‘vlw‘T I IIIIIIIICI ' I "- ' ['3' I: III‘. "MI 'I'II‘I‘II. I II:

I WWI RlIl ll I I IlllIlg l: ' UPI l ll” lIIIlIl‘IlIIIII I ‘I'n- "1’" . .
llllll l .‘ . ' 2...: '1 ‘I'l-IVI 'IIIWfiI-I‘luv.‘}l"w , f 7 II". n 1
I‘ ‘I "I :I‘I’II'IIjIzy “ . ‘

I’II III‘é-II

      
  
 

'II'II IlllII "IlllI‘II III .

  
    
 
  
  
  
          

III, III 'I'4I Ifl‘é“ : 'iil 1; z':
III‘IIlIIIIllIII III'I I;II. IIIIluI iIIIIII II.I.I.'... II
IIIllellllIl‘IuIl IllIIIl’IIg-II'I-III. 3: ’r .. .
«III? III Ids“ . :_. IIIIIIII‘l-IIIEII‘II}???é
IIIII'IIII'IIIIIIIIIII III III I, ,IIIIIIIIIIIII
I ‘I

'< .o-' 4 a
Lumm‘
I. .,.-.A.-.—

    
   

-
.._o...:‘

”up
0

‘ 35- "

ac a

4 v-

4

o
1.1. ‘.:.-;’~:-é—‘ ."0 ‘:
1523's“ 9;.— 5
. .

IIIT'III "I II

- , I: '9;
IIILIII' l "I": ”3' u“ 3’

. . Ill! 3.
I . ' ‘I o $ .. 7.;
' . .“MIII‘III'EI-I '. IIIWI‘IlIl I-I‘Q'f' I HI'
I 9‘3“: I‘ll)“ I: .1. I”. E‘Il 'EIIIE II” I ”If!” ”A "IE I ' II" III?"

.
~ ..
.24

...4 |-‘

       
   
 
 
 

a

— v'
.p‘.—.
M
~_—.

  

          
      
  

...
:.
‘7’“:
_.,.
.—
..
...."”
;:.
.-...
3......_
‘22:;
t...—
L...
_ _
“'1” .2.
'- '-::’__'
_ _ .__.. -
w- ..-'r‘:"“:."-...' -.
W; -
:72”, ,. :3, .
:37.“
“W

ElII- I M}? ‘ '
II, llyl‘IlT‘IllIl‘ IllIJIBI IIIIII “SUI
II mfg-“Ar II JIIfiItI {Ill-I?! LIVE-:3 {Ifi‘lig‘dh
I 1me In M IIII‘IIII III:
'III; I IIIIIIII ”III III], III. .II IIIIHIW

III" I‘IIII‘.‘IIII II
. .. I

   

III

..' I; (IIIII‘I ' I IIIv" I I ‘II

I' ' .. -'..I- I III:‘II9III.',.':71.1"“I"IIISI'I‘Ii . ‘I §1II IIIIII' WI

. . . II ,lIIIII'II‘I".”I;.|'.;“IZ"Ilu‘ligl‘llv, 5%; III I? 'III HIII I III-l
; 'I“ 'II ‘“III'Il' IIJII Ill“ 3": L'lIIIIil lIlll ll’ .‘ 'llell I
‘ I'III'I'I .III III," MN“ I
.. III II II I ,. ‘
IlIIlII “I I‘ll III Il; III [IL {I‘III ‘I.-1 I" ZI‘ICI I,
III I III! ”II lIll I'I-llI IIIIIII llIlIIIIlI l" til II'Ij‘I' ' :
‘IIIIIIII II III IIIIII III IIIIIIII IN E‘ ‘1'.
II III III] III
III ‘IIII. II III IIIIIII" IIIIIIII‘I'I III
IIHI II: III! III'lIl IIIIIIIIIIIIII”I

AMI
ll. I II‘III'I “III IIIIII II
III; IlIlIll IIIIZI III'III II {III IIIIIIIII IIIIII IIIIIIIII

lIIrI‘II IIII III IIIIE- IIIII'I.

   

l-Y IIIII .E‘I

, II
III!

 

WALK;.:'II 2M1. . _._;_,;{‘-_; -. l “l _\v: E m.
. .- ‘ Ell k

YW’???‘ I" «ml-ma '

“ll“l“ll“‘ll‘l‘l‘ll“‘L“L“l“l“‘l «I

 

 

3 1293 l xivcm a}?

This is to certify that the

dissertation entitled

A HIERARCHICAL MODEL OF THE
CAUSAL RELATIONS BETWEEN ENVIRONMENTAL
ATTITUDES, BELIEFS AND BEHAVIOR:
WITH SPECIFIC DATA RELATING TO THE

RECYLING OF BEER CONTAINERS
presented by

Kadayam H. Padmanabhan

has been accepted towards fulfillment
of the requirements for

Ph 0 D. degree in Marketing

 

 

 

 

WMf‘iajOl' pfOiZCSSOYg

Date 7-24-81

MS U is an Affirmatiw Action/Eq ual Opportunity Institution 0-12771

. 5-. ur-

 

IV‘ESI.) RETURNING MATERIALS:

Place in book drop to
LIBRARJES remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

"*3 64 '
w... ”Twawm

 

 

 

 

 

 

 

A HIERARCHICAL MODEL OF THE
CAUSAL RELATIONS BETWEEN ENVIRONMENTAL
ATTITUDES, BELIEFS AND BEHAVIOR:
WITH SPECIFIC DATA RELATING TO THE

RECYCLING OF BEER CONTAINERS

By

Kadayam H. Padmanabhan

A DISSERTATION

Submitted to
Michigan State university
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Marketing and
Transportation Administration

1981

ABSTRACT

A HIERARCHICAL MODEL OF THE
CAUSAL RELATIONS BETWEEN ENVIRONMENTAL
ATTITUDES, BELIEFS AND BEHAVIOR:
WITH SPECIFIC DATA RELATING TO THE
RECYCLING OF BEER CONTAINERS

By

Kadayam H. Padmanabhan

The relationship between attitudes and between attitudes and
behavior is complex. This complex relationship has been studied in
the form of a hierarchical model by Hunter, Levine, and Sayers (1976).
The model hypothesizes that attitudes or concepts are logically
related to each other, and arranged hierarchically from the most
concrete and specific to the most abstract and general. Poole and
Hunter (1980) extended the hierarchical model to a theory of the
attitude behavior relationships by postulating that specific behaviors
are causally attached to the specific attitudes at the bottom of
attitude hierarchies.

The purpose of this research is to explore the relationship
between attitudes such as a belief that we suffer energy crisis and
a specific behavior, viz. frequency of returning empty beer containers.
This is done by testing mediation and network hypotheses underlying
the hierarchical model.

The population studied is Greater Lansing, and included both
students and non-students. Data was collected through mail survey.

Appropriate belief and behavior scales were formed by grouping

Kadayam H. Padmanabhan

individual questionnaire items on the basis of content similarity,
internal and external consistencies. A confirmatory factor analysis
was conducted to check the groupings. Also computed were the means,
standard deviations, and reliabilities. Correlations between scales
were computed, corrected for attentuation due to error of measurement.
Path analysis was performed on the belief and behavior correlations.
matrix using PATHPAC (Hunter and Hunter, lé77).

The path model fitting the data differed only in trivial ways
from the model predicted by the hierarchical theory. Causal links
follow logical relations. The influence of higher order environmental
beliefs on behavior is indirect and mediated by more specific environ-
mental beliefs. Causal chains run from general environmental beliefs
to specific environmental behavior.

Findings confirmed the prediction that beliefs at the top of
the hierarchy are adopted before beliefs lower down; that beliefs at
increasingly lower levels, i.e. beliefs that are more specific will
have increasingly lower means.

The major conclusions are: (l) Causal relations between
environmental beliefs have a hierarchical structure stemming from the
belief that resource shortage is real at the top and flowing down to
a large number of more specific beliefs. (2) Causal influence among
environmental beliefs goes from top down; from the most abstract and
general belief to more concrete and specific beliefs and finally from
highly specific beliefs to the adoption of corresponding behaviors.

The hierarchical model is shown here as a theory of the

relationship between general and specific beliefs, and between beliefs

Kadayam H. Padmanabhan

and behavior. Study upholds the hierarchical model prediction that
attitude-behavior correlations are highest if they are measured at

an equivalent level of generality. Specific behavior (i.e. frequency
of returning empty beer containers) is best predicted by correspond-
ingly specific environmental beliefs. The study stresses the need
rto understand the hierarchy of general and specific beliefs if we
wish to understand how behavior choice occurs.

This is the first attempt at developing a broad network of
environmental attitudes in the form of a hierarchical model and
tracing the mediating causal links between abstract environmental
attitudes and specific environmental behavior. General environmental
beliefs form the basis for more specific beliefs which in turn
determine behaviors. Campaign messages must incorporate the mediating
beliefs which will enable change at the higher level to induce change
at the lower level.

In regard to marketing strategy, study stresses the need to
create goodwill for the product as a whole rather than one given brand,
and for the company as a whole rather than one given product. This
will enable companies to fully exploit the wide-ranging impact of

causally prepotent beliefs regarding industry and company as a whole.

Copyright by
KADAYAM H . PADMANABHAN

1981

ACKNOWLEDGEMENTS

A number of persons contributed toward the completion of this
dissertation.

Dr. Donald Taylor, as member of the dissertation committee,
offered numerous suggestions in regard to'the overall conduct of
research and especially in the area of sampling. His suggestions
were incorporated in the research. I am also grateful to Dr. Taylor
for the understanding and encouragement he provided me in his capacity
as Chairman of the Department of Marketing.

Dr. Gilbert Harrell, as co-chairman of the dissertation
committee, led me firmly into the field of consumer behavior theory.
At his insistence, bulk of the early drafts were rewritten so would
conform to a theoretical mold--and with good result. I am grateful
to Dr. Harrell for welcoming my research ideas with friendliness and
openness.

Dr. John Hunter is one of a kind. As co-chairman of the
dissertation committee, he has been my guiding light throughout the
conduct of this research. His enormous intellectual discrimination,
formidable analytical skills and total dedication in guiding me
have all been an inspiration and a joy.

I thank Clyde Anderson and Larry Chapin for assisting me
through computer work. They helped me along even as I made

inconvenient and frequent demands on their time.

ii.

I wish to thank Judy Gilroy for her excellent typing assis-
tance. She completed the work with efficiency and dispatch.

I owe to my wife, Vasantha and daughter, Priya for completing
this dissertation and more. Their continuous love and care guided me
emotionally through my long year in school, notwithstanding occasional
‘naggings nearness seems to bring. If my labor has been worthwhile,

their love has been rewarded.

iii

LIST OF

LIST OF

Chapter

I.

//II. GENERAL ATTITUDES, SPECIFIC

III.

IV.

TABLES . .

FIGURES

AN OVERVIEW

INTRODUCTION

Background
Hierarchical Model

TABLE

Purpose of Research . .
Importance of Research

OF CONTENTS

HIERARCHICAL RELATION TO BEHAVIOR . .

Outline .

Role of Attitude Concept

ATTITUDES,

General Attitudes, Specific Attitudes, and
Their Relation to Behavior . . .
Hierarchical Structure of Attitudes . . . . . .

Summary .

ENVIRONMENTAL ATTITUDES, ENVIRONMENTAL BEHAVIOR
AND CONTAINER RETURNS . .

Environmental Attitudes and

Environmental Behavior

Hierarchical Model and Adoption of
Environmental Beliefs
Beer Containers and the

Summary .'

SCOPE AND METHODOLOGY . .

Outline .

