) 1 2 3 4 5 6

#946170
9.1% at £61 1 2 3 4 5 6

$21 7i-‘5—élol 91111 1 2

Q

“7‘21 3111 41 Cl

Q

$942 3d 251
419:1 7i-S-éiol 531111

(.0
as
U"
03
Q

269

3. e41 113}?! 4.111%-%...

$0.46): gm 91:1 1 2 3 4 5 6 7 {19:16}: 441:)
:ii‘iélflxl
9.13 at £51 1 2 3 4 5 6 7 43151243161

7112-] 713210! 91121 1 2

w
4s
01
0‘
Q

421 71549 0| 94151

4. 49:11:] 6691.14165- a...

$94611 4,126 9.11:} 1 2 3 4 5 6 7 4.216111 411:}
$9461?“

31% ad EDI 1 2 3 4 5 6 7 FEE ad ZJEI
'TL‘EJ NESOI 9117—1 1 2 3 4 5 6 7 $194 7i'S-éJOI $3151

 

>I< Elg- EEO" EIISII Ciel-ti- 412121 %JflfiiE 3&0" ’O'E SHEFQAIE.

1. LIE- :15} Elli-616.1% Bl 61311611 “431414111 564111617] 5141” 0" lill6Il...

Q E_ Biol %‘1 2.1%:
Eloltl 1 2 3 4 5 6 7 $2.15 EIOIEI
2. Ll°ll4fl “4134111 %flflilflili‘Pfl

@1183 EDIE) 1 2 3 4 5 6 7 (Ala-El EDIE)
3. LI: 636.6 61613-361 61616141 “EEIIEZI 5141 4167M 3141““ Ell6ll...

6* EE-t- 9101 an 216:—
EDIE) 1 2 3 4 5 7 21‘5- EDIE)

O)

4. 616117.11 “fitllflZl—‘i— «(16171514 7}“

it EE-t— E'JOI 21:1 ‘21-:-
EDIE) 1 2 3 4 5 6 7 91E rEIOIEI

270

 

6. #011411 “36649 31915149)

films; EDIE) 1 2 3 4 5 6 7 916663 EOICI

7. U165 33%} 611511436511 11131161141 ”91%| al‘flfil‘l’fll Cll6ll...

Q EE-t— BIOI call 21%
EOIEI 1 2 3 4 5 6 7 215 EDIE.)

>:< :3 E210" 1:161 0161-5- 412121 %JilfilE 51-011 '65 616—41412.

1945171 9.03% 66171 $1611, 61—3— 7166:- 61414 61.4.6.- 6121 94:11, 714161 7.461
1:111 411:1." L11:- 61°11...

(1)5612: 1111111 (211111111 (313711364121 (4)849 (5161612; 6.161

 

3 ----- 4 ----- 5 II??- Ell E-E EOIEi

3. “.6173 65553— 191-5161:? %%€— 661-616 9.1561." LIE 61611...

(1161669.: ELIEH (2)3161 (313711215161 (416.164 (5)5612: £161

“216% E11161-65: E61156?! 3341—1:- iLE’JEEE—El 113.61% $20101 6.1131.” LIE 0]

(DEEP—E ELHIH (2)91511 (3)3113Es‘cl (4)5321 ($515123 HQ

5. 1361-3 6.16191 6173203621171 01:: 63E 466-01 551616131 6321611171?

 

£164 ERIN 9151 1 2 3

.3:
UI
n:

21.61 42.6011 555:1

271

 

6. “%Z‘liifl 7361141 ’EJEPI Ii‘l 33611615131 437—16151.” bl“:- °l°ll...

(DEEEE E'JDI (215.1111 (312362351111 (4)6161 (5)6132: tit-i

7. “23111-6:- E’rlfi'iig: 91611 $173313*% 701516131 3.1173521]
@6151.” E- 0101]...

1111.11

111:9.- %‘Zi6l71l @6161

(1)5812§ '3.“le (2)'.='JEH (3)3Xi 3%”:1 (4)319 (5)5512; {32—1

8. 121—151.15 50.00065— 716— 61 46 3161131 , 514.3 :1 65501] 6161.3. $175311;
%— 61611 7l-7—614l6161161?
(1) 0-10.000 5.14 (2) 10.000 - 20,0005! (3120. 000 - 30,000.34

(4) 30.000 - 40,000 a (5) 40.000 - 50.000111

>:< 0111161 55-91 L161.

1.0131541 "éé—E? (‘a”) (01)

2. 6431.51.91 LiOIE. (“:1 4ii)
3. 0431.39} 75.61%? ( )

4. 043151-91 %tZIE? (3H9 $50" 03. 51121512412)
1) CHEM 2) %éE/N 3) 63012.5 4) 7|El ( )
5. 01:131-1— 71239l 6?. 333

11111

%- ‘F‘iJ—E? (a 5563‘ ‘L

d3
03".
19

4 3.5.1.5 1:116:11 0113—111 £5 131612 ’51°1 ZF-fl’tl-‘P-

( ) “"911 ( E ‘9)

‘-

6361611 25.4441 1141636161;

272

i" P.

 

 

)

 

ues ionnaire
3* “lei-E" “3% 31% °."+"§2| Type (A 3.6% a); ’ElqziflAlla-ype;

>14 0161.321 6 ~1ng 75161 25.3412

 

Post—Test

aa- aa new 6141

5—5. EOIXI.

filial-El 5:219! 515% OIEHOII J{iMli‘l “.JQOII %3

X 51% 7|ޤ0ll Ell":l Dial-E'-

 

%‘

L.

6134161611 6161 61615—21 131101.). 1:74
6

l

L.

E
61.61. 634161611 61161 016161-91 41216161

33.
1

1_.

1

31
L.
3.1

2

1

'EISiXI $2!

3| 9125 %flfi' 4 5.1% EHEMIEI EILllll.

2101: 52111 6611412. 01615-91 ’éUZIOI 2
1. 1.1414121 11641416171 614161

filMOICI

2. 61614121 6141616171 614161

a ‘0’5 6ll—‘i‘-i.l*l$>_?

A

.erU.
525m.
WWQ
3W5:

..2

5E
MEN
ad

IEISIXI %Q
1

51410111

A

.’.EHlXiE Eéfilf
§|AlOI El

7

273

3. ’54:] 2191:3149} 5.116161 151636415101] 6116.} 011615-91 43110111 1:741:3- ‘o'fi 6H

3151:1119?

%QiMBI %QfiMBI
@658 aflEE
1:117:11: HEEOIEI 1 2 3 4 5 6 7 will: EEQOICi
5653 5653
Q PJEXI %Q51 1 2 3 4 5 6 7 Q ElECl
flil61xl %Q 561% Bit
mumq 1 2 3 4 5 6 7 HMNH

$517115 E§61E .’tHleE 5%“:-
§|A171 01141:} 1 2 3 4 5 6 7 §IA1OIE1

4. 61:16 6161614161 6.1616 6.346.161 616.1 61616.16 41216111 :53- ‘O’E 611
51541419?

513611614121 EIJEII§IA19I
@658 @653
Elliili JHEQOIH 1 2 3 4 5 6 7 Ellilli 11§Q0|Cl
@6153 64619.18
Q c357“ $1361 1 2 3 4 5 6 7 Q B551
fl§|611l %Q ’LIEIQ 9.161
6|410ICI 1 2 3 4 5 6 7 511110161

¢HII1§ EEfi-t— .’tHIXIE Egali-
§|A171 01111:) 1 2 3 4 5 6 7 6| WM

5. Maxi 514.141.6171 5111...] 616.1 64161 61210lb1 .2416:— 61616111671

@3101 Llfltl 1 2 3 4 5 6 7 @2101 it}
HEIQ 4‘- 21121 1 2 3 4 5 6 7 ’LIEIQ 4‘- 91:1
$gamq 1 2 3 4 5 6 7 agamq
amflq 1 2 3 4 5 6 7 gmfiq

6. 6614121 1641416171 614161 6161 616161 1.1716161 £17419: 616161411171

55101 HEB) 1 2 3 4 5 6 7 5301 if.)
HEIQ 4‘- 21111 1 2 3 4 5 6 7 .‘JEIQ 4‘- 21m
egamq 1 2 3 4 5 6 7 agamq
gmgq 1 2 3 4 5 6 7 imam

274

 

 

7.6%1flflfi4W11661W1*14N%L%N’Mmagqm?