Scope of Research . . .
Methodology of Research

Environment

iv

Page
vi.

vii

U'IbUJN

\l

11
15

16

16

20
23
29

30
30

3O
31

Chapter Page

v. RESULTS AND DISCUSSIONS . . . . . . . . . . . .‘. . . . . 37
Overview . . . . . . . . . . . . . . . . . . . . . . . _ 37
Intensity, Direction and Distribution of

Environmental Beliefs and Behavior . . . . . . . . . 37
Testing the Hierarchical Model . . . . . . . . . . . . 42
Testing the Mediation Hypotheses . . . . . . . . . . . 49

Testing the Network Hypotheses . . . . . . . . . . . . 54
Impact of General vs. Specific Environmental .
Attitudes on Environmental Behavior in the

Context of Hierarchical Model . . . . . . . . . . . . S9
Relation Between Present Findings and -
Prior Research . . . . . . . . . . . . . . . . . . . 60
VI. SUMMARY OF CONCLUSIONS . . . . . . . . . . . . . . . . . 63

Intensity, Direction, and Distribution of
Environmental Beliefs and Container

Return Behavior . . . . . . . . . . . . . . . . . . . 63
Testing Mediation Hypotheses . . . . . . . . . . . . . 64
Testing Network Hypotheses . . . . . . . . . . . . . . 66
General vs. Specific Attitudes in the Context '

of the Hierarchical Model . . . . . . . . . . . . . . 67
Comparison with Prior Studies . . . . . . . . . . . . . 67
The Hierarchy Model and Container

Return Behavior . . . . . . . . . . . . . . . . . . . 68

VII. IMPLICATIONS AND RECOMMENDATIONS . . . . . . . . . . . . 7O

Implications for Container Recycling . . . . . . . . . 70
Optimal Environmental Education Campaigns . . . . . . . 7l
Implications for Marketing Decisions . . . . . . . . . 74
APPENDICES
Appendix
A. Questionnaire . . . . . . . . . . . . . . . . . . . . . . 76
B. Grouping of Questionnaire Items . . . . . . . . . . . . . 87
C. Factor Intercorrelations and Loading Matrix . . . . . . . 91

BIBLIOGRAPIH . O O O O O O O O O O O O O O O O O O O O O O O O O 108

LIST OF TABLES

Table J. ,Page
1. Belief and Behavior Scales . . . . . . . . . . . . . . 39
2(a). Observed Scale Correlations . . . . . . . . . . . . . . 43
2(b). PrediCted Scale Correlations . . . . . . . . . . . . . - 45
2(c). Observed Minus Predicted Scale Correlations . . .i. . . 46

vi

LIST OF FIGURES

Figure ; Page
1. The Hierarchical Structure of Concepts Associated

With war in General . . . . . . . . . . . . . . . . . 13

2. Path Model . . . . . . . . . . . . . . . . . . . . . . 48

vii

£15. OVERVIEW

The purpose of this research is to explore the relationship
~between abstract attitudes such as a belief that we suffer energy
crisis and specific behaviors such as returning beer cans. The
guiding hypothesis is that such correlatiOns will be low because the
relationship between abstract attitudes and behavior is mediated by
intervening attitudes of more specific nature. It is hypothesized
that there is a hierarchy of attitudes that causally mediates relations
between more abstract attitudes and less abstract attitudes and
between attitudes and behavior. The purpose of the empirical study
is twofold: (a) to lay out in detail the entire network of environ-
mental beliefs relevant to a particular environmental behavior,

returning beer containers, and (b) to test the mediation hypothesis

using path analysis.

(\

 

CHAPTER I g

 
   
  

[ariety of phenomena. While

, .attitudes are often the

es factors shape behavior.
e 914 to'structure of
trolggvio

   
 

a central position in the

r - 5.."
field of applied psychology as well as in the marketing literature.
I

Its importance in marketing theory and practice is underscored by
the pivotal role it occupies}n the major des'criptive models of

. - C
cansur‘behavior. The notign that we tend to buy those products

c. 9
we like aurwoyl @050 we dislike is a truism, borne out by our

direct experience as well as convincing evidence from marketing
studies.
The relationship between attitude and behavior is sometimes
complex. Apparent inconsistenciss often develop between attitudes
and behavior because studies fail to recognize the complexities
involved--in how attitudes are assessed, the conditions in which
behavior is undertaken and the societal constraints governing behavior.
A frequently debated question is whether behavior is predicted
better by general attitudes or by specific attitudes. Fishbein (1973)

suggested that attitude and behavior would show a strong relationship

U

 

9|

when both are measured at an equivalent level of generality. Crespi
(1971) and Weigel et a1. (1974) went further to propose that attitude-
behavior consistency is high when both measures are highly specific.
The general vs. specific attitudes debate can be most meaningfully
resolved through an understanding of how attitudes (both general and

~specific) are themselves related.

Hierarchical Model

 

The complex relationship among attitudes is highlighted by
the presence of attitudes with varying levels of generality or
specificity. This complex relationship has been studied in the form
of a hierarchical model by Hunter, Levine, and Sayers (1976). The
model primarily examines the properties of a set of attitudes toward
a set of concepts that is organized in the form of hierarchical chains.
The attitude concepts are hypothesized as being logically related to
each other, arranged at different levels from the most concrete and
specific to the most abstract and general. Poole and Hunter (1980)
extended the hierarchical model to a theory of the attitude behavior
relationship by postulating that specific behaviors are causally
attached to the specific attitudes at the bottom of attitude
hierarchies. The extended hierarchical model predicts the exact
relationships between general and specific attitudes on the one hand
and between attitudes and behavior on the other.

According to the hierarchical model, the correlations between
a general attitude and a specific behavior is mediated by intervening

attitudes of a more specific nature. Each step in the chain reduces

the level of correlation in accordance with the product rule of

path analysis. Thus a general attitude will not be highly correlated
with any particular behavior. However, a general attitude is
superordinate to many attitudes leading to many specific behaviors.
Thus it will be correlated with a large number of specific behaviors.
‘If an index of the many logically subordinate specific behaviors

were formed, then that index would have a high correlatiOn with the
general attitude. This has been confirmed by the studies of Fishbein

and Ajzen (1974), and Tittle and Hill (1967).

Purpose of Research

 

The purpose of this research is to test the hierarchical
model by studying the relationship between a particular environmental
behavior and a set of environmental attitudes. We will:

(1) Lay out the concrete network of interrelated attitudes
for a particular environmental behavior: container return

(2) Test the network hypotheses implied in the hierarchical
model

(3) Test the mediation hypotheses emerging from the
hierarchical model

We will measure the direction, intensity, and distribution
of consumer attitudes regarding environment in general and beer
container returns in particular. These attitudes will then be
related to the frequency with which beer containers are returned,

and the distribution of such behavior among consumers.

In testing mediation hypotheses, the following two major
propositions will be examined:

(a) That the causal impact of environmental beliefs follows
logical relations

(b) That the influence of higher order beliefs on behavior
-is indirect and mediated by more specific beliefs

In testing network hypotheses, the following two major
propositions will be examined: I

(a) That causal relations between environmental beliefs
are structured in a set of hierarchical relationships

(b) That the causal structure of environmental beliefs is
arranged from most abstract and general to the most concrete and

specific

Importance of Research

 

This represents the first attempt at developing a broad
network of environmental attitudes in the form of a hierarchical
model and tracing the mediating causal links between abstract
environmental attitudes and specific environmental behavior. Do the
various attitudes and behaviors form an integrated system? HOw
favorable are consumers toward environmental conservation? Does an
orientation to environmental conservation increase the frequency of
returning empty containers?

Identification of superordinate and subordinate attitudes
can go a long way in clarifying the basis and direction for an

effective environmental education program. It is especially helpful

in developing a long-term approach to environmental education.
In the long run, environmental education based on entire network
of related environmental attitudes is likely to be both more

effective and efficient than one based on isolated environmental

behaviors.

CHAPTER II

GENERAL ATTITUDES, SPECIFIC ATTITUDES,
AND THEIR HIERARCHICAL RELATION
TO BEHAVIOR
Outline

This chapter briefly outlines the role of attitude as a
leading concept in the field of applied psychology, and the nature
of the relationship between attitude and behavior. Second, it
describes a hierarchical attitude model in some detail, along with
research support in other areas.

The chapter closes with a summary propositions emerging from
attitude-behavior studies and specifically from the hierarchical

conceptualization of the attitude-behavior relationship.

Role of Attitude Concept

 

Gordon Allport began his classic paper on attitudes in the

1935 Handbook of Social Psychology with this statement: “The concept

 

of attitude is probably the most distinctive and indispensable concept
in contemporary American social psychology." "Perhaps," comment
Murphy et al. (1937), "no single concept within the whole realm of
social psychology1 occupies a more nearly central position than that

of attitudes."2 Indeed, writers like Bogardus (1931), Folsom (1931),

 

1Thomas and Znaniecki (1918) have been credited for the
establishment of the concept of attitude as a permanent and central
feature of social psychology.

2Murphy et al., Experimental Social Psychology
(New York: Harper, 1937), p. 889.

and Thomas and Znaniecki (1918)3 went so far as to define social
psychology as the scientific study of attitudes.

The concept is central in other disciplines too.

" . . . the attitude concept has come to play an important part in
all of the behavioral sciences and many of the applied disciplines."
A(Fishbein, 1967, p. v). Perhaps, no other concept from the behavioral
sciences has been used so widely by theoriSts and researchers as

the term attitude.

The study of attitude is well-entrenched in marketing theory
and practice. The acceptance by theorists is evident in the pivotal
role that the concept of attitude plays in the major descriptive
models of consumer behavior. Attitudes can directly affect purchase
decisions and experience with a product or service can shape an
attitude. Purchase decisions are based upon attitudes existing
at the time of purchase.

\)57/ Janis and Hoffman (1970) Observed a high relationship between
attitude toward smoking and extent of smoking behavior. Similar
results were found by Veevers (1971) for attitude toward drinking
alcoholic beverages and drinking behavior; by Kothandapani (1971)

in regard to attitude toward personal use of birth control methods

and use (or nonuse) of birth control methods; by Fishbein and Coombs

(1974) in regard to attitudes toward presidential candidates and

 

3E. S. Bogardus, Fundamentals of Social Psychology, 2nd ed.,
(New York: Century, 1931), p. 444; J. K. Folsom, Social Psychology
(New York: Harper, 1931), p. 701; and Thomas and Znaniecki (1918),
p. 526.

voting behavior in presidential election; Albrecht et a1. (1972)
in regard to attitude toward legalization of marijuana and petition
signing.

It has also been reported that good commercial affects both
attitude and behavior (Lair, 1965); that attitudes toward trading
-stamp usage are reflected in trading stamp usage (Udell, 1965);
and that the use of price as an evaluative criterion in cOnvenience
goods purchases can be predicted by an attitude battery (Craig, 1971).
Studies have also consistently demonstrated that the affective
dimension is significantly more effective in explaining variance
in market share than the cognitive dimension.

Reviews by Ajzen and Fishbein (1977), Schuman and Johnson
(1976) and Kelman (1974) all concluded the relationships of at least
moderate strength are the rule rather than the exception.

\/k:::;/// The relationship between attitude and behavior is complex.
Attitudes and behavior are influenced by other attitudes and beliefs,
by situational constraints, and societal expectations or norms.
Apparent inconsistencies between attitudes and behavior often arise
because the studies have not provided a refined and detailed assessment
of all relevant attitudes, the situation in which action takes place,
and the societal constraints governing the action situation.

General Attitudes, Specific Attitudes,
and Their Relation to Behavior

Fishbein (1973) observed that a strong relationship between
attitude and behavior depends on the extent to which attitude and

behavior are measured at an equivalent level of generality.

10

He suggested that when the attitude object is general or compre-
hensive, the behavior criterion should be equally general or
comprehensive.

Crespi (1971) suggested that attitude-behavior correlation
is highest when the attitude and the behavior are highly specific.
‘Weigel et al. (1974) found that the attitude with the highest-
correlation was that attitude most highly Specific to the behavior.
Liska (1974) noted that only the relevant specific attitude has a
direct causal impact; more remote attitudes have indirect effects.

Weigel et a1. (1974) assessed attitudes toward objects
ranging from general (“a pure environment") to specific ("the
Sierra Club") and later gave subjects an Opportunity to volunteer
for activities that implied different levels of commitment to the
Sierra Club. Heberlein and Black (1976) assessed purchases of
lead-free gasoline in relation to attitude scales at differing
levels of specificity. In both studies, the most general attitude
scale yielded the weakest relationship to behavior, and the most
specific scale yielded moderately high correlations.

General attitudes are not irrelevant. Studies Show that
general attitudes become indirectly relevant to a large number of
specific behaviors, though not maximally relevant to any one of
them. Consequently, while a general attitude may not strongly
affect any one specific behavior, its effect on a large number of
behaviors can be very large in total. An abstract attitude may be
a poor predictor of any particular behavior, but will be a good

predictor of an index where multiple acts are used as the criterion

11

of behavior (Tittle and Hill, 1967). Weigel and Newman (1976)

found that their measure of attitude toward environmental issues

had an average correlations of only .29 with single behaviors while
the correlation with an index of several such behaviors was .62.
Fishbein and Ajzen (1974) found that general attitudes toward objects
-exhibit a mean correlation of .14 with single behaviors contrasted_

with a correlation of .66 with a multiple behavior index...

Hierarchical Structure of Attitudes

Background

 

The structural relationship between attitudes of different
generality has been studied in the form of a hierarchical model by
Hunter, Levine, and Sayers (1976). An empirical study of such a
system by Poole and Hunter (1979) suggested only slight modifications
to that model. Poole and Hunter (1980) extended the hierarchical

model to include a theory of the attitude behavior relationship.

we}.

The basic premise of the model is that people respond to
psychological objects or concepts in three different ways: belief,
attitude (i.e., emotion or affect), and behavior. Where two concepts
are related logically, the researchers have long suspected that the
corresponding beliefs, attitudes, and behaviors will also be related.

Hunter et al. derive their model from a more general theory
of affective cognitive organization. A logical hierarchy is one of
several possible cognitive organizations, each of which could give

rise to different models of influence between attitudes. The model

12

considers the properties of a set of attitudes toward a set of
concepts that is organized in the form of hierarchical chains.
Concepts can be frequently organized into logical classes or
subclasses that form the superordinate subordinate relationship-
with each other. The concepts are arranged at different levels
Afrom the most concrete and specific to the most abstract and general.
Their main example is shown in Figure 1. .Hunter et a1. present a
specific mathematical model that will predict the course of changing
attitudes associated with concepts that are logically related to

each other in such a hierarchy.

Origin

The idea that attitudes are arranged in logical hierarchies
is suggested by two major research groups. First, several social
psychologists have proposed that attitudes are hierarchically
arranged. Allport (1937) posited an attitudinal continuum with
opinions, attitudes, interests, and values as successive points
along the continuum. McGuire (1960, 1968) has always made a strong
argument for long-run logical consistency in belief structures.
Rokeach (1969) has suggested that beliefs, attitudes, and values
are arranged hierarchically--beliefs being specific and lehst central,
and values being general and most central.

A second source of support for the hierarchical model can be
found in research on semantic memory by investigators of human
information processing. There is evidence supporting the hierarchical

nature of the storage network in semantic long-term memory (Collins

l3

 

 

 

 

 

 

 

war
I L ' l
War in . war in
Indochina Middle East
war in War in war in War in
Vietnam Cambodia Israel Yemen

Figure l.--The Hierarchical Structure of Concepts
Associated with War in General.

SOURCE: Hunter, Levine, and Sayers (1976).

l4

and Loftus, 1975; Collins and Quillan, 1972; Lindsay and Norman,

1972; Wood, 1972). Given that (a) memory often contains hierarchical
structures, and (b) feelings and emotions about concepts or experiences
are often part of a memory trace, it is not surprising that attitudes
can be viewed organizationally in the same fashion as concepts are

-stored in memory.

Assumptions

 

The hierarchical model makes the following assumptions:

(a) Attitudes can be associated with concepts found at all
levels.

(b) There is a strong downward influence in the hierarchy.
Upward influence is much weaker than downward influence; nor is there
any sideways influence. Only downward influence is recognized.

(c) One can attempt to change attitudes by presenting
independent messages about concepts at any level in the hierarchy.

(d) Attitudes can be influenced by attitudes about concepts
at higher levels in the hierarchy. In other words, we assume that
change in any given attitude is produced by two possible agents:
a direct external message or the attitude toward the concept
immediately above the given concept in the hierarchy.

(e) Belief structures exert a strong influence on attitudes,
but attitudes will not affect beliefs.

(f) Their equations consider only the case in which the

structure of belief system is given by the beginning of the time

15

period under consideration. However, their equations could be

adapted to changing structures.

Summary

Attitudes are important determinants of behavior. _Significant¥

 

in,”

euro...

correlations between_atti§g§§_é§dikgha¥inI_ha!emhBEELIERQIEEQI
many—Etgdies. However, the relationship between attitude and.
behavior is sometimes complex.

Hierarchical models explain complex relationships. Beliefs,
attitudes and behaviors are organized in the form of hierarchical
chains. The impact of attitudes on behavior is predicted from this
hierarchical structure.

The hierarchical model predicts that:

(a) Attitudes are hierarchically arranged and internal
causal influence goes down only along hierarchical channels.

(b) Behaviors are associated only with attitudes at the
bottom of the hierarchy.

(c) The influence of higher order attitudes on behaviors

is indirect and mediated by more specific attitudes.

CHAPTER III

ENVIRONMENTAL ATTITUDES, ENVIRONMENTAL :
BEHAVIOR AND CONTAINER RETURNS

Environmental Attitudes and
Environmental Behavior

 

 

Background

 

There has been a lack of commonly accepted definitions,
objectives, and mechanisms for applying research results to the
needs of environmental planning and decision making (Brookfield,
1969; Canter, 1971). However, understanding of environmental
perception and behavior has increased rapidly in recent years.
The objectives of these studies were to:

(1) Assess environmental attitudes in individuals and
groups I

(2) Identify variables that could explain observed
differences in attitudes

(3) Investigate the impact of environmental attitudes
on environmental behavior

Environmental Cognition, Attitude,
and Behavior

 

Maloney and Ward (1973) first developed an ecology inventory
to measure ecological attitudes and knowledge. After analyzing the
data on people's attitude and knowledge regarding ecology, environ-

ment, and pollution, Maloney and Ward concluded that the average

16

17

person did not appear to know very much about the environment and
that knowledge about the environment has little effect on behavior.

However, Dispoto (1977) found that knowledge was more
predictive of activity than was emotion. He inventoried college
students' specific factual knowledge of environmental issues, verbal
»commitment toward environmental issues, and actual behavior in
improving the environment. It is not environmental emotiOn but
environmental knowledge that accounted for a consistently moderate
amount of the variance in how many environmental groups joined in
environmental efforts.

Ideally, both knowledge and attitudinal factors should be
incorporated in environmental education. The success of public
policy decisions, educational programs, and other efforts dependent
upon environmentally-oriented actions of specific individuals may
well hinge upon understanding of the relationships among their
personality characteristics, attitudes, values, knowledge and
environmental behavior. Some support to this direction is also
provided by Arbuthnot and Lingg (1975) who found that compared to
the French, there was a high degree of consistency between Americans'
environmental action and environmentally specific attitudes.

Environmental Attitude-Behavior
Relationship

 

 

Arbuthnot (1977) correlated selected attitudinal and personal-
ity characteristics, attitudes toward environmental problems,

environmental knowledge and behavioral commitment to the use of

18

recycling centers. Recyclers are no more likely to score high on
Pro-Ecological Attitudes, but they did score higher on Environmental
Cynicism and Ecological Responsibility.

Thus, while nonrecyclers appear to share many of the recyclers'-
concerns, the recyclers were more concerned about the future
-consequences of present policies and felt more compelled to take
action (which they did). At the same time, the recyclers felt that
their individual actions may have little lOng-range impact in the
face of the anti-ecological activities of large corporations and
governmental agencies. The nonrecyclers have a more benign attitude
about the motives and impact of the industrial and technological
systems in which they live.

Perry et a1. (1976) found a positive correlation between
attitude toward stopping air pollution and action against air
pollution.

General Attitudes, Specific
Attitudes, and Environmental
Behavior

Bruvold (1973) studied a community that had used water

 

reclaimed from domestic sewage to supply the public golf course,
recreation park, and swimming pool. He found that attitudes of
users of reclaimed water were more favorable than those of
nonusers.

He also found that attitude toward reclaimed water in the
community was correlated with an index of several specific

behaviors. This confirms the prediction that a behavior index

19

will correlate more highly with attitude than will behavior scores
taken singly.

Heberlein and Black (1976) tested the hypothesis that specific
environmental attitudes are more highly correlated with environmental
behavior than general environmental attitudes. They observed purchase
abehavior of lead-free gasoline and regular gasoline as well as
attitude toward air pollution. They found that attitude measures
that are more specific to a given behavior are better predictors
of that behavior than are more general measures.

Weigel and Newman (1976) found that their attitude measure
correlated .62 with a behavioral index but had an average correlation
of only .29 with the separate behaviors.

General Attitudes, Specific

Attitudes, and Environmental
Behavior--A Summary

 

 

 

General attitudes are important to study, not because they
predict a single behavior well, but because they influence a wide
range of specific beliefs and hence a large number of behaviors.

Both specific and general attitudes ought to be included in a study
to predict behavior, and the entire causal model from general
attitudes to specific attitudes to behavior ought to be traced.
Including only specific beliefs in a study is likely to give high
attitude-behavior correlations but will not show how the belief and
action relate to other attitudes and behavior. Including only general
attitudes is likely to be disappointing because not much of the

variance in behavior can be predicted. By including both, one can

20

better predict behavior from attitudes, yet show how the beliefs
and actions are part of a large cognitive configuration.
Hierarchical Model and Adoppion
of Environmental Beliefs
The order of adoption of environmental beliefs can be
{predicted from the hierarchical model. Environmental beliefs form.
a logical hierarchy and are adopted in logical order, i.e. from
tOp down. Once a given belief is adopted, then logic leads to the
adoption of implied beliefs. The adoption of the implied belief
will thus lag the adoption of the given belief by the time interval
for the logic of the system to work down one step. Ideally, the
adoption of the implied belief might be as fast as thought. But
there are various factors which slow things down. First of all,
the implied belief might not come to mind. If a person does not
consider a belief, then he will not think of it in relation to
other beliefs and hence will not be subjected to the logical influence
of the implying belief. This means that salient beliefs will be
brought into logical agreement much faster than non-salient beliefs.
Second, not all logical relations are binding. One belief may be
necessary for the adoption of another, but the more specific belief
may require adaption of some additional premise. Thus adoption of
the implied belief will not come about until the second premise is
accepted. Third, persons may doubt a belief even though they see
that logic dictates that belief. They may think the equivalent of
"It sounds logical, but may be I'm missing something. I'll wait

and see what other people think." The more often the person thinks

21

about the belief, the shorter the time until the doubt is
dispelled. And there are other factors.

Thus the lag in time for logical processes to work their
way into causal impact may range from minutes to weeks, months, or
even years for very esoteric beliefs. Thus one would expect great
.individual differences in the extent to which adoption works its way
down in any given hierarchy.

In the environmental area, nearly all of the relevant beliefs
involve either personal inconvenience or imposing behaviors on
others. Both are naturally repugnant and hence it is likely that
most beliefs in the environmental domain would tend to be rejected
unless persuasion is brought to bear. Thus the earliest message
that any given consumer is likely to hear on the topic of environment
is likely to be counter to his prior leanings. A message which
persuades the consumer to adopt a relatively specific belief is
likely to have only a temporary effect, since logical processes
would tend to bring the changed belief back into agreement with
higher order beliefs which are still negative. Thus the first
permanent change would occur when the consumer is first convinced
to adopt the belief at the top of the hierarchy. Permanent change
would then work its way down from there.

After a long campaign in which all aspects of a given
hierarchy are made salient to all consumers, we would expect belief
patterns to settle down into two types: one group that accepts the

belief at the top of the hierarchy and all the beliefs that follow

22

from it, and a second group which rejects the belief at the top Of
the hierarchy and everything under it.

The longitudinal pattern of development of environmental
beliefs in most populations can thus be predicted from the I
hierarchical model. In the beginning, nearly all consumers will
reject nearly all the beliefs in the hierary. Then some consumers
will adopt the belief at the top and begin moving down the hierarchy
as logic and the salience of events dictate. Over time, more and
more consumers adopt the top belief and begin the downward adoption
process. Finally, all of those who are going to adopt the t0p belief
will have done so and will have completed the logical development of
all the beliefs in the hierarchy. At this point, there will be
just two kinds of belief system: believe all or believe none.