%Eflluflq 1 2 :3 4 5 6
666*Faq 1 2 3 4 5 6
%QQNH 1 2 3 4 5 6
QNEG 1 2 3 4 5 6

8. ”31%| 6.1133146] c1161 C1119] ”$210111 1:793

%Emliflq 1 2 :3 4 5 6
LIEIQ gF 811:1 1 2 3 4 5 6
%QQNU 1 2 3 4 5 6
gmgq 1 2 3 4 5 6

qqqq

EQOI gm

1.1ng 4‘- 91111

591615101111
$501361

666666?

\lflx‘lq

>’< 111% 7lfi§0| "1"}.1 €1§17|91 QQE {‘JNIQE AIQOH
ES :r‘UHQ 7156111 914171 %EXI. :1 6152 013110“ K“MEI I=fl§01l 521315”
E Etfl 9151:1412. 01113591 QQOI EE- 3—‘1‘ Ii'lOIKI 53.53 535-51 £19.17“ QQSW

91224. %‘flfi' 4 312 5161MB 1511—1121.
1. ’35! 11161615171 411151666...

$flfifilflllfiu- 1 2 3 4 5
#121613“
633166 1 2 3 4 5

31121 71-‘32‘JOI 91:1 1 2 3 4 5

2 16111641711317] {11'55‘ $7...

$21 $111 6W 31:1 1 2 3 4 5
61216176
33% QG 1 2 3 4 5

'T‘i’J 71EQOI 91151 1 2

w
.p.
U"

275

%EOI €111

QEIQ -’1‘- 21111

%QQOIQ
£01361

Q QIMQEL Olir t'xil

x)

Nix]

\l

\‘lfi

?213111 Q51

E‘QQ ad Q61
71%! 71%QOI 91121

512.1 8131 Q 111

:HEJQ ad QE1
719:] 7134801 21111

 

3.6% 23319;”1’1-‘66...
371946111 QXI 91111 1 2
$21617”

9113 31 Q51 1 2

$61 71§Q0| 21111 1 2

4. 61:13! 2361?; 4111-116- %..

$94311 QXI 9161 1 2
3119:1617“
93% ad Q61 1 2

$21 71's-Q0I 91111 1 2

X 111° 7|21§0| =1171 BQNE

£1218 Oi‘é-a-LIW? 5‘94 Q55 W951: 511011

1.4151311 1115.751517] 31611::

{5161 215-”F6111

Sgt-EH 1 2 3
2. 11164171 113-1441:6171 3141::
612161 1.151 $612

61x11 2161.1 2

3 66111111 75.6171 3161”
{1611 6:461?“

°-:-E1 1 2 3
4 2.61611 63-66617] 61416:
61216.1 1161 $612

E1W3 ‘21 1 1 2

5 ‘66 £116.61 1: 5171617171
{64 -’._r‘--’F61II

Sit-C1 1 2 3

3 4 5 6 7 311216111 4.11:1
3 4 5 6 7 $612 31 £61
3 4 5 6 7 311?; 7173-QOI 2M
3 4 5 6 7 $6161.21 4151
3 4 5 7 $213; 31' 2.1121
3 4 6 7 311?; 71EQOI 71121
3 $0111 Q Eo—“r-fim 24:XIOII E1115.1 013156° |
0’11 6116561412.

271 617371+E%...

(A)
.p.
U"

4 5 6

276

212
Elm?!

Fl)!
-1>

32

SEQ
1161

2".
17.114

1111
1>

O3
611
at
.29.
1‘3.
.191.
2:.
111
r0
Iii
111.
0i
3.131
N
-r-‘
128.
-_9-.

Q2115. L1%."- 661% £8 %612
61x11 211:1 1 2 3 4 5 6 7 Ulill 9.1111

7. ‘flllil E125§I41E 51171 33173 71717157737...

6161 Eff-616! DIEM Q

QJ-EQ 1 2 3 4 5 6 7 65-7-6131 24121

8. 6:16! '21113114170 71 2171171 1161011...

613121 1143 $618 61% $6613

U|X|1 211:1, 1 2 3 4 5 6 7 HI I: 21111

>:< 111:. $6101 0413131394 612119-JXI61E 11016 E 6H-.— LIME.
1.7116617311101111 21121611: 67111661 x1116

7.5161 HENRI CHEW

0 51:1 1 2 3 4 5 6 7 flilflfl

6163-50111 61-3-51 01:1
211.51g 0| 211:1 1 2 3 4 5 6 7 21§1610| 1:1

1 £6 51311 66:1 16 61°16 616 2111612-

( ) “"511 ( %1 Q)

4 onggwct 21113121 2.511011 gum—1411 1:11:31” 7JA1fi1Llc1,

277

APPENDIX F
KOREAN VERSION OF EXPERIMENTAL STIMULI

Stimuli Type A

 

278

(JIB'I-fil- El 1IIAI LII—g. %% Type A ‘I’J° "I‘fisfl ZS-fJAL‘L)

3373-?- %7i624193°11*1t 519 71%)“ EH? “611% 6.673%}? 611E 3'6 E‘ilé}
2163145} 33733123 fllEt— 71%]94 973303 65119} $716673 71% 65-? 7‘3
E01] [113} 71‘6““ 611711] $76} (“3%, 11%, 113131. 11““)% 75% 11:2: 1?— 3}
E 111511145}. 211 6917694 941316 51%44 26%145}.

 

,2... _.;._H__. u

1?

56.1.1 .

W 1
'7' .

F7
:1

(336 £37134 9691 7|?3) : 671616716 (26661

O
21,”; aging S‘éfil 275215104 EfEJP—I 266763 HHENJ 91201, 2
2:12} =1.‘
7

7191 EH7I. 4:6, 621% H1131?“ 2%AI9IIL QEHNS} E93294
11%!2 16126317“ ill-“.5611 21E %EOIL} NEE.

 

¢~ 7|?E1): 671, 6, M1171 11116. 61
616621 61661 66716011 61¢ 56612 216 661qu 71216.

 

thll (75%: 33171315? 7|“) 6626666 6712I 8H6

31’“ 9J7| $|3H a—lQ‘B—l 3% Hal} EIHIE Afgfiffll 3 £31} £794 %7l,
5, &EI|7I Hflé. §1§f§§21 43%?! i‘flEEQg 64%?“ EC} E‘d 7517“
Sggaé E'Iéaffl 911-1:- %{SOIL} HESS.

279

 

 

>6 666'— 676 616116 67666616 «>161 471 716.121 6666 51-8-31 601
616 6766616 61661. 120016 16 66)

1. 66161616666171 66: 66 1666121 219171211

2. 6.16661 $flfl§l7l 66: %éil 166 6.1666 7121)

3. 6126 E16661: 661 16661.21 219—1 7121)

4. 6’3 21666: 661 1616 6666 7121)

%‘5’3’31 “.11 (61:1 ‘66 "121) %‘5‘3’3: 6:1 (’51-:- 62.161794)

 

"EU! Fulflfli‘wl ilk}
59:13] 319'; Elk}

 

61111 6161616171 616121 111.1611 61
a 61616 616 316121 111661 610.1616,
mfl£%,%fl 6666,262 66
6.6171 666 671616 6161.

311'! 3313’} H] 111511 ""1

5:.
O

1.11 ~11 <21
91 610.1, 911-611 3611 0111.
v1 11 1.11 11 .11 10111 011 616-6
62.1% 1
671 6161
111.111 :1 6 6916 01 6101161 1111

6%.] %%£‘§%7§fl 3126 1331313913}.

7. ~ 1’:
T ‘3 61‘

21 21-

61 1.116160] 71111 016, 3.31.31

91 61161 11.161 601 12611 016
6 6—

I]

QM} w-flflflfl 33149} 38¢! £115
5141:: 21% 3612} flflfl %fifl’éfl?