A cross sectional study of environmental beliefs and
behaviors is a snap shot of the population at some point in this
developmental sequence. If the study is conducted at an intermediate
point, then there will be great individual differences in the extent
to which each consumer has progressed along the developmental path
that he will eventually follow. This diversity has two sets of
implications: implications for correlations between beliefs and
implications for mean belief adoption. The correlational impli-
cations are contained in the hierarchical path model laid out
elsewhere.

The hierarchical model postulates that all beliefs in the
hierarchy will be correlated, though for different reasons. A

belief that is directly superordinate to another belief will have

23

a direct causal impact on the implied belief. A belief that is
superordinate to another belief but is separated by one or more
mediating beliefs will have an indirect causal impact on the remotely
implied belief. Finally, two beliefs that have a joint superordinate
belief (i.e. beliefs which are directly or indirectly subordinate
-to the same third belief) will be ”spuriously" correlated because
of this common causal antecedent. Thus the hierarchical model
dictates the nature of the path analysis that should fit the
correlations between the belief scales in this study.

The means of the beliefs should also be patterned in terms
of the hierarchy. A belief score is high if the person has adopted
the belief and low if the belief is rejected. Thus the mean for a
belief is highly correlated with the count of the number of consumers
who have adopted that belief. Since consumers typically adopt the
top belief first, that belief should have the highest mean. Beliefs
directly implied by the top belief are adopted next and should have
the next highest means. As beliefs become more and more specific,
the mean should be lower and lower. This pattern is only a rough
one since different branches of the hierarchy may have very different
degrees of salience, but on the average groups of beliefs at levels

of greater specificity should have lower means.

Beer Containers and the Environment
Bottle Return Laws
"No deposit. No return. Born circa 1935.

Died in Oregon September 30, 1972.
May it rust in peace."

24

So reads a tombstone mockingly erected in Porland by
environmentalists in 1972, when Oregon became the first state to
outlaw the sale of beer and soft drinks in non-returnable bottles
and cans.1 The "bottle bill" also bans detachable pull-tab tops.
Since then, 1200 deposit bills have been introduced in the 50 states
~and the U.S. Congress. Mandatory deposit to encourage recycling has
also been strongly recommended by the NatiOnal Commission on Supplies
and Shortages (NCSS).2 There was an unsuCcessful attempt at the
federal level to enact a national anti-litter law in 1974 and again

in 1975.

Michigan Deposit Law

 

Here in Michigan, the voters passed the beverage container
deposit bill in the 1976 general election and the bill went into
effect in 1978. under the law, all soft drink and beer containers
sold, offered for sale, or given to consumers in Michigan must be
returnable and carry a cash deposit. The bottles (glass and plastic)
and cans of beer are thus tagged with a deposit of 5¢ or 10¢.
Returnable containers used by more than one manufacturer (termed
"certified" bottles) carry a deposit of 5¢, and those that can not
be used by more than one bottler (termed "uncertified" bottles)

carry a deposit of 10¢. Certified bottles are used primarily for

beer.

 

1Business Week (7/28/73): pp. 76-77.

2Progressive Grocer (March 1977): pp. 37-

25

Bottle Return and Environmental
Benefits

The objective of bottle bills is cleaner environment and
reduced expenditures of materials and energy. The Environmental
Protection Agency (EPA) has argued that by 1980 national deposit
legislation would reduce the annual litter of beverage containers
from 5.3 billion to 1.6 billion.1

Nonreturnables cost 6¢ to make, whereas the returnables
cost 8¢.2 The differential manufacturing cost is more than offset
by repeated use of returnables. After 10 trips, a refillable bottle
saves more than two-thirds of the energy needed to make a one-way
bottle. It also takes 87% less energy to make a recycled aluminum
can than one of virgin metal, while the energy savings from a
recycled steel can is 39%. A national deposit bill would result
in an annual savings of 530,000 tons of aluminum, 1.5 million tons
of steel, and 5.2 million tons of glass.2

Similar differences are noted between glass and can containers.
GAO has estimated a return rate of 90% for bottles and 70% for cans--
suggesting that glass containers are more efficiently reused than
metal containers. Glass also offers advantages in terms of greater

simplicity, and lower stabler raw material and manufacturing costs.

 

1Business Week (2/21/77): pp. 84-85.

2Advertising Age 43 (3/1/71): pp. 6-

26

Bottle Return Controversy

 

Resource shortage and public litter have evoked widespread
national concern. There is persistent demand for a massive switch
to returnable, refillable bottles. The deposit bill was, however,
trenchantly Opposed before it was adopted in Michigan. Passed at
[the end of an intense and prolonged debate on the economic and
ecological issues involved in waste disposal, the bill continues
to evoke heated debate. The two Opposing groups continue to debate
the pros and cons of the deposit bill. Indeed, ever since Oregon
enacted its Minimum Deposit law in 1972 to reduce roadside litter,
the throwaway ban has proved to be one of the most controversial

moves in the ecology bottle.

Bottle Return and Industry_Critics

Appearing before the U.S. Senate sub-committee for the
National Soft Drink Association, University of Chicago
Professor Richard A. Posner disputed the contention that federal
deposits or product changes in nonreturnable containers would
necessarily help reduce litter and solid waste or even save
energy.1

The basic argument is that people simply do not like to
return empties. They will do it during an altruistic campaign, but
in the day-to-day course of affairs, the consumer likes the

. 2 .
convenience and mobility of one-way containers. The National

 

1Food Processing 39(3) (March 1978): pp. 30-31.

2Industry Week 173 (8/17/70): pp. 58-

27

Soft Drink Association claims that 15 years ago, average trippage
for beverage bottles was 24-34 trips, but the range now is 10-12
trips, and in some urban areas it is more like 3-6 trips.l

It is contended that forced deposit tends to reduce conSumer
choice. It may further lead to higher prices and disappearance of
whouse brands. Overall, it is claimed "the consumer welfare will be
diminished by a forced substitution of returnables for nonreturn-

ables."2

Bottle Return and Economic Costs
Industry fears the costly overhauls of filling lines,
restructured production costs, and slashed production rates.
Industry argues that installing just one new refillable bottle line
costs about $13 million. The soft drink bottles would suffer the
most, because it is packaged in a larger number of plants than is
beer. For instance, Anheuser-Busch packages Budweiser beer in
15 plants, whereas Coca-Cola packages coke in 1000 filling plants.3
The switch to refillables would also be a switch from a
capital intensive system to one that is labor intensive. At
Anheuser-Busch's Newark, N.J. brewery, which packages some 5 million

bbl of beer each year, it takes only 6.5 peOple to work the can line,

 

1Commerce America 1 (11/18/76): pp. 16-

2Food Processing 39(3) (March 1978): pp. 30-31.

3Business Week (2/21/77): pp. 84-85.

 

28

13.5 to handle one-way bottles, and 20 to man a refillable bottle
line.1

It also causes extensive disruption for retailers and
wholesalers in the form of extra labor for handling, cluttered

store front, insufficient backroom space, sanitation, etc. The

.end result is a reorganization and slowing down of production lines.

Is Bottle Return Necessary?
Industry and consumer critics naturally question the need
for bottle return law.
While it is conceded that one-way bottles wind up as eyesores,

they represent ngumore than 4% of the total littergk How about

__ hl.._.—

 

litter in the form of strewn paper, plastic metal, and rubber
products?2 Beer and soft drink containers are apparently picked
mainly because they are the most visible. The energy savings derived
are also insignificant. Only 16,000 BTUS are expended for every
dollar spent on beer, whereas 70,000 BTUS are consumed for every
dollar spent on other grocery products.3

The success in Oregon can not also be duplicated elsewhere.
In the first place, Oregon has only 1% of the nation's population,

and its beverage markets are dominated by local brewers and bottlers

 

1Business Week (2/21/77): pp. 84-85.
2Supermarket (September 1978): pp. 107-

3Supermarket (June 1977): pp. 67—

29

who stand to lose very little from the deposit bill. Whereas the
impact could be disastrous in other states.1

The critics have therefore concluded that both litter
reduction and resource recovery are necessary to manage the nation's
soaring garbage load. The answer lies in recycling materials,

not banning one-way containers.

Summagy

Consistencrbegisse ,environmenfaLégtitudesLansLenEironmental
behayiornhasflbeen noted in a number-of.studies. Rgpyplers, for
instance, tend to have more pro-ecological attitudes.

Specific attitudes are better predictors of environmental
behavior than are general attitudes. However, general environmental
attitudes influence a broad range of specific attitudes and thus
m;ny behaviors. To obtain greatest insight into cognitive
configurations, one ought to look at both specific and general
environmental beliefs.

Returnable beer containers are said to yield a number Of
environmental benefits in terms of material and energy saved.
Returnables have, however, remained extremely controversial. Many

critics have contended that the container deposit requirement

imposes high economic cost and industry disruptions.

 

1Business Week (7/28/73): pp. 76-77.

CHAPTER IV
SCOPE AND METHODOLOGY

Outline
This chapter covers two broad areas--scope of the research
and the methodology employed in completing'the research. ‘In the
first part, we will look at the basic nature of research, specific
product examined, and the population studied.
In the second part, we will outline the construction of
survey instrument employed, nature of sample selected, and the

methods used in analyzing data and modelling.

Scope of Research

 

The purpose of this research is to explore the relationship
between abstract attitudes such as a belief that we suffer an energy
crisis and specific behaviors such as returning beer cans.

We will investigate the impact of consumers' environmental
beliefs on their beer package choice and container return frequency.
Environmental beliefs range from broad natural resource and conser-
vation issues to specific issues involved in container deposit law.
Beer is a frequently used product, and hence subject to routinized
purchase habits. Beer is a highly visible and extensively promoted
product. The environmental implications of beer packaging have been
widely publicized. These messages have linked environmental attitudes

to beer purchasing behavior.

30

31

The population studied is Greater Lansing. Of total 433
respondents in the sample, 225 are students in Michigan State
University and 208 are others living in Greater Lansing area.

The procedures employed in selecting the two samples and combining
them into one large sample are eXplained in the methodology section

Methodology of Research

 

Instrument COnstruction

 

Prior studies provided a general outline of environmental and
other issues involved in beer container returns. A pilot study was
conducted to more sharply define the issues. Ten people were
interviewed: six beer drinkers/buyers, two beer distributors, and
two bottle managers in Shop Rite Super Market and Meijer Thrifty
Acres.

The preliminary questions covered

- Consumer beliefs regarding
:resource shortage
:energy shortage
:seriousness of shortage
:conservation need
:conservation willingness
:need for society to control resource use
:role of beer companies in resource shortage
:significance of resource used in metal can or

uncertified bottle

32

- Consumer support for container return
- Frequency of container return by consumer

In the second stage, a written questionnaire was developed.
and pretested. The objective was to measure the consumers' attitude
toward a number of environmental issues. A preliminary draft was
'given to 30 respondents including MSU students, white and blue collar
workers, and professionals. The respondents were asked to be
critical, especially in regard to poorly Constructed or ambiguous
items. Should they have any trouble in understanding the meaning
of any word or phrase, they were encouraged to speak out or write
so on the margin.

These procedures greatly helped in identifying problem
questions or sections, and making the final survey instrument to
be clearer, easier, and quicker to complete. Certain inconsistencies
in the language were noted and hence modified. The pretest responses
were then analyzed in terms of means, standard deviations, and
correlation coefficients, so we could eliminate unnecessary items
or modify them.

The questionnaire was then formalized for the final survey.
All the attitude questions were answered on a five-point Likert
scale from strongly disagree to strongly agree. The direction of
one-half the items were reversed, and the items were reordered based

on random numbering.

33

The General Population Sample

 

Two separate samples were initially collected: .a non-
student sample and a student sample.

Formal sample selection procedures were employed to obtain
the general population or non-student population. The number and
-strength of 3-digit telephone districts in the Greater Lansing area
were obtained from Michigan Bell. Random number tables were then
used to generate the last four numbers. The sample distribution
among the telephone districts was allowed to correspond to the
strength of different telephone districts. Once a bank of numbers
is selected, calls were made using random digit dialing within the
bank until a specified number of households in each district was
reached.

Calls were made to the households explaining the purpose of
the survey and eliciting consumers' consent to receiving and completing
the questionnaire by mail. Approximately 65% of the households
consented to participate in the survey.“ The questionnaires were
then sent out by mail along with stamped return envelope. Approxi-
mately 70% of the mailed questionnaires were returned. Of the total

212 questionnaires returned, 208 were found complete and usable.

The Student Sample

In respect of student population, convenience sampling
procedures were employed. Although randomized selection on an
individual basis was not attempted, an effort was made to select

students belonging to different courses, levels, and backgrounds

34

at Michigan State University. These variables were systematically
varied to obtain a sample which, it is believed, broadly represents
the student population at MSU.

Of the total 234 questionnaires returned, 225 were found

complete and usable.

The Combined Sample

 

The responses were then coded on computer coding forms.
Attitude items were reversed, wherever necessary, so that all the
statements would lie in one direction. The responses were then
transferred from computer sheets to computer cards.

The student and non-student samples were studied separately,
and significant differences in responses, if any, were examined.
Such an examination revealed no significant differences in responses.
Hence, student and non-student samples were merged at this stage
and the reported data analysis is for the combined sample. This
combined sample consisted of 337 student and non-student respondents

who both consumed and purchased beer.

The Measurement Model

.Our next step was to derive appropriate attitude and behavior
scales based on the sample responses. Provisional scales were formed
using content analysis. These scales were then tested and modified
using confirmatory factor analysis. PACKAGE (Hunter and Cohen, 1969)
was used to find correlations among the attitude items. The
questionnaire items were then grouped according to the construct

being tapped. Multiple groups procedures were used to form item

35

clusters on the basis of content similarity, internal consistency,
and external consistency or parallelism (Hunter and Gerbing, in
press). The grouping of questionnaire items into different attitude
and behavior scales is summarized in Appendix B. The corresponding
clusters were scored and correlated, and this matrix was corrected
for attentuation due to less than perfect reliability (Nunnally,
1967). Correction for attentuation provides estimates of the
correlations between scales that would be Obtained if measurement
error were not present. This corrected matrix constituted the input

for path analysis.

The Causal Model

 

The first step in a path analysis is to formulate a causal
model predicting the structure of the correlations between the
variables studied. We used the logical precedence rules of the
hierarchical model to generate the causal model. The most general
belief scale was placed at the top. Those beliefs directly implied
by the general belief were added next. And so on. This model was
subjected to path analysis and showed almost perfect fit. However
the empirical analysis suggested a few links between beliefs in
different logical chains that require premises which were not thought
of prior to the study and were thus not predicted. The modification
added to the hierarchical structure, they did not contradict it.

Path analysis was performed on the belief correlation matrix
using PATHPAC (Hunter and Hunter, 1977). Path coefficients are

derived by ordinary least square estimation (Heise, 1975) using

36

simple or multiple regression of each variable onto its causal
antecedents. If a variable has only one antecedent, then the path
coefficient is the correlation between the dependent variable and
its antecedent. Where a variable has a number of antecedents,
then the path coefficients are the beta weights obtained from the
.multiple regression of the dependent variable onto the posited
antecedent variables within the model.

The path coefficients are used to generate a predicted
correlation matrix which is subtracted from the observed correlatiOn
matrix to provide a residual matrix from which the goodness of fit
of a proposed model can be judged. In the reproduction of the
correlations from the path diagram, the errors would not be expected
to be uniformly distributed unless the sample size were so large
that the estimation could be regarded as perfect. Otherwise, the
estimated correlation from the model depends on the length of the
causal paths which go into that estimate. The longer the causal

path, the greater the cumulated error in the estimate Of the

predicted correlation.

CHAPTER V
RESULTS AND DISCUSSIONS

Overview

The chapter begins by describing the intensity, direction,
and distribution of environmental beliefs and behavior observed in.
the population studied. Special attention is paid to the level of
environmental awareness, specific environmental beliefs, perceived
role of beer companies, and beer container return behavior.

The second section describes the procedures employed and data
used in testing the hierarchical model. The model is tested in two
ways: by testing mediational hypotheses and by testing network
hypotheses. The third section tests the mediational predictions
and the fourth section tests the hierarchical network hypotheses.

The fifth section resolves the dispute over general vs.
specific attitudes within the context of hierarchical model.

The sixth and final section compares the path model derived
in this study with results from prior studies.

Intensity, Direction and Distribution
of Environmental Beliefs and Behavior

 

 

Belief and behavior scales were formed by grouping individual
belief and behavior items on the basis of content similarity,
internal and external consistencies. LA confirmatory factor analysis
was conducted to check the groupings. The results of confirmatory

factor analysis are contained in Appendix C.

37

38

Also computed were the means, standard deviations, and
reliabilities for the belief and behavior scales. TheSe are shown
in Table 1. Item responses were averaged rather than summed to
preserve the original response scale. Thus scales are also scored

from l=Strongly Disagree to S=Strongly Agree. Category 3 is neutral.

Environmental Awareness

 

A mean soore of 3.79 was found for the belief that we are
experiencing serious resource shortage (506). A mean of 3.27 was
found for the belief that resource shortage is real (505). So,
while most consumers believe that we are passing through a period
of serious resource shortage, there is considerable disagreement as
to whether the current alleged resource shortage is real. A number
of consumers believe that the current resource shortage is hoax.

The highest positive mean Scores were obtained in respect
of two general environmental beliefs: that the consumers must
conserve resources (504), 3.88, and that such conservation would
make a difference (507), 3.82. A majority of consumers believe
that they should conserve resources and that, if they conserve, it
would help. A conservationist sentiment seems to prevail in the
majority of consumers studied.

The lowest standard deviation of .60 was obtained for the
belief that consumers' conservation would make a difference (507).
Whereas a standard deviation of .65 was obtained for the belief
that consumers must conserve resources (504). There is a substantial

agreement regarding what consumers can and should do for resource

39

TABLE l.--Be1ief and Behavior Scales.

 

 

 

Scale . a b
No. Content Mean SD Reliability
501 Society must control

resource use 3.63 .63 .76
502 Resources more important

than jobs 3.18 .93 .83
503 Industry will not act

on their own 3.54 .89 .74
504 Consumers must conserve

resources 3.88 .65 .61
505 Resource shortage is

real 3.27 .78 .70
506 Resource shortage

is serious 3.79 .72 .73
507 Consumers can help 3.82 .60 .65
508 Industry not doing

their best 2.79 .64 .65
509 Resources used in cans

significant ‘ 3.65 .79 .80
510 Prefer glass over metal

containers - 3.24 .64 .57
511 Beer companies prefer

can containers 3.03 .78 .71
512 Favor container return 3.24 .85 .75
513 Certified bottles

conserve resources 3.36 .68 .76
514 Government is doing

nothing 2.62 .90 .74
516 Frequency of container .-

return 4.24 .85 .69

 

aOn a scale from l=Strongly Disagree to 5=Strongly Agree.
Category 3 is neutral.

bCoefficient Alpha.

4O

conservation. Consumers who question such beliefs are not numerous

and/or do not believe so strongly.

Role of Government and Beer

Compgnies

 

Lowest mean scores were obtained in regard to the beliefs
‘that government is doing nothing about resource shortage (514), 2.62,
and that beer Companies are not doing their best (508), 2.79.
Consumers are divided as to whether either the government or the
beer companies are doing their best._ We could not detect any
inclination to blame either industry or government for the current
shortage.

A considerably higher standard deviation, .90 was obtained
for the belief that government is doing nothing (514) than for the
belief that beer companies are not doing their best (508), .65.
Consumers' beliefs regarding what government is doing differ more

widely than their beliefs regarding what beer companies are doing.

Need for Societal Control

 

A mean score of 3.54 was obtained for the belief that
industry will not act on its own (503) and 3.63 for the belief that
society must control resource use (501). The respondents are thus
divided in their beliefs regarding what industry does'and the need

for societal control of resources.

Jobs vs. Resource
A mean score of 3.18 was obtained for the belief that

resource is more important than jobs (502). Consumers seem roughly

41

equally divided in their support for resource priority over jobs.
There is a minority of consumers who strongly believe that resource
is more important than jobs.

Overall, both the issues of resource priority over jobs and
of societal control of resources, seem to divide the consumers into
.two distinct groups, one supporting and the other opposing. In fact,
highest diversity among all the beliefs was obtained in regard to
resource priority over jobs, as reflected'by a standard deviation
of .93.

Specific Environmental Beliefs
and Package Type Preferences

 

We obtained a mean score of 3.65 in regard to the belief
that resource used in cans is significant (509). Lower mean scores
are recorded for a number of other specific beliefs: that certified
bottles do conserve resources (513), 3.36; that they prefer two way
beer containers (512), 3.24; and that they prefer glass over metal
containers (510), 3.24.

A majority of consumers do recognize that resource used in
cans is significant. There, however, are those who disagree.

Consumers are significantly divided in their belief that
certified bottles conserve resources, in their preference for two
way beer containers and for glass over metal containers. There is
a minority of consumers who strongly believe that certified bottles
conserve resources, and who prefer two way beer containers and
glass over metal containers. One of the highest response variabil-

ities is obtained in regard to their preference for two way

42

containers. Consumers seem to be as divided in their preference
for two way containers after the passage of the bottle bill as they

were before.

Container Return Behavior

 

The highest positive mean score of 4.24 is Obtained in respect
of reported behavior, viz. the frequency of container returns (516).
Consumers return beer containers with an extremely high frequency
and thus contribute significantly toward the recycling effort.

There is only a small minority of consumers who do not return empty
containers.

The high frequency with which consumers return beer containers
in the face of their divided support for two way containers probably
results from the economic cost involved in not returning empties.

So that even those who do not favor two way containers return them
lest they lose the deposit. We did not ask the frequency with which
beer containers are returned solely for resource conservation rather

than for the deposit.

Testing the Hierarchical Model
Correlations between environmental belief and behavior
scales were computed and corrected for attentuation due to error
of measurement. These correlations are presented in Table 2.
The highest correlations are found between attitudes at the
same level of generality, as predicted. However, simple correlations

between belief scales do not specify the direction of causality.

43

 

OOH N, H m mm w on mm m ma m m 5H 5 b an mam

m OOH 0 HL m- m HH: «H. H: H H m: m m- m m- mHm
H o 00H 5. . a- mH 0H 0 am MH- 4. mm- EH- SH m a- ch
.m :H- a- ooH mH m- mm mm m: m m c H- 4 MH HH mHm
mm N- c- mH 00H Hm em me a cH mm as cc mm om ma NHm
c m mH m- Hm ooH cm cH cm H- mm m 4 mm m Hm HHm
6. HH- 0H mm am cm OOH ca em m an MH a mm 6 4H on
mm NH- 0 mm . map AH cc OOH N am mm on mm mm mH mm mom
m H- an m- S cm an N 06H cm- HN e m me mH cm mom
mH H MH- m 6H H: N am am. OOH mm cc me an mH He pom
m H c- m mm mm cm mm HN rm _ 66H on mm Hm mH ma . com
a m- mm- c as m MH on a oc on ooH Hc m mH «a mom
EH 8 SH- H- cc 4 a an n me am He 06H cm Hm oc «cm
a m- SH 4 mm. mm mm mm «c m- Hm m an 00H mm me mom
a m m MH om m c mH mH mH mH mH Hm mm cOH mm mom
mm m- a- HH me Hm «H mm on He me we cc me mm ocH Hom

 

on mam vam mam mam Ham cam mom mom mom mom mom vom mom mom Hon XMHUMZIm unm:H

 

.mcoiwucamuuou meadow ooaomoolléHmVN 5mg.