6 6166161, 0% 916661 66.16 2161
11.11 666161011 616612.32. 6x16 61
6 3'1‘Vé~?~’r— 616160111.

01 616166 2 60.1 6216 61111 6
76661 216121 6161 71-6616 61 614101
73.14131} 6.9.61, 21 6 67361631 6766.

6, 631511711 3116 613:16-96 2161616:— %—
67666011 31% 667661
6 716601111.

 

 

 

(X13. iii: 3%?- Egéfifla Eggfllh 1998-20005 7l-E)

280

 

>6 W7} 6.166 4711 7166 19986161 2000de
61111 67166716 (2666 11016 6116761) 6617166.— 51-3—34 611:1.

1_ 31111751616LL§17I 73—6 (Hyundai): - 159
2. 7341373761 $QE§17I £16! (Samsung): 109

3. $33" 1311111 73%! (Binggrae): ~136
4. 656:! 31:5 8%! (Nongsim): 125

386 7|$1 El. 73 1| 1*(Corporate Environmental Index)

1' ngHF 0: £7} EH7I. 45E. E2} 2% 6487|Eil

* Bail?- -100= £7} CH7I. 457%, E2} 2‘551-8 7I-E7‘I 5- HH9—I EQE
54% E‘IEEF @913

* 13% XI?- 100: :67} UPI. 453 5%} 2% 34% 7|§il I.39—l 2%EQ
3 5:32}? @9-

Corporate E nvironmental Index

 

    
  

 

 

 

15° "W 125 '“11”—1
m 109 . y - 1'
"(:2 100 7
E
z
2 50 . 7
8 lSamsung
E 12 0 7 7 1 ' .
:5: g Samsung Nongsim - ”“95”"
to I
9 as '50 7 . ,
S
c
"2’ —100 , '
C ‘ ~
g . . .
'g -150 -136
UJ 459' ' ' i
-200
Company

(* For the purpose of microfilming this paper, the labels ”Hynudai and "Binggrae" were
removed from the chart above so that the chart is different from original one used in
the experiment.)

281

Stimuli Type B

(21181-11 113'. 611111 Luggg Type 321% "l‘i‘jafl 25-’;}*l.9_)

 

282

BEE; 61666166161: 6:1 71661 :16 616 666.16” 71166 6661
66-116. 6666 7116-:— 7166 6666 6616 67166716 66
.601 66 71661 1117161 616 (:16, 66, 2631. 66116 6 616.- 66611
6:— 71166116. 21 6166 6111-1:— 6166 66—116.

 

1

1.12117 1 5‘71014111 91“.?! 7H2) :271Ea7l-ffi- (2636

‘33 31% iDS ”SE—IN illfiloi Cl%4‘9-| Egfifli Hfléfill 219-01. 3
all 27M EH7| Tfl. E‘c’fS 137-lat?“ Slfiklilll éflfilfiIQl i'fliP—l

>166 612161711 666: 66- 66016 6215.

l—

 

 

mat (:Zgl7l—ff- 21:—#— 7I21): 37L 5. 61.6l7l 8H3. 61
gr

SQI’J Elfii-l 357F150" 54¢ E‘a‘filfl 21=§él0|l4 7M5.

 

 

2.2/3! £42 3:32; 51:6: 7161); 66262663 716 HHE

5W 9J7| $I3H 544994 85 7|§El QHIS Algalfll :1 all E7194 87L
E. .*_*E!||7| HHS. ilélfiflll 3E3 iflflgQg 348’“ EC} fl‘d 7547“

Eggflé E‘Iéfill RAE %QOIH ”21%.

283

 

>:< 66—11—

3375‘72‘1311511 %Zfi‘igl‘flrl‘c 01311 47H 71%194 %XJé-é- 51-3-34 €01

41176} Q736V3€3 141E515}. (200151 1% 311.1%)

1. 134951” Ttflifll 735*: 5%!

9‘5”.“

612151 E15531

(3531-74 949M521)

Edam-$165.3!” '38: %4‘4 (E; 33%;; 7%)
(5643' BEE-35‘: £2! (35171-74 9191 7|21)

55.31 (51-3 Egifi 7|Zl)

 

33‘3””? ”F“ (%

20

'5 °-t'=-1 “'21)

66316: 66 (7216 621617121)

I.—

 

66 517—6166
66 6121 Elk}

EDI Eilfl’fliwl ilk}
511355! 315 Elk}

 

314%:

 

11% ’1’7 I}! ‘61 I137} 7]

216 366 11

713%, %311 fiflE’él, 1313’— ”3°13 1’5-

if?! 673366 6°

61619} 3&4“! 61"!
$1.56] ”363111—11, EH

31617] 66- 26716171 361:}.

321 361.11.} a] 511 ~51] 1;] 5?: 13 x] g), :6 13116
91 6'11 (1.}; 911?} E? ‘2} 6.16-EEO] 36H 016‘,
31>: 51-71 :66 616151.10] 101111 01.601, :11 E11"?!
1111 6 7611 016 :16 r 66 1
9] 613 $16-$10] 121111 01% {”671 51212: ‘
6:561! :1 51‘— °66 016— 666161 1111

9113113515}.

551‘}! 16616671 311/454 filial a}
‘5 514% Xl‘rl 367d 34’69} 5678?.«73
7166- 61681111, 0% 9.0366 66-8.-
OI'ESH 4L“ fl'fg 411.3]01] 21411.10‘3; 31:1}
6 8H9 5E739-7’1‘7 §I*Hi 01C}.

0] 314-3—8- Zl %‘fl %7-‘1351 *HH 6}
731311] 2163+ 521% 716711-3- 15} ilk/‘16]
7517’???“ 96°01, 211% 67331-611} @736
.6511?" 3313 *}°d‘é°11%13}lr: 6
%fiiidl 34%" %Vani
%fE- 7l°d%°l'4.

 

 

(113 iii: 86-?- Eaéfifla 8537121, 1998-2000's.! 7|?)

284

 

 

X Eilciigilifie 733-5333 fiofiofiflflt iii-iii 4711 719394 1998011/‘1 ZOOOEE
4°] 51V}??? 71%:- (3‘337‘6 “J26“ 31%i1) jail]??? ‘31773—34' €13}.

1. 138757” $flfi§i7l 2'5! (Hyundai): - 159

2. SACHS” g'flfiflfl %é! (Samsung): 109

3. 616 315 2756+ (Binggrae): —136

4. 61:18:11 E131 %Q (Nongsim): 125

X 7| 33 E} fiXII"(Corporate Environmental Index)

* ngl-i- 01 ='=1“7PEH7|. 515$, 5%} 2% 3487I-fi-il

* 35.1%“?- —100: €1.17} I3H7l, -’1‘-§, 5%} 2%618 7|Eil 1=- HH94 SEE
Q3 “32? @93—

* Ea XI?- 100: ='67} EH7I. $761. 521‘ 2% 51% 7l-Exl E994 295.3%
g %Ié’fl @513-

Corporate Environmental Index

  
 

 

 

150 7 125
_ 109
g 100
§
2 50 A
8 FIHyiJnd; 1
C _ __ _____1____ . ‘
%- g 0 . . ' Iamggrae I
E 'g Hyundal Binggrae , .
° 2 -50 E
9 ‘0 1
a
C
a -100 1
‘ 1
2
E -150 1 —136 1
W 459 1

-200 ~ - , . ~ ._._. , -‘

Company

(* For the purpose of microfilming this paper, the labels "Samsung" and "Nongsim"
were removed from the chart above so that the chart is different from original one used
in the experiment.)

BIBLIOGRAPHY

286

 

. Lml_ __._

 

 

BIBLIOGRAPHY

Afsah, S. (2002). PROPER (Program for Pollution Control Evaluation and Rating): A
Model for Promoting Environmental Compliance and Strengthening Transparency
and Community Participgtion in Developing Countries. World Bank. Available:
www.worldbank.org/nipr [2002, 4/2].