44

 

OCH 0 h m , on mm mm m 5 mm ma NH m m cam

O OOH O O- . OO. OH HO OO NO HO H- O OO O HHO
a . O : OOH OH O OH OO OH Om OO OH HO OH OH OOO
O O- OH OOH ON ON OH O- O HH O H- O O OHO
O- OO O OO OOH OO Om OO OO OH O O OH OH OHO
NO OH OH Om OO OOH OO O OO NO ON OO ON ON OOO
OO HO ON OH . Om. OO OOH O OO OO OH OO Om OO OHO
O OO OH O- OO O a OOH NO ON OO- O HO HO OOO
O. ON ON . O OO ON ON OO OOH OO O- OO HO O OOO
NO HO OO HH OH NO OO OO OO . OOH HO OO OO OO HOO
OH H- OH O O OO OH OO- O- HO OOH OO OO OO OOO
OH O HO H- O OO OO O OO OO OO OOH OO HO OOO
O OO OH O ON ON ON HO HO OO OO OO OOH OO OOO
O O OH O OH on . OO O O OO OO HO OO OOH OOO

 

mam Ham mom mam can mom mam mom mom Hom mom vom mom mom MONUMZIm omnwouomm

 

.mcoflumamuuoo mamom oo>uomnOnl.onm mammfi

45

 

 

OOH O h O OH NH Om O OH NO HH OH OH O OHO
O OOH O OH OO NN OH HO ON NH v n O m HHO
n O OOH N O O HH OH ON NN HH ON 5H NH NOO
O OH N OOH ON NH O O O O N m N H OHO
OH OO O ON OOH hv VN Om ON OH 5 HH O O OHO
5H NN O NH 5v OOH mv O O HN 5 NH O O OOO
Om MH HH O vN mv OOH MH HN Ov 5H ON ON OH NHO
O HO OH O Om. O OH OOH NO 5N O OH HH O OOO
VH ON ON O ON O HN NO OOH mv OH ON OH NH OOO
Nm NH NN O OH HN Ov 5N MO OOH Om OO Hv ON HOO
HH v HH N b 5 NH O OH Om OOH vm OO Hv OOO
OH 5 ON m HH NH ON OH ON OO vm OOH OO Hv VOO
OH O NH N O O ON HH OH HO OO OO OOH Oh OOO
O m NH H O O VH O NH ON Hv Hv Oh OOH OOO
OHO HHO NOO MHO OHO OOO NHO OOO MOO HOO FOO VOO OOO OOO

 

.mcoHumHmHHoo meom cmuoHOmumll.anN mamas

46

.OO. mH msoHuMH>wO COMO—HOO mo Esm 08H;

 

O O O O OHI O O HI NI O N NI OI O OHO
O O HI OHI O OI O HI N O OI OI ON O HHO
O HI O HH NI OH O HI O OH O N NI O NOO
O OHI HH O O OH O NHI NI O H VI O O OHO
OHI O NI 0 O HI O NI O VI OI VI OH O OHO
O OI OH OH HI O O VI VH HH NH ON OH VH OOO
O O O O . O O O OI V O HI NH O OO NHO
HI HI HI NHI NI VI OI O O HI OOI NHI OH HI OOO
NI N O NI O VH V O O O VNI NH O VI OOO
O O OH O VI HH O HI O O O N N OH HOO
N OI O H OI NH HI OOI VNI O O HH O HI NOO
NI OI N VI VI ON NH NHI NH N HH O O O VOO
OI ON NI O OH OH O OH O N O O O O OOO
O O O O O VH OO HI VI OH HI O O O OOO

 

OHO HHO NOO OHO OHO OOO I NHO OOO OOO HOO NOO VOO OOO OOO

 

.mCCflUMHQHHOU wHMOm ”GUUHUGHQ mans": 00>H0mQOII.AUvN "Hung?

47

A path analysis was conducted to trace the causality, and specifically
distinguish between direct, indirect, and spurious effects.

The path analysis model is shown in Figure 2. The model seems
to fit the data quite closely. Table 2 presents the statistics
required to assess goodness of fit. Table 2(a) presents the actual
.or observed correlations between scales corrected for attentuation.
Table 2(b) presents the correlations predicted by the model. Table 2(C)
presents the differences or errors in fit: The overall chi square
test shows no significant departure from the model.

The hierarchical model predicts that beliefs at the tOp of
the hierarchy are likely to be adopted before beliefs lower down.

Thus the model predicts that beliefs at increasingly lower levels,
i.e. beliefs that are more specific, will have increasingly lower
means.

To test this prediction, the path model was used to form
three sets of beliefs: the most abstract beliefs, the intermediate
beliefs, and the most specific beliefs.- The means for the beliefs in
each set were averaged. The abstract beliefs were: 505, 506, 504
and 507. The intermediate beliefs were: 502, 501, 503, 512, 508
and 509. The specific beliefs were: 511, 510 and 513. The belief

averages are shown below:

Abstract beliefs 3.69
Intermediate beliefs '3.34
Specific beliefs 3.20

Thus more specific beliefs are less likely to have been adopted.

This bears out the prediction of the hierarchical model.

48

 

Shortage
is Real
(505)
.70
Shortage
is Serious
/(506) \
.59 .58
\
Consumers Consumers
Must Conserve Can Help
.31/ (504) \ )507)
.60 .41
Resources \ /
Must Have Society Must
Priority Control f\\‘\\\\\\\
Over Jobs Resource Use .32
(502)K / (501) \ \
.29 .43 .49
\‘Indust gK/ 3g‘avor Fre enc
rY qu Y
Will Not Act Container-.39-9» of
On Their Own ' Return Container Return
(5%3) (5%2) (516)
.62 .43
Iv
Industry Resources
Not Doing Used in Can
Their Best Significant
(50's) \ (50'9)
.30 ‘34 .46
Companies Favor
Prefer Cans+______.50 Glass Over
(511) Metal
(Slim
.25
Certified Bottles
Conserve
Resources
(513)

Figure 2.-—Path Model

49

The detailed examination of the path diagram shows that it
confirms the hierarchical model. This examination will be presented

in two steps: mediational hypotheses and network hypotheses.

Testing the Mediation Hypotheses

 

The mediation hypotheses concern the immediate linkages
ibetween beliefs. The form to be considered here is a belief triad'
in which belief A tends to imply belief C only because there is an
intermediate belief B such that A tends to imply B while B tends
to imply C. The correlational form of such a hypothesis is that
the correlation between A and C will be much lower than either the
correlation between A and B or the correlation between B and C.
The path model shows a mediation effect of this type by showing
links from A to B and B to C but not from A to C. That is, the
correlation between A and C is the result of the indirect causal
path from A to B to C rather than a direct impact of A on C.

In the context of our study, the basic proposition is that
the influence of higher order environmental beliefs on behavior is
indirect and mediated by more specific environmental beliefs.
Belief Triads Showing that Causal
Mediation Follows Logical
Mediation

For purposes of illustration, we will first consider selected
belief triads in detail. In each such triad, the beliefs have the
logical structure: A tends to imply B and B tends to imply C, so
that A only indirectly tends to imply C. We will show that the

correlation between A and C is correspondingly lower than the

50

correlation between A and B or the correlation between B and C.
The rest of the key mediation relations will then simply be listed.

The belief that consumers can do much (507) and that they
favor container return requirements (512) have a correlation of .16,
whereas the intermediate scale that they support societal control
of resource use (501) has a correlation of .41 with the first_belief
and .49 with the second. Similarly, the beliefs that resource
shortage is serious (506) and that resource conservation is more
important than jobs (502) have a correlation of .15, whereas the
intermediate belief that consumers must conserve resources (504)
has a correlation of .59 with the first belief and .31 with the
second.

The beliefs that resource shortage is real (505) and that
the consumer can do much (507) have a correlation of .40, whereas
the intermediate belief that resource shortage is serious (506) has
a correlation of .70 with the first belief and .58 with the second.
Similarly, the beliefs that resource shortage is real (505) and that
consumers must conserve resources (504) have a correlation of .41,
whereas the intermediate belief that resource shortage is serious
(506) has a correlation of .70 with the first belief and .59 with
the second.

These relations between correlations are the basis for the
distinction between direct and indirect causal effects in path
analysis. The path diagram shows the mediational effects as causal.

chains. These chains run from general beliefs to specific behavior

51

and hence show the predicted mediation of causal influence. Other

important mediating relations are listed next.

Mediational Beliefs

 

The impact of belief that resource shortage is real (505)
on the belief that consumers must conserve-resource use (504) and
that consumer conservation would make a difference (507) is mediated
by their appreciation of the seriousness of resource shortage (506).
It is not enough for consumers to believe that resource shortage is
no hoax, they must also appreciate the seriousness of shortage,
before they become inclined to conserve resources. Before consumers
can be brought to think that they can and-should help in the resource
crisis, the seriousness of shortage must be communicated to them.

Two parallel beliefs that consumers must conserve resources
(504) and that consumer conservation would make a difference (507)
seem, in turn, to jointly mediate the impact of the belief that we
are in resource shortage (506) on the support for societal control
of resource use (501). Consumers that simultaneously believe that
they can and should help in the resource problem are more likely to
believe that society must control resource use. Correspondingly,
those who fail to believe that they can and should help in the
resource problem are less likely to believe that society must control
resource use. Even when consumers appreciate the seriousness of
shortage, they would not support societal control of resource use,
unless they are convinced they can and should help in the resource

problem.

52

The impact of the belief that we are in resource shortage
(506) on the belief that resources are more important than jobs (502)
is mediated by the consumers' willingness to conserve (504). Even
in the face of a recognized serious resource shortage, only
conservation-minded consumers would agree that resources are
.more important than jobs.

The belief that society must control resource use (501) seems
to shape a broad spectrum of beliefs, in regard to the role played
by industry (503, 508 and 511) and disposition toward container
return requirements (512, 509, 510 and 513). This belief is a key
link in the causal path. The impact of consumers' general environ-
mental beliefs about beer containers and what industry's role is
believed to be, is mediated by the belief as to whether or not
society has priority right over individual.

The beliefs that consumers must conserve resources (504) and
that consumer conservation would help (507) impact on their belief
that industry will not act on its own (503), but such an influence
is mediated by the belief that society must control resource use (501).

The impact of the belief that society must control resource
use (501) on the belief that resources are more important than jobs
(502) is mediated by the belief that industry will otherwise not act
on its own (503). Even those who recognize that society must control
resource use would favor resources over jobs only if they also
believe that industry will otherwise not act on its own.

The impact of consumers' beliefs that consumers can and

should help in resource shortage (504, 507) on favoring return

53

containers (512) is mediated by their belief that society must
control resource use (501). Even those who believe that they can
and should help in the resource problem would favor return containers
only if they also agree that society must control resource use.

The belief that society must control resource use (501)
-impacts on the belief that resources used in can are significant
(509), but such an impact is mediated by the preference for container
return requirements (512).

The impact of support for return containers (512) on
preference for glass over metal containers (510) is not direct,
but mediated by the beliefs that resources used in cans are signi-
ficant (509) and that industry is not doing its best (508). It is
not enough for consumers to favor return containers before they
prefer glass over metal containers, they should also believe that
resources used in cans are significant and also that business is not
conserving as it is.

The impact of the belief that resources used in cans are
significant (509) on the belief that certified bottles conserve
resources (513) is not direct, but mediated by the consumers'
preference for glass over metal containers (510). It is not enough
for people to understand that resources used in cans are significant,
they must also prefer glass over metal and believe that business is
not conserving either, before they can recognize that certified

bottles do save resources.

54

Environmental Beliefs Logically
Related

 

The mediation hypothesis assumes that environmental beliefs
are logically related. The path model which fits the data does follow
such logical relations. For example, consumers who perceive the '
resource shortage as real would at least consider such shortage as
potentially serious. Consumers who believe there is a serious
resource shortage should be willing to conserve.

Consumers who believe that society must control resource use
would understandably favor container returns. Conservation-minded
consumers not only favor container return requirements, but also
are conscious that resources used in can are significant. Their
perceptions and preferences are consistent.

That even conservation-minded consumers would provide only
lukewarm support to sacrifice jobs in favor of resources, is
explicable in the background of severe economic stress in Michigan

at the time of the study.

Testing the Network Hypotheses

 

The network hypotheses are the global structure hypotheses of
the hierarchical model: that the causal structure will follow the
logical structure from general to specific beliefs. The belief
which is logically precedent to all others is the mostgeneral belief
in the inventory: the belief that resource shortage is real (505).
According to the hierarchical model, this belief should also be

causally prepotent and hence should appear at the top of the path

55

model, as it does. Other beliefs should then follow this one in
order of generality with specific behavior appearing at the bottom
of the path diagram.