Afsah, S., Laplante, B., & Makarim, N. (1996). Programme-Based Pollution Control
Management: The Indonesian PROKASIH Programme. Asian Journal of
Environmental Management, 4 (November)(2), 75-93.

 

 

Afsah, S., Laplante, B., Shaman, D., & Wheeler, D. (1997, July). Creating Incentives to
Control Pollution, [DEC Notes]. Policy Research Department of World Bank.
Available: www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Afsah, S., Laplante, B., & Wheeler, D. (1996). Controlling Industrial Pollution: A New
Paradigm (No. 1672 PRD working paper). Policy Research Department of World
Bank. Available: www.worldbank.org/nipr/work_paper/1672/index.htm [2000,
4/26].

Afsah, S., Laplante, B., & Wheeler, D. (1997, March). Regulation In The Information
Age: Indonesian Public Information Program For Environmentgl Management.
Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htrn [2000, 6/5].

Afsah, S., & Vincent, J. R. (1997, March). Putting Pressure on Polluters: Indonesig'g
PROPER program. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htrn [2000, 6/5].

Afsah, S., & Wheeler, D. (1996). Indonesia's New Pollution Control Program: Using
Public Pressure to Get Compliance. East Asian Executive Reports, 18(6), 9-12.

Ajzen, I. (1985). From Intention to Actions: A Theory of Planned Behavior. In J. Kuhl &
J. Beckmann (Eds.), Action Control: From Cognition to Behavior (pp. 11-39). New
York: Springer-Verlag.

Ajzen, I. (1988). Attitudes, Personality. and Behavior. Chicago: Dorsey.

Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human
Decision Processes, 50, 179-211.

Ajzen, I., & Fishbein, M. (1980). Understanding Attitudes and Predicting Social
Behavior. New Jersey: Prentice-Hall Inc.

287

 

 

 

Anderson, L. W. (1988). Attitudes and their Measurement. In J. P. Keeves (Ed.),
Educational research, methodology, and measurement: an intemationalhandboolg.
Oxford, England: Pergamon Press.

APA. (1997). Publication Manual of the American Psychological Association (4th ed.).
Washington, DC: American Psychological Association.

 

Aronson, B., Turner, J. A., & Carlsmith, J. M. (1963). Communicator Credibility and
Communication Discrepancy as Determinants of Opinion Change. Journal of
Abnormal aLd Social Psychology, 67, 31-36.

Arora, S., & Cason, T. N. (1995). An Experiment in Voluntary Environmental [
Regulation: Participation in EPA's 33/50 Program. J oumad of Environmental
Economics and Management, 28(3), 271-286.

Arora, S., & Cason, T. N. (1996). Why Do Firms Volunteer to Exceed Environmental
Regulations? Understanding Participation in EPA's 33/50 program. Land
Economics, 72(4), 413-432.

 

Arora, S., & Gangopadhyay, S. (1995). Toward a Theoretical Model of Voluntary
Overcompliance. J oumal of Economic Behavior and Organization, 28, 289-309.

Associated Press. (1991, June 9). Presbyterians Ratify Teaching on Sex, Ecology. Boston
Globe, pp. 4.

Badrinath, S. G., & Bolster, P. J. (1996). The Role of Market Forces in EPA Enforcement
Activity. J oumial of Regulatory Economics, 10(2), 165-181.

Bagozzi, R. P. (1994). Measurement in Marketing Research: Basic Principles of
Questionnaire Design. In R. P. Pagozzi (Ed.), Principles of Mag‘etig Research (pp.
149). Cambridge, MA: Blackwell Publisher.

Bagozzi, R. P., Baumgartner, H., & Yi, Y. (1991). Coupon Usage and the Theory of
Reasoned Action. In R. H. Holman & M. R. Solomon (Eds.), Advances in
Consumer Resear_ch_ (Vol. 18, pp. 24-27). Provo, Utah: Association for Consumer
Research.

 

Batie, S. S. (1997). Environmental Issues, Policy and the Food Industry. In L. T. Wallace
& W. R. Schroder (Eds), Perspectives on Food Industry/Govemment Linkages.
Boston: Kluwer Academic Publishers.

Batie, S. S., & Ervin, D. E. (1997, June 8-9). Flexible Incentives for Environmental
Management in Agriculture: A typology. Paper presented at the Flexible Incentives
for the Adoption of Environmental Technologies in Agriculture, Greenbelt,
Maryland, Gainesville, Florida.

288

Berberoglu, G., & Tosunoglu, C. (1995). Exploratory and Confirmatory Factor Analyses
of an Environmental Attitude Scale (EAS) for Turkish University Students. _T_h§
Journal of Environmental Education, 26(3), 40-43.

Bettman, J. R. (1979). An Information ProcessLng Theory of Consumer Choice. MA:
Addison-Wesley Company.

Blacconiere, W. G., & Northcut, W. D. (1997). Environmental Information and Market
Reactions to Environmental Legislation. Journal of Accounting, Auditing_&
Finance, 12(2, Spring), 149-178.

Blackman, A., & Bannister, G. J. (1998). Community Pressure and Clean Technology in
the Informal Sector: An Econometric Analysis of the Adoption of Propane by
Traditional Mexican Brickmakers. Journal of Environmental Economics an__d

Management, 35(1), 1-21 .

Boniface, D. R. (1995). Experiment design and stfistical methods for behavioral and
social research. London; New York: Chapman & Hall.

Booth, W. C., Colomb, G. G., & Williams, J. M. (1995). The Crafi of Reseafl. Chicago:
The University of Chicago.

Borenstein, S., & Zimmerman, M. B. (1988). Market Incentives for Safe Commercial
Airline Operation. American Economic Review, 78(5), 913-935.

Bovee, C. L., & Arends, W. F. (1992). Contemporary Advertising. Homewood, IL.:
Irwin.

Brown, S. P., & Stayrnan, D. M. (1992). Antecedents and Consequences of Attitude
Toward the Ad: A Meta Analysis. Journal of Consumer Research, 19(June), 34-51.

Brucks, M. (1985). The Effects of Product Class Knowledge on Information Search
Behavior. Journal of Conaumer Research, 12(June), 1-16.

Bruner, G. C. I., & Hensel, P. J. (Eds). (1992). Marketing Scales Handbook: A
Compilation of Multi-Item Measures. Chicago, IL: American Marketing
Association.

Bryant, B. (1995). Environmental Justice: Issue, Policies and Solutions. Washington,
DC: Island Press.

Burros, M. (1996, 7 February). A New Goal beyond Organic: Clean Food. New York
Times, pp. B1, B5.

Cahill, L. B., & Kane, R. W. (1994). Corporate Environmental Performance Expectation

in the 19908: More than Just Compliance. Total Oualifl Environmental
Managgnent, 3, 409-420.

289

 

Calder, B. J ., Phillips, L. W., & Tybout, A. M. (1981). Designing Research For
Application. Journal of Consumer Resea_r_ch, 8(September), 197-207.

Caldwell, L. K. (1990). Between Two Worlds: Science, the Environmental Movement,
and Policy Choice. Cambridge, UK: Cambridge University Press.

Campbell, D. T., & Stanley, J. C. (1963). Experimeni and Quasi-Experimental Desiggs
for Research. Boston: Houghton Mifflin Co.

Chan, T. S. (1996). Concern for Environmental Issues and Consumer Purchase
Preferences: A Two-Country Study. Journal of International Conflmer Marketing,
9(1), 43-55.

Che, Y. K., & Earnhart, D. (1997). Optimal Use of Litigation: Should Regulatory
Information Be Withheld to Deter Frivolous Suits? Rand J oumal of Economics, 28,
120-134.

Clifford, M. (1990). Kicking Up a Stink: South Korean Government Reels from Anti-
Pollution Backlash. Far Eastern Economic Review, October 18, 72-73.

Cohen, M. A. (1998). Monitoringand Enforcement of Environmentafl’olicy. [Working
paper of World Bank]. Owen Graduate School of Management, Vanderbilt
University. Available: www.worldbank.org/nipr/workJaper/index.htrn [2000,
4/26].

Cohen, M. A., Fenn, S. A., & Nairnon, J. S. (1995). Environmental and Finflcial
Performance: Are They Related? Washington, DC: Investor Responsibility
Research Center, Environmental Information Service.

Cook, T. D., & Campbell, D. T. (1979). Quasi-Experimentation: Design & Analysis
Issues for Field Settings Boston: Houghton Mifflin Co.

Cormier, D., Magnan, M., & Morard, B. (1993). The Impact of Corporate Pollution on
Market Valuation: Some Empirical Evidence. Ecological Economics, 8, 135-155.

Cote, J ., & Tanushaj, P. S. (1989). Culture Bound Assumptions in Behavior Intention
Models. In T. Srull (Ed.), Advances in Consumer Resea_r_c_h (Vol. 16, pp. 105-109).
Provo, Utah: Association for Consumer Research.

Creswell. (1994). Research Desig: QualitzLive & Qumfigative Approaches Thousand
Oaks, CA: Sage.

Cropper, M. L., Simon, N. B., Alberini, A., & Sharma, P. K. (1997, December). I_h_§
Health Effects of Air Pollution in Deth India, [No. 1860 PRD Working Paper].
The Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/workjaper/index.htrn [2000, 4/26].

Crowne, D. P., & Marlowe, D. (1964). The Approval Motive. New York: Wiley.

290

 

 

 

Dasgupta, S. (1999, November). Opportunities for ImprovingEnvironmental Compliance
in Mexico, [No. 2245, World Bank Discussion Paper]. Policy Research Department
of World Bank. Available: www.worldbank.org/nipr/work_paper/index.htrn [2000,
4/26].

Dasgupta, S., Hettige, H., & Wheeler, D. (1997, December). What Improves
Environmental Performance? Evidence from Mexican Industry, [No. 1877 PRD
Working Paper]. Policy Research Department of Word Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 6/5].

Dasgupta, S., Huq, M., & Wheeler, D. (1997, February). Bending the Rules:
Discretionary Pollution Control in China, [NO. 1716 PRD Working Paper]. Policy
Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Dasgupta, S., Laplante, B., & Mamingi, N. (1998, April). Capital Market Responses To
Environmental Performance In Developing Countries, [No. 1909 PRD Working
Paper]. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htrn [2000, 4/26].

Dasgupta, S., Laplante, B., & Meisner, C. (1998, March). Environmental News in
Argentina, Chile, Mexico and the Philippines, [Executive Summary]. Policy
Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Dasgupta, S., Lucas, R. E. B., & Wheeler, D. (1997, November). Small Plafnts, Pollution
and Poverty: New Evidence from Brazil and Mexico, [No. 2029 PRD Working
Paper]. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htrn [2000, 4/26].

Dasgupta, S., & Wheeler, D. (1996, November). Citizen Complaints a‘s Environmental
Indicators: Evidence fi'om China. Policy Research Department of World Bank.
Available: www.worldbank.org/nipr/work_paper/compwp/index.htm [2000, 4/26].

Davis, J. J. (1994). Consumer Response to Corporate Environmental Advertising. J oumal
of Coasumer Marketing, 11, 25-47.

Deily, M. B., & Gray, W. B. (1991). Enforcement of Pollution Regulations on a

Declining Industry. Journal of Environmental Economics yd Management, 21,
260-274.

Deutsch, C. (1998, 19 July). For Wall Street, Increasing Evidence That Green Begets
Green. New Yogr Times, pp. A7.

Donaton, S., & Fitzgerald, K. (1992). Polls Show Ecological Concerns is Strong.
Advertising Age, 63, 49.

291

 

Eagly, A. H., & Chaiken, S. (1993). The Psychologyof Attitudes. Fort Worth: Harcourt,
Brace, Jovanovich.

Environmental Protection Agency. ( 1 99 1). Environmental Education. EPA Journal,
1 7(4).

 

Environmental Protection Agency. (1997). EPA Green Lights Program Snapshot for
January 1997. Washington, DC: Environmental Protection Agency.

Fazio, R. H., Powell, M. C., & Williams, C. J. (1989). The Role of Attitude Accessibility
in the Attitude-to-Behavior Process. Journal of Consumer Research, 16(December),
280-288.

F eldman, S., Soyka, P., & Ameer, P. (1996). Does Improvinga Firm's Environmental
Management System and Environmental Performance Result in a Higher Stock
Price? Fairfax, Va.: ICF Kaiser International.

Fialka, J. J. (1998, May 4). EPA Puts Records About Polluters On the Internet. Wall
Street J oumd, pp. A8.

Fishbein, M. (1967). Attitude and the Prediction of Behavior. In M. F ishbein (Ed.),
Readirgs in Attitude Theory and Measurement. New York: Wiley.

Fishbein, M. (1980). A Theory of Reasoned Action: Some Applications and Implications.
In J. H. E. Howe & M. M. Page (Eds), Nebraska Symposium on Motivation, 1979
(Vol. 27, pp. 65-116). Lincoln: University of Nebraska Press.

Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An
Introduction to Theory and Research. Reading, MA: Addison-Wesley.

Fishbein, M., & Ajzen, I. (1977). Attitude-Behavior Relations: A Theoretical Analysis
and Review of Empirical Research. Psychological Bulletin, 84(September), 888-
918.

Fisher, A., McClelland, G. H., & et al. (1991). Communicating the Risk from Radon.
Journal of the Air & Waste Management Association, 41(11), 1440-1445.

F ombrun, C. J. (1996). Reputation. Boston, MA: Harvard Business School Press.

Foulon, J ., Lanoie, P., & Laplante, B. (1999, October). Incentives for Pollution Control:
_R_egulation and I?) or (?). Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Frankel, C., & Coddington, W. (1994). Environmental Marketing. In R. Kolluru (Ed.),
Environmental Strategies Handbook (pp. 643-677). New York: McGraw-Hill.

Gabel, H. L., & Sinclair-Desgagne, B. (1993). Managerial Incentives and Environmental
Compliance. Journal of Environmental Economics and Management, 24, 229-240.

292

General Accounting Office. (1987). Hazardous waste: facility inspections are not
thorough and complete (Report RCED-88-20). Washington DC.

General Accounting Office. (1991). Environment enforcement: penalties may not recover
economic benefitwined by violators (Report RCED-91-166). Washington DC.

General Accounting Office. (1993). Environmental enforcement: EPA cannot ensure the
accuracy of self-reported compliance monitoring data (Report RCED-93-21).
Washington DC.

General Accounting Office. (1994). Toxic substances: EPA needs more reliable source
reduction data and progress measures (Report to the Chairman, Subcommittee on
Health and the Environment, Committee on Energy and Commerce, House of
Representatives). Washington DC.

Gerstenzang, J. (1997, 21 November). Survey Bolsters Global Warming Fight. lg
flgeles Times, pp. A4.

Goldberg, M. E., & Hartwick, J. (1990). The Effects of Advertiser Reputation and
Extremity of Advertising Claim on Advertising Effectiveness. Journal of Consumer
Research, 17(September), 172-179.

Goldsmith, R. E. (1987). Two Students of Yeasaying. Psychological Reports, 60, 239-
244.

Goodman, A., & Streeter, A. (1999). Companies of the Year. Tomorrow, 9(1), 14-16.

Gray, D. B., Borden, R. J., & Weigel, R. H. (1985). Ecolgical Beliefs and Behaviors:
Assessment and Change. Westport, CN: Greenwood Press.

Green, D. P., & Cowden, J. A. (1992). Who Protests: Self-Interest and White Opposition
to Busing. Journal of Politics, 54, 471-496.

Greenwald, M., & Katosh, J. P. (1987). How to Track Changes in Attitudes. American
Demographics, August, 46.

Gresser, J. (1979). Managing Industrial Development with Environmental Management
in the Republic of Korea (Report No. 79-3): World Bank, Urban and Regional
Economics Division.

Grodsky, J. A. (1993). Certified Green: The Law and Future of Environmental Labeling.
The Yale Journal on Regulation, 10, 147 -227.