Two major propositions are involved in the hierarchical
network hypotheses:

A. That causal relations between environmental beliefs are
structured in a set of hierarchical relationships.

B. That the causal structure of environmental beliefs is
arranged from most abstract and general to the most concrete and
specific.

Hierarchical Relations Among
Environmental Beliefs

 

 

(a) Prepotent belief

 

The most general attitude in the survey is the belief that
resource shortage is real (505). The hierarchical model predicts
that this belief should be prepotent in causal influence. It was
predicted to appear at the top of the path diagram and it does.

This belief is substantially correlated with a number of environmental
beliefs, and especially with key general environmental beliefs in
regard to the role of consumer, need for societal control of resource
use, etc. It is correlated with the specific environmental beliefs
such as the consumers' disposition toward container return require-
ments, though to a lesser extent as predicted.

The belief that resource shortage is real (505) impacts
directly on the consumers readiness to recognize that resource

shortage is serious (506). Alternatively, if they believe that

56

the shortage is a hoax, they are unlikely to recognize its seriousness.
Belief that resource shortage is real (505) significantly impacts on
the consumers' willingness to conserve (504), the belief that consumer
conservation will make a significant difference (507), that society
must control resource use (501), and specifically their support of
container return requirements (512). A correlation of .47 between_
consumers' beliefs that resource shortage is real and that they

favor container return requirements, suggests the key role played

in our study by the former belief.

(b) Higher level vs. lower level concepts

Hierarchy implies also that higher level concepts are
correlated with a large number of other concepts, both among
themselves and with lower level concepts. This is an extension
of the previous suggestion regarding the existence of prepotent
concept (or concepts).

The following general belief scales

505 Resource shortage is real

506 Resource shortage is serious

504 Willingness to conserve

507 Consumers' conservation would help

501 -Society must control resource use
are not only substantially correlated with each other. They are
also significantly correlated with a number of other scales. The

following correlational matrix highlights this observation:

57

505 506 504 507 501 ; 503 512 509 502

*
505 1.00 .70 .41 .40 .42 f .08 .47 .20 .15
506 1.00 .59 .58 .48 E .31 .28 .28 .15
504 1.00 .45 .60 E .37 .40: .38 7.31
507 . 1.00 .41 i -.09 .16 .24 .19
501 1.00 3 .43 .49 .32 .35

Substantial correlations are thus obtained in respect of
general environmental beliefs (501, 504-07)--beliefs regarding
resource shortage, resource conservation, and society's control
over resource use. They are also significantly correlated with a
number of specific beliefs (502-03, 509, 512). For instance, the
beliefs that resource shortage is serious and that consumer conser-
vation would make a differenceo-both seem to play key roles in the
hierarchical network. Consumers have to believe that society must
control resource use if they are to favor container return require-
ments. They must also be willing to participate in resource

conservation.

(c) Hierarchical strings

 

The hierarchical hypothesis would not only suggest that
higher level concepts branch out on to different realms, but also
that each branch or string of beliefs can be considered separately.
There are two distinct strings of environmental beliefs,
one dealing with industry's role and the other dealing with the role
of the consumer. The belief that industry will not act on its own
(503) is correlated with the belief that society must control resource

use (501), that industry is not doing its best (508), and that

58

beer companies prefer metal cans over glass bottles (511).
Similarly, support for container return requirements (512) is
correlated significantly with the beliefs that society must control
resource use (501), that resources used in can containers are
significant (509), the preference for glass over metal containers
.(510), and that certified bottles save resources (513). Those who
support societal control of resource use favor container return
requirements; reinforced by the beliefs that resource used in can
container is significant, that certified bottles save resources,

and that glass is preferable to metal for containers.

(d) The family of environmental beliefs

 

The hierarchical hypothesis asserts that all of the environ-
mental beliefs belong to the same family. Hence all are causally
linked, either directly or indirectly. This is confirmed by the

path analysis.

(e) Hierarchical direction not reversible

 

The hierarchical model not only suggests that environmental
beliefs are layered at different levels, but also that the causal
linkages are unidirectional. All causal arrows in the path diagram
of Figure 2 point from more general to less general beliefs. This
path diagram is consistent with the hierarchical model assumption
of top down causal flow. This path diagram fits the data and hence

tends to confirm the hierarchical model.

59

Structuring of Environmental Beliefs
from Most Abstract to Most Concrete

 

 

The network of environmental beliefs has general, abstract
beliefs at the top. The prepotent concept is the belief that resource _
shortage is real. Immediately below are the following general
beliefs:

505 Resource shortage is serious .

504 Consumers must conserve

507 Consumers' conservation would help

501 Society must control resource use

At the lowest level are found the following most specific,
concrete beliefs:

511 Beer companies prefer metal containers

513 Certified bottles conserve resources

As we come down from the top level we find gradually
increasing specificity in the belief concepts. For instance, if
we examine the string of beliefs on the role of beer companies, we
have at the top the belief that beer'industry will not act on its
own, then the belief that beer industry is not doing its best, and
terminating with the belief that beer companies prefer metal over

glass containers.
Impact of General vs. Specific Environmental
Attitudes on Environmental Behavior
in the Context of Hierarchical Model

The hierarchical model predicts that attitude behavior

correlations are highest if they are measured at an equivalent level

60

of generality. Specific behavior will be predicted best by
specific beliefs.
Presented below are the first order correlations between
container return behavior and a number of environmental beliefs.
512 501

507 504 506 505

I‘d-I'l-I'
****I*

516 .39 .32 .18 .17 .08 .09_
Container return behavior has its highest correlations of .32 and
.39 with the belief that society must control resource use (501) and
the consumers' disposition toward container return requirements (512).
In the next level, container return behavior has intermediate level
correlations of .18 and .17 with the beliefs that consumers must
conserve (504) and that their conservation efforts would help (507).
Finally, container return behavior has its lowest correlations of
.09 and .08 with such general beliefs that resource shortage is
real (505) and it is serious (506).

This data supports the prediction derived from the hierarchical
model: specific behaviors will be best predicted by correspondingly
specific beliefs.

Relation Between Present Findings
and Prior Research

 

 

Dispoto (1970) assessed the belief that resource shortage is
real. While the superordinate role of this belief in the environ- -
mental area is upheld in our study, our research shows that it is
mediated by the belief that resource shortage is serious. Consumers

not only have to agree that resource shortage is real but also

61

that such shortage is serious before they can be persuaded to
become conservers.

Arbuthnot (1977) hypothesized that consumers' willingnessl'
to conserve resources is based on beliefs that resources have priority .
over jobs and that society must control resource use. The evidence
in our study reverses this causal ordering. Our data suggest_that.
consumers become conservers before they will support societal control
of resources. We find here a reasoning consumer: if he believes
that resource shortage is serious and that his action would make
significant difference, then he will support societal control of
resource use.

Arbuthnot (1977) has hypothesized that consumers conserve
because the actions of large corporations are antiecological. Our
data show that although many consumers criticize industry, industry's
actions are not the basis for consumer conservation. Many conservers
have considerable appreciation for the seriousness of the shortage
and are concerned with the role they can play without waiting to see
what industry might or might not do. Even the conservation minded
consumers are not strident in their criticism of industry. The
majority believe that industry is doing its best, though needing
to be prodded all the time.

Bruvold (1973) suggested that conservation minded consumers
recognize resources as more important than jobs, and hence are
willing to conserve. Our path analysis suggests that consumers'
willingness to conserve is not affected by their belief that jobs

are more important than resources. Consumers will support resource

62

priority over jobs if they are conservation minded and if they
believe that industry will not act on its own. Two obServations

may be made in this connection: (1) There is no widespread support
for resource priority over jobs. Even the heavy conservers are
extremely sensitive when it comes to jobs. Their avowed support

for societal control of resources will not prompt them to give up
jobs to save resources. (2) The environmental activities of large
corporations are not going to make conservers conserve any less.
Conservation by large corporation may supplement consumer conservation
efforts.

Polling correlations (Business Week, February 21, 1977) have
been interpreted to mean that consumers favor container return,
because (1) they are willing to conserve and (2) the actions of
large corporations are antiecological. Our data confirm the causal
direction of the first interpretation but show the second correlation
to be "spurious." Our data show that consumers who favor container
return are those who believe society must control resource use,
those who are willing to conserve, and those who believe that their
own conservation efforts would be useful. On the other hand, as
noted in the discussion of Arbuthnot (1977), consumers' commitment

to container return does not depend on their belief about industry.

CHAPTER VI

SUMMARY OF CONCLUSIONS

Intensity, Direction, and Distribution
of Environmental Beliefs and
Container Return Behavior

 

 

 

Consumers agree that we are passing through a period of
serious shortage but disagree whether such shortage is real. However,
there is substantial agreement that consumers should conserve and
that it would help if they did so. A significant conservationist
sentiment seems to prevail in the population studied.

A majority of the consumers believe that both government and
beer companies are doing what they can. They are not inclined to
blame the resource shortage on either government or industry. Though
consumers are somewhat more critical of the government than of the
industry.

Consumers are divided when it comes to their job. While
there is a significant minority of consumers who strongly believe
that resource is more important than jobs, jobs are too important
to be taken lightly even by conservationists. The emotional nature
of jobs is revealed by the wide range of how consumers feel.

While a substantial majority believe that the resources used
in containers are significant, there are many who disagree. Consumers
are divided in their beliefs that certified bottles conserve resources,
and also in their preference for two way containers and glass

containers over metal containers.

63

64

Consumers seem to be returning empty containers with high
frequency. Apparently, many of those who do not favor return

containers return them anyway to get their deposit back,

Testing Mediation Hypotheses

 

Internal Influence of a Family of
-Beliefs Goes Down Only Along
Hierarchical'Channels

 

 

(1) Direct relationships between environmental beliefs are
weaker than indirect relationships.

(2) This reflects the mediational nature of such beliefs.
A number of mediating beliefs can be identified.

*Before consumers who understand that resource
shortage is real become conservers, they must
also appreciate the seriousness of the
shortage.

*Even in the face of recognized serious shortage,
only conservation minded consumers would agree
that resources are more important than jobs.

*EVen those who believe that they can and should
help will favor refillables only if they agree
that society must control resources.

(3) According to the hierarchical model, the causally
prepotent belief should be the most general belief. In this
inventory, the most general belief is the belief that the resource
shortage is real. This belief is substantially correlated with a
number of abstract and concrete environmental beliefs, and, in

particular, is significantly related to the disposition toward

container return requirements.

65

Influence of General Environmental
Beliefs on Container Return
Behavior is Indirect and

Mediated by More Specific

Beliefs

 

 

 

 

(1) Environmental behavior is substantially correlated only
with the specific environmental beliefs at the bottom of the hierarchy.
.The highest correlation with frequency of container returns is with
consumers' disposition toward container return requirements. The
next highest is with the belief that society must control resource
use.

Environmental behavior is correlated only minimally with
environmental beliefs at the top of the hierarchy. For instance,
the frequency of container returns has a very low correlation with
the belief that resource shortage is real and serious and a low
correlation with consumers' inclination to conserve resources.

(2) Such low correlations between specific environmental
behavior and abstract environmental beliefs is explained by the
presence of mediating concrete beliefs. For example, even consumers
who believe they should and can help in conservation would not
necessarily return beer containers unless they believe that society
must control resource use. The longer the chain of mediating beliefs,
the lower the correlation.

Environmental Beliefs are
logically Related

The path model shows that causal relationships follow logical
relationships. For example, it is understandable that consumers who

perceive shortage as real would come to consider it as serious.

66

Similarly, consumers who believe that societal control is necessary
would come to favor container requirements.

Consumers' perceptions and preferences seem consistent. The
fact that even conservation minded consumers give only lukewarm
support for resource priority over jobs reflects the severe economic

.stress experienced by a large number of Michigan consumers.

Testing Network Hypotheses

 

Environmental Beliefs are Structured
in a Set of Hierarchical Relations

 

 

(1) As predicted by the hierarchical model, the path
analysis shows environmental beliefs to form one family.

(2) The prepotent belief is that resource shortage is real.
This belief impacts on every other belief in the hierarchy. Specific
environmental behavior and more specific beliefs can be logically
traced all the way back to the prepotent belief.