Hamilton, J. T. (1995). Pollution as News: Media and Stock Market Reactions to the

Toxics Release Inventory Data. Journal of Environmental Economics and
Management, 28(1), 98-113.

293

Hamilton, J. T. (1996). Going by the (Informal) Book: The EPA's Use of Informal Rules
in Enforcing Hazardous Waste Laws. In R. Pethig (Ed.), Conflicts and Cooperation
in Managing Environmental Resources. New York: Springer-Verlag.

Hanrahan, D., Wheeler, D., Keene, M., & Shaman, D. (1998). Develoging Partnerships
for Effective Pollution Management. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Harford, J. D. (1997). Frim Ownership Patterns and Motives for Voluntary Pollution
Control. Managerial and Decision Economics, 18(6), 421-432.

Hartman, R. S., Huq, M., & Wheeler, D. (1997, December). Why Paper Mills Clean Up:
Determinants of Pollution Abatement in Four Asian Countries, [No. 1710 PRD
working paper]. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm.

Hartman, R. S., Wheeler, D., & Singh, M. (1994, December). The Cost of Air pollution
Abatement, [No. 1398 PRD working paper]. Policy Research Department of World

Bank. Available: www.worldbank.org/nipr/work_paper/compwp/index.htrn [2000,
5/ 1 7].

Helland, E. (1998). The Enforcement of Pollution Control Laws: Inspections, Violations,
and Self-Reporting. The Review of Economic and Statistics, 141-153.

Henderson, G. V. (1990). Problems and Solutions in Conducting Events Studies. Journal
of Risk and Insurance, 93(3), 282-306.

Hettige, H., Hug, M., Pargal, S., & Wheeler, D. (1996). Determinants of Pollution
Abatement in Developing Countries: Evidence from South and Southeast Asia.
World Development, 24 (December)(12), 1891-1904.

Hoffman, A. (1993). Who Loves Love Canal. Tomorrow, 3(3), 58-64.

Hoffman, A. (1996). A Strategic Response to Investor Activism. Sloan Management
Review, 37(2), 51-64.

Hoffman, A. J. (2000 a). Competitive Environmental Strategy: A Guide to the Changi_ng
Business Landscape. Washington, DC: Island Press.

Hoffman, A. J. (2000 b). Integrating Environmental and Social Issues into Corporate
Practice. Environment, 42(5), 22-33.

Hofrichter, R. (Ed.). (1993). Toxic Struggles: The Theory and Practice of Environmental
Justice. Philadelphia, PA: New Society Publishers.

Holbrook, M. B., & Havlena, W. J. (1988). Assessing the Real-to-Artificial
Generalizability of Multi-attribute Attitude Models in Tests of New Product
Designs. Journal of Marketing Research, 25(February), 25-35.

294

 

 

Hunter, J. E., Danes, J. E., & Cohen, S. H. (Eds). (1984). Mathematical Models of
Attitude Changflnd Coflive Structures. Orlando: Academic Press.

Hunter, J. E., & Gerbing, D. W. (1982). Unidimensional Measurement, Second Order
Factor Analysis, and Causal Models. In B. M. Staw & L. L. Cummings (Eds),
Research in Organizational Behavior (pp. 267-320). Greenwich, CT: JAI Press, Inc.

Hunter, J. E., Gerbing, D. W., Cohen, S. H., & Nicol, T. S. (1980). PACKAGE 1980; A
System of FORTRAN Routines for the Analysis of Correlational Data. Waco, TX:
Baylor University.

Hunter, J. E., Levine, R. L., & Sayres. (1984). Attitude Change in Hierarchical Belief
Systems and its Relationship to Persuasibility, Dogmatism, and Rigidity. In J. E.
Hunter & J. E. Danes & S. H. Cohen (Eds), Mathematical Models of Attitude
Change and Cognitive Structures. Orlando: Academic Press.

Huq, M., & Wheeler, D. (1993). Pollution Reduction Without Formal Regulation:
Evidence from Bangghglesh, [No. 1993-39 PRD working paper]. Policy Research
Department of World Bank. Available: 9"
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

 

 

J affe, A., . B., Portney, S. R., Portney, P. R., & Stavins, R. (1995). Environmental
Regulation and the Competitiveness of US. Manufacturing. Journal of Economic
Literature, 33(1), 132-163.

 

J eon, G. J. (1998). Information as a Policy Tool: A Flexible Means of Improving
Environmental Oaality in Korea. Unpublished Ph. D. Dissertation, Naderbilt
University.

Jones, J. D., Jones, C. L., & Phillips-Partick, F. (1994). Estimating the Costs of the Exxon
Valdez Oil Spill. Research in Law and Economics, 16, 109-150.

Kassaye, W., & Verma, D. (1992). Balancing Traditional Packaging Functions with the
New Green Packaging Concerns. Advanced Management Journal, 57, 15-23.

Katz, J. (1994). Levi Straus & Co.: Global Sourcing. Cambridge, Mass: Harvard
Business School.

Kaufman, L. (1999). Selling green: What Managers and Marketers Need to Know about
Consumer Environmental Attitudes. Environmental Quality Maflgement, 8(4,
summer), 1 1-20.

Kennedy, P. W., & Laplante, B. (1995, August). Equilibrium Incentives for Adopting
Cleaner Technology Under Emissions Pricing. Policy Research Department of
World Bank. Available: www.worldbank.org/nipr/work_paper/index.htrn [2000,
6/5].

295

 

Kennedy, P. W., Laplante, B., & Maxwell, J. (1994). Pollution Policy: The Role for
Publicly Provided Information. Journal of Environmental Economics and
Management, 26(1), 31-43.

Keppel, G. (1991). Desig and Analysis: A Researcher's Handbook. Englewood Cliffs,
NJ .: Prentic-Hall, Inc.

Kerr, K. (1990). Thinking Green is No Longer a Hippie Dream. AdWeek‘s Marketing
Week, 31,18-19.

Khanna, M., Quimio, W. R. H., & Bojilova, D. (1998). Toxics Release Information: A
Policy tool for Environmental Protection. Journal of Environmental Economics and
Management, 36(3, November), 243-266.

Kiker, C. F., & Putz, F. E. (1997). Ecological Certification of Forest Products: Economic
Challenges. Ecological Economics, 20, 37-51.

Klassen, R. D., & McLaughlin, C. P. (1996). The Impact of Environmental Management
on Firm Performance. Management Science, 42, 1199-1214.

Knight, R. (1998). Profits and Principles: Does There Have To Be a Choice? London:
Shell International Ltd.

Konar, S., & Cohen, M. A. (1997). Information as Regulation: The effect of Community
Right to Know Laws on Toxic Emissions. Journal of Environmental Economics aLd
Managgment, 32(1), 109-124.

Krupp, F. (1990). Win/W in on the Environmental Front. EPA Journal, 16(5).

Kuhn, R. G., & Jackson, E. L. (1989). Stability of Factor Structures in the Measurement
of Public Environmental Attitudes. The Journal of Environmental Educziion, 20(3),
27-33.

Lafferty, B. A., & Goldsmith, R. E. (1999). Corporate Credibility's Role in Consumers'
Attitudes and Purchase Intensions When a High versus a Low Credibility Endorser
Is Used in the Ad. Journal of Business Research, 44, 109-116.

Laffont, J. J. (1989). The Economics of Uncertainty and Information. Cambridge, MA:
MIT Press.

Lanoie, P., Laplante, B., & Roy, M. (1997). Can Capital Markets Create Incentives For
Pollution Control? Ecological Economics, 26, 31-41.

Laplante, B., & Lanoie, P. (1995). Market Response to Environmental Incidents in

Canada: A Theoretical and Empirical Analysis. Southern Economic Journal, 60(3),
657-672.

296

Laroche, M., Kim, C., & Zhou, L. (1996). Brand Familiarity and Confidence as
Determinants of Purchase Intention: An Empirical Test in a Multiple Brand
Context. Journal of Business Research, 37(October), 115-120.

Lavelle, T. J. (1994). Federal Agencies in the Realm of Pollution Prevention and
Community Right-to-Know. Federal Facility Environmental J ustice, 5(4), 489-510.

Lavine, H. J ., Huff, W., Wagner, S. H., & Weeney, D. (1998). The Moderating Influence
of Attitude Strength on the Susceptibility to Context Effects in Attitude Surveys.
Jom'nal of Personality and Social Psychology, 75, 359—373.

Lee, C. (1993). Beyond Toxic Wastes and Race, Confronting Environmental Racism:
Voices from the Grassroots (pp. 41-52). Boston, MA: Bullard, Robert D., South
End Press.

Leeming, F. C., Dwyer, W. O., & Bracken, B. A. (1995). Children's Environmental
Attitude and Knowledge Scale: Construction and Validation. The Journal of
Environmental Education, 26(3), 22-31.

 

Ir

Livemois, J ., & McKenna, C. J. (1998). Truth or Consequences: Enforcing Pollution
Standards with Self-Reporting. Journal of Public Economics.

Lynn, F. M. (1990). Public Participation in Risk Management Decisions: The Right to
Define, the Right to Know and the Right to Act. Risk-Issue Health SAF., 1(2), 95-
101.

Mackenzie, S. B., & Lutz, R. J. (1989). The Role of Attitude Toward the Ad as a
Mediator of Advertising Effectiveness: A Test of Competing Explanations. Journal
of Marketing Research, 23(May), 130-143.

MacKinlay, A. C. (1997). Event Studies in Economics and Finance. Journal of
Economics Literature, 35(1), 13-39.

McKinney, M. J. (2000). What Do we Mean by Consensus? In P. Brick & D. Snow & S.
V. D. Wetering (Eds), Crossing the Great Divide (pp. 33-41). Washington DC:
Island Press.

Magat, W. A., & Viscusi, W. K. (1990). Effectiveness of the EPA's Regulatory
Enforcement: The Case of Industrial Effluent Standards. Journal of Law &
Economics, 33, 331-360.

 

Magat, W. A., & Viscusi, W. K. (1992). Inforrfltional Approaches to Regplation.
Cambridge, MA: The MIT Press.

Malik, A. S. (1990). Markets for Pollution Control When Firms are Non-Compliant.
Journal of Environmental Economics and Management, 18(2), 97-106.

297

 

Maloney, M. P., Ward, M. P., & Braucht, G. N. (1975). A Revised Scale for the
Measurement of Ecological Attitudes and Knowledge. American Psychologigt, 30,
787-790.

Mani, M., Pargal, S., & Huq, M. (1996, November). Does Environmerg Regulation
Matter? Determinants Of The Location Of New Manufifacturingjlants In India_In_
12241, [No. 1718 PRD Working Paper]. The Policy Research Department of World
Bank. Available: www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Marcinowski, F., & Napolitano, S. (1993). Reducing the Risks from Radon. Air & Waste,
43, 955-962.

 

McCoy, C. (1998, 11 February). Two US. Members of Mitsubishi Group and
Environmental Activists Resch Pact. Wall Street Journal, pp. A8.

McGuire, W. J. (Ed.). (1969). The Nature of Attitudes and Attitude Change (2nd ed.).
Cambridge, MA: Addison-Wesley.

Mellor, M. (1993). Building a New Vision: Feminist, Green Socialism. In R. Hofi'ichter
(Ed.), Toxic Struggles: The Theory and Practice of Environmental Justice (pp. 36-
46). Philadelphia, PA: New Society Publishers.

Miller, D., & Ratner, R. K. (1998). The Disparity Between the Actual and Assumed
Power of Self-Interest. Journal of Personalijy and Social Psycholggy, 74, 53-62.

Mishra, G. K., Newman, D. P., & Stinson, C. H. (1997). Environmental Regulations and
Incentives for Compliance Audits. Journal of Accountiag and Public Policy. 16(2),
1 87-214.

Mitchell, A. A., & Olson, J. C. (1981). Are Product Attribute Beliefs the Only Mediator
of Advertising Effects on Brand Attitude? Journal of MarketingResearch,
18(August), 318-332.

Moscovitz, D. (1993). Green Pricing: Why Not Customer Choice? The Electricity
Journal, 6(8), 42-49.

 

Muller, T. E., & Taylor, W. (1991). Everybody Talks about the Environment: But How
Environmentally Responsive are Consumers? Marketing-Proceedings of the ASAC
Conference, 12(6), 202-211.

Muoghalu, M. I., Robison, H. D., & Glascock, J. L. (1990). Hazardous Waste Lawsuits,
Stockholder Returns, and Deterrence. Southern Economic J oumal(October 1990),
357-370.

Nash, J., & Ehrenfeld, J. (1996). Code Green: Business Adopts Voluntary Environmental
Standards. Environment(January/February), 16-20, 36-45.

298

 

Naysnerski, W., & Tietenberg, T. (1992). Private Enforcement. In T. H. Tietenberg (Ed.),

Innovation in Environmental Policy (pp. 109-136). Cheltenham, UK: Edward Elgar.

 

Naysnerski, W., & Tietenberg, T. (1992). Private Enforcement of Federal Environmental
Law. Land Economics, 68(1), 28-48.

Newell, S. J. (1993). Developing a Measurement Scale and a Theoretical Model Defining
Corporate Credibility and Determining Its Role as an Antecedent of Consumers'
Attitude toward the Advertisement. Unpublished Doctoral Dissertation, Florida
State University, Tallahassee.

Newhouse, N. (1990). Implications of Attitude and Behavior Research for Environmental
Conservation. The Journal of Environmental Education, 22(1), 26-32.

Noah, L. (1994). The Imperative to Warn: Disentangling the "Right to Know" fi'om the
"Need to Know" about Consumer Product Hazards. Yale Journal of Regplation,
11(2), 293-400.

Noye, C., & Das, M. (1993). Does Concerns for Environment Translate into
Environmentally Friendly Purchase Behavior? Proceedings of the ASB Conference,
212-223.

O'Conor, D. (1995). Managing the Environment with R4apid Industrialisation: Lessons
from the East Asiam Experience. Paris: OECD (Development Centre of the
Organisation for Economic Cooperation and Development).

OECD. (1991). Environmental labeling in OECD Countries. Paris: Organization for
Economic Cooperation and Development.

Osgood, C. E., & Tannenbaum, P. H. (1955). The Principle of Congruity in the
Production of Attitude Change. Psychological Review, 62, 42-55.

Ottman, J. (1993). Industry’s Response to Green Consumerism. Journal of Business
Strategy, 13, 3-7.

Pargal, S., Hettige, H., Manjula Singh, & Wheeler, D. (1997, July). Formal and Informal
Regplation of Industrial Pollution: Comparative Evidence from Indonesia_ and the
United States, [No. 1797 PRD Working Paper]. Policy Research Department of
World Bank. Available: www.worldbank.org/nipr/work_paper/index.htrn [2000,
4/26].

Pargal, S., & Wheeler, D. (1996). Informal Regulation of Industrial Pollution Developing
Countries: Evidence from Indonesia. Journal of Political Economy, 104
(December)(6), 1314.

Pattern, D. M. (1998). The Impact of the EPA's TRI Disclosure Program on State

Environmental and Natural Resource Expenditures. Journal of Accounting and
Public Policy, 17(winter)(4,5), 367-382.

299

 

rpa _. ..

 

Paulhus, D. L. (1984). Social Desirable Responding: Some New Solutions to Old
Problems. In D. M. Buss & N. Cantor (Eds), Persoaality Psychology: Recent
Trends and EmergingDirections (pp. 201-209). New York: Springer Verlag.

Pearce, D. W., Markandya, A., & Barbier, E. B. (1989). Blueprint for a Green Economy.
London: Earthscan Publications.

Pearce, D. W., & Turner, R. K. (1990). Economics of Natural Resources and the
Environment. Hertfordshire, UK: Harvester Wheatsheaf.

Pease, W. S. (1991). Chemical Hazards and the Public's Right to Know: How Effective Is
California's Proposition 65? Environment, 33(10), 12-20.

Perman, R., Ma, Y., & McGilvray, J. (1996). Natural Resource & Environmental
Economics. New York: Longrnan.

Prince, J. (1991). Whittling Down Waste. Progressive Grocer, 70, 41-44.

PROPER-PROKASIH Team, BAPEDAL, Jakarta, PRDEI, & Bank, W. (1995,
November). What is Proper? Reputational Incentives for Pollution Control in
Indonesia. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Protess, D. et al. (1987). The Impact of Investigative Reporting on Public Opinion and
Policy Making: Targeting Toxic Waste. Public Opinion Ouaterly, 51(2), 166-185.

Roddy, G., Cowan, C. A., & Hutchinson, G. (1996). Consumer Attitudes and Behavior to
Organic Foods in Ireland. Journal of Intemationagionsumer Marketing, 9(2), 41-
63.

Roper Organization Inc. (1990). The Environment: Public Attitudes and Individual
Behavior.

Rosendahl, I. (1990). Retailers Joining Fight to Clean Up Environment. Drug Topics,
134,(6).

 

Rosenthal, R., & Rosnow, R. L. (1991). Essentials of BehaLioral Research: Methods a_pc_l
Data Analysis (2nd ed.). New York: McGraw-Hill.

Russell, C. S. (1990). Monitoring and Enforcement. In P. R. Portney (Ed.), Public
Policies for Enviromentflrotection (pp. 243-274). Washington DC: Resources
for the Future.

Segerson, K., & Tietenberg, T. (1992). The Structure of Penalties in Environmental
Enforcement: An Economic Analysis. Journal of Environmental Economics and
Management, 23(2), 179-200.

300

 

 

Shane, P. B., & Spicer, H. H. (1983). Market Response to Environmental Information
Produce Outside the Firm. The Accounting Review, LVIII, 523-538.

Shepard, B. H., Hartwick, J ., & Warshaw, P. R. (1988). The Theory of Reasoned Action:
A Meta-Analysis of Past Research with Recommendations for Modifications and
Future Research. Journal of Consumer Research, 15(December), 325-343.

Shridhar, P. (1996, 4-8 Jan 1996). The Right Research: Measuring the Success and
Effectiveness of Public Information Progzams Paper presented at the Conservation
'96: Responsible Water Stewardship, Orlando, FL.

 

Smith, R. E., & Swinyard, W. R. (1983). Attitude-Behavior Consistency: The Impact of
Product Trial Versus Advertising. Journal of Maaketing Research, 20(August), 257
267.

Solomon, M. R. (1992). Consumer Behavior. Boston: Allyn and Bacon.

Soros, G. (1998). The Crisis of Global Capitalism: Open Society endangered. New York:
Public Affairs Information Service.

 

Spano, S. (2001). Public Dialogue aard Participatory Democracy. Cresskill, NJ: Hampton
Press, Inc.

Stammer, L. (1997, 9 November). Hanning the Environment Is Sinful, Prelate Says. w
Angeles Times, pp. 1A.

Steiner, R. L., & Bamhart, R. B. (1972). The Development of an Instrument to Assess
Environmental Attitudes Utilizing Factor Analytic Techniques. Science Education,
56(3), 427-432.

 

Stephens, T. (1994). The Concept of Environmental Justice. Sugar Law
Center(December), 9-1 0.

Sternthal, B., Tybout, A. M., & Calder, B. J. (1994). Experimental Design:
Generalization and Theoretical Explanation. In R. P. Bagozzi (Ed.), Principle of
Marketing Research (pp. 195-223). Cambridge, MA: Blackwell Publishers.

Tietenberg, T. (1985). Emissions Trading: An Exercise in Reforming Pollution Policy.
Washington, DC: Resources for the Future.

Tietenberg, T. (1990). Economic Instruments for Environmental Regulation. Oxford
Review of Economic Policy, 6(1), 17-34.

Tietenberg, T. (1995). Design Lessons from Existing Air Pollution Control Systems: The
United States. In S. Hanna & M. Munasinghe (Eds), Property Rights in a Social
and Ecological Context: Case Studies and Design Applications. Washington, DC:
The World Bank.

301

Tietenberg, T. (1996). Private Enforcement of Environmental Regulations in Latin
America and the Caribbean: An Effective Instrument for Environmental
Management? Washington, DC: Inter-American Development Bank.

Tietenberg, T. (1998). Disclosure Strategies for Pollution Control. Environmental and
Resource Economics, 11, 578-602.

Tietenberg, T. (1998, October). Tradable Permits and the Control of Air Pollution in the
United State. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/newappr.htm [1998, October].

Tietenberg, T., & Wheeler, D. (1998, October 23-25). EmpoweringThe Community:
Information Strategies For Pollution Control. Paper presented at the Frontiers of
Environmental Economics Conference, Airlie House, Virginia.

Wallace, P. E. (1993). Disclosure of Environmental Liabilities under the Securities Laws:
the Potential of Securities-Market-Based Incentives for Pollution Control 50.
Washington & Lee Law Review, 50(Summer), 1093.

Wang, H., & Wheeler, D. (1996, September). Pricing Industrial Pollution in China: An
Econometric Analysis of the Levy System (#1644 PRD working paper). The Policy
Research Department of World Bank. Available: www.
worldbank.org/nipr/work_paper/index.htm [2000, 4/26].

Wapner, P. (1995). In Defense of Banner Hangers: The Dark Green Politics of Green
Peace, Ecolagical Resistance Movements (pp. 300-314): State University of New
York Press.

Wells, W. D. (1961). The Influence of Yeasaying Response Style. Journal of Advertisiag
Research, 1(June), 1-12.

Wheeler, D. (1992, January). The Economics of Industrial Pollution Control. World
Bank. Available: www.worldbank.org/nipr/work_paper/index.htrn [2000, 6/ 15].

Wheeler, D. (1997, June 27). Information in Pollution Maaagement: The New Model,
[No. 16635-BR]. Policy Research Department of World Bank. Available:
www.worldbank.org/nipr/work_paper/index.htrn [2000, 4/26].

Wheeler, D. (1999). Greeninalndustry: New Poles for Communities, Markets, and
Government. New York: Oxford University Press.

Wheeler, D., & Afsah, S. (1996). Going Public on Polluters in Indonesia: BAPEDAL's
PROPER PROKASIH Program, East Asian Executive Reports (V 01. May 1996).
Washington, DC.: International Executive Reports.

White, L. (1967). The Historical Roots of Our Ecological Crisis. Science, 155, 1203-
1207.

 

302

Williams, M. (1992). Environmentally Safe can Enhance Sales. Advertising Age, 63, 8-9.

Wimmer, R. D., & Dominick, J. R. (1994). Mass Media Research (4th ed.). Belmont,
CA: Wadsworth Pub. Co.

Winters, L. C. (1989). Does it Pay to Advertise to Hostile Audiences With Corporate
Advertising? Journal of Advertising Research(June/July), 11-18.

Woodward, K., & Nordland, R. (1992, 30 November). New Rules for an Old Faith.
Newsweek, 71.

World Bank. (1998, October 02, 1998). New Ideas in Pollution Regalation (NIPR).
Policy Research Department of World Bank. Available:
http://www.worldbank.org/nipr/ [2000, 4/26].

World Business Council on Sustainable Development. (1998). Pragmatism Is the Driving
Force. Tomorrow, 8(6), 41.

Yacoob, M., Brantly, E., & Whiteford, L. (1996). Public Participation in Urban
Environmental Management: A Model for Promoting Community-Bag
Environmental Management in Peri-Urban Areas (R0103; WASH-TR-90): Camp,
Dresser, and McKee International, Inc, Arlington, VA ; Peace Corps, Washington,
DC Information Collection and Exchange Div.

Yi, Y. (1990). Cognitive and Affective Priming Effects of the Context for Print
Advertisements. Journal of Advertising, 19(2), 40-48.

Zucarro, C., & Fortin, D. (1992). Reassessing the Socially Responsible Consumer.
Marketing-Proceedings of the ASAC Conference, 13(6), 209-217.

303