(3) Abstract beliefs are highly correlated with each other
and with beliefs at the next lower level.

(4) Environmental beliefs branch out into distinct hierarchi-
cal strings such that each string can be studied separately.

(5) The hierarchical direction is not reversible. This is
shown by the fact that causal arrows are downward from more abstract
to more concrete beliefs, confirming the prediction of top-down

causal flows.

67

Environmental Beliefs are Structured
from Most Abstract to Most Concrete

 

 

Environmental beliefs are arranged from most abstract and
general to the most concrete and specific. This is shown by the
fact that abstract beliefs (e.g. seriousness of resource shortage,
need for societal control of resource use, etc.) are all found at
'the top. At the bottom, we have the more concrete beliefs_regarding
container return requirements, role of beer companies, etc. As we-
go down the hierarchy, we find gradually increasing concreteness
in beliefs.

General vs. Specific Attitudes in the
Context of the Hierarchical Model

 

 

The hierarchical model would predict that attitude-behavior
correlations will be highest if both are measured at an equivalent
level of specificity, and that correlations with specific behaviors
are smaller if the attitude is more abstract. This is shown by the
fact that container return behavior has the highest first order
correlation with consumers' disposition toward container return
requirements, has intermediate correlations with beliefs that
consumers must conserve resources and that it would help if they
conserved, and has its lowest level of correlations with beliefs

that resource shortage is real and serious.

Comparison with Prior Studies
Compared with prior environmental studies, our path model
suggests the need to not only know whether a person believes that

the resources shortage is real, but also whether they believe the

68

resource shortage is serious, before they can be persuaded to
become conservers. Not only must the consumers be willing to
conserve, but he must believe it would help before he can be
persuaded to support societal control of resource use.

Whereas prior studies hypothesized that consumers are inclined
to conserve because they recognize the need for societal control
over resource use, our study suggests the causal order to be the
reverse. A consumer first becomes a conserver, and then because
of their conservation mindedness they come to support societal
control of resource use.

Contrary to the findings from prior studies, our consumers
are not cynical. They do not blame either industry or government
for the current resource shortage. Those who are conservers are
concerned with what they themselves can do.

The Hierarchy Model and Container
Return Behavior

 

Our study shows that there is a hierarchy of environmental
attitudes which mediates the environmental attitude behavior
relationship. General environmental beliefs are connected to
specific environmental behaviors through causal chains leading first
to more specific beliefs and then to behaviors. In such a structure,
general beliefs play the dominant role in determining behavioral
choice over the long run. Such general attitudes can counteract the
effects of external messages on specific beliefs.

The hierarchical model is a theory of the relationship

between general and specific beliefs, and between beliefs and

69

behavior. We need to understand this hierarchy of general and
specific beliefs if we wish to understand or explain how behavior
choice occurs. By recognizing the mediational effect of intervening

beliefs, one could avoid misreading attitude behavior correlations.

CHAPTER VII
IMPLICATIONS AND RECOMMENDATIONS

Implications for Container Recycling

The results show that consumers in Michigan are conservation
oriented. They are prepared to make reasonable adjustments in
their lifestyle so long as their jobs are,secure. In the background
of current economic distress, strong concern is expressed over
jobs. The majority of consumers are not dogmatic critics of either
government or industry. They just want to help.

The lower means for specific (compared with general) environ-
mental beliefs may indicate the need to educate consumers regarding
specific environment-related issues, like bottle vs. can containers,
certified vs. uncertified bottles, etc. Consumers need to be more
concretely educated regarding what beer containers, certified
bottles, etc. entail in terms of environment and ways in which
consumers can help conservation.

The lower means may also stem from the lag in time as environ-
mental logic works its way down the hierarchy. It is likely then
that beliefs would be adopted in logical order, i.e. from the top
down. Adoption of an implied belief will lag the adoption of a
given belief by the time interval required for logic to work down

one step.

70

71

Given the substantial conservationist sentiment found in the
consumers, we ought to develop environmental programs which will help
consumers along in finding concrete ways in which they can conserve

resources .

(Optimal Environmental Education Campaigns

Campaign Objective

 

Causal influence among environmental beliefs goes from top
down. An optimal campaign must adapt to this logic. Permanent
belief change can occur only at the top of the hierarchy. Hence
the first thrust of a campaign should be aimed at achieving adoption
of the most general beliefs. Once part of the population has adopted
this belief, then messages can be aimed at lower beliefs. HOwever,
messages aimed at the top of the hierarchy must be continued until
all those who can be persuaded at the most general level have been
reached. It is the general environmental beliefs that form the
basis for more specific beliefs which in turn determine behaviors.

Over time, the campaign should become more and more broadly
based as more and more specific beliefs and specific behaviors are
added to the growing base of the hierarchy as developed in the
campaign message. As the program approaches time for specific
behaviors to be stressed, consumers should be provided with concrete

illustrations of how they could act to conserve resources.

Mediating Messages

 

The presence of mediating beliefs must be recognized in

designing the message for any environmental education program.

72

Campaign messages must incorporate the mediating beliefs which
will enable change at the higher level to induce change at the
lower level. Consumers will work through such logical chains by
themselves, but only slowly. If education campaign lays out the

mediating steps explicitly, then consumers might change more quickly.

Failure of Direct Behavioral Campaigns

 

Much of advertising theory is behaviorist in orientation.
These theories would argue that if you want to change a given
behavior, then aim your ads at that behavior. The hierarchical
model suggests that in the environmental area this strategy will
fail.

Short lived promotional programs aimed at isolated behaviors
will not lead to durable results. An environmental campaign aimed
at a specific behavior will only temporarily modify that behavior
if it leaves superordinate beliefs unchanged. Once the campaign
is over, the behavior will drift back to match these unchanged

beliefs.

Charting the Campaign

 

The planning of an optimal campaign requires knowledge of
the hierarchy in which desired behaviors are embedded. The deter-
mination of this hierarchy is in part a matter of logic: the
deduction of the list of beliefs which are logically antecedent
to the desired beliefs. However, once a portion of the hierarchy
has been mapped out, then it should be fleshed out by talking to

people in the population under consideration who are most likely

73

to have developed hierarchies of their own, i.e. pe0ple who have had
considerable experience with the relevant things, processes, or
events. In this research, there was a lot of information gained by
talking to beer distributors, managers of liquor departments, etc.

A partial hierarchy can be used to make part of this interview a
structured interview to flesh out the part of hierarchy already
known and to begin the process of identifying the terminology used
by that population.

The timing of messages depends on how many persons in the
population have adopted the top beliefs or the intermediate beliefs,
etc. While the order of adoption of beliefs can be predicted from
the model, the exact timing can not. Thus the campaign should
include plans for successive polling to chart the hierarchy prior
to decision points.

In charting such a program, two factors must be kept in mind:
time involved in achieving visible results and differences between
individuals who we are trying to influence. Environmental education
takes time to yield visible results. A prepotent belief once
modified takes time to filter through hierarchy of beliefs and to
be translated into meaningful environmentally-oriented activities.
Patience and persistence are called for. Hierarchies of environ-
mental beliefs may also differ between individuals--in terms of
differences in beliefs, in how they are related, and the strengths

of such relationships.

74

Implications for Marketing Decisions

 

Study suggests need to create goodwill for the product as a
whole rather than seek preference for one particular brand.

Consider the following hierarchy of Games

Games
Puzzles Board Video Computer
Games Games - Games
Sports Pong Space Combat
“(l/HEENME:\\‘ Games Games Games
Atari Intellivision Atari Space
Football Football “///’\\\\\;War
Space Asteroids
Invaders

The market for games is limited because many adults reject games

as entertainment (they are too "childish"). Here ads directed at

the top might break down resistance and greatly increase the customer
base. After all ads at bottom compete for fixed customer base,
whereas ads at top increase customer base. Such an approach may

be appropriate for companies in slow-growth industries, e.g. tobacco,
tire, coffee, etc. This will especially appeal to the industry
leader who is likely to benefit first from any expansion in industry
demand. General Foods has shifted its advertising from building

market share to building the size of the overall coffee market.

75

In such an effort, companies may join together to promote
the industry product as a whole rather than indulge in brand-level
competition in a stagnant industry. Industry level promotion is
already done, for instance, in the case of cotton, milk, commercial
banks, etc.

Similar arguments can be made regarding company wide adver-j
tising rather than product advertising. If the causally prepotent
belief is at the level of company as a whole, then institutional
advertising or image advertising may go a long way in laying ground
for general goodwill. Once such a base is established, the company
would use messages that link the company to lower level products.

In building store loyalty, retailers realize that economic and other
specific appeals become important to customers only after they
perceive an initially acceptable image. A number of retailers tend
to promote an overall theme rather than any one or more identifiable
product line, e.g. Sears promoting price image, KrMart promoting
quality image, Macy's promoting "middle-priced department store"
image, etc. Institutional advertising of the type launched by
Mobil, Exxon, etc. is also aimed at building corporate goodwill

and later product loyalties.

APPENDI CES

APPENDIX A

QUESTIONNAIRE

SUBJECT: Survey of Environmental Attitudes and Consumption
Behavior

The purpose of this survey is to describe consumers' feelings
about several issues, their purchasing choices and consumption
behaviors. Some people believe that the environment is in serious
state and requires early action, while others say that the environ-
mental issues such as energy are overplayed, and that there are more
important public issues. Here, we seek to determine where people
stand on several issues, their purchasing and consumption behaviors.

The survey consists of a set of questionnaires on the topics
of natural resources, energy, beer containers, and beer purchase
behavior. The instructions for filling out the pages are found at
the top of each page. There are no right or wrong answers to the
questionnaire statements: only your opinion is desired. Please
respond to all the items.

Sometimes you may feel as though you have seen the item
before. This will not be the case, so please do not look back and
forth through the items. Also do not try to remember how you
checked similar items earlier. Please make separate and independent
judgement for each item. 1

Thank you very much for your cooperation.

76

INSTRUCTIONS:

77

 

Natural Resources.

 

deposits, oil, coal, etc.
that area.

closest to your own opinion:

SA a Strongly Agree

D = Disagree

A 8 Agree SD = Strongly

N 8 Neither Agree nor Disagree

Below you will find a list of statements about

By "natural resources," we mean forests, metal
We would like to know your opinions in
Against each statement, circle the response that is

Disagree

 

C 2.
C 3.
C 4.
C 5.
C 6.

/c 7.

/ c 8.
/ c 9.
c 10.

We are entering a period of scarcity

and shortage of most natural resources.

Industry is largely responsible for
the natural resource problem.

Where natural resources are privately
owned, society should have np_control
over what the owner does with them.

Industry will act in the public
interest if the government will
let them.

More emphasis should be placed on
individual's economic rights than on
society's natural resource rights.

Society must ultimately control what
citizens do with the nation's natural
resources.

We must enjoy life with the natural
resources we now have, and let the
future take care of itself.

Government is doing nothing to
alleviate the nation's natural
resource shortage.

We will have plenty of natural resources,

if we just invent new processes for
finding and developing them.

If an industry is to provide jobs in
certain areas, we must be ready to
put up with some natural resource
wastage.

SA

SA

SA

SA

SA

SA

SA

SA

SA

A N D
A N D
A N D
A N D
A N D
A N D
A N D
A N D
A N D
A N D

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

C 11.

C 12.
C 13.

/'c 14.

/c 15.

C 16.

78

Industry will never do anything in the
society's interest if it reduces their
profits.

The "scarcity of natural resources" is
just a hoax played on the consumers.

Consumers can do much to alleviate
the natural resource shortage.

We should turn to conserving natural
resources only if it does not change
our life style.

Fear of natural resources shortage
should not discourage us from using
natural resources and enjoying life
today.

Many companies are working together
to conserve natural resources.

SA

SA

SA

SA

SA

SD

SD

SD

SD

SD

SD

79

 

INSTRUCTIONS: Below you will find a list of statements about
Energy. By "energy" we mean oil, coal, natural gas, etc. We

would like to know your opinions in that area.

Against each

statement, circle the response that is closest to your own
opinion: '

SA = Strongly Agree D

A I Agree SD

N = Neither Agree nor Disagree

Disagree

Strongly

Disagree

 

Many companies are working together
to conserve energy resources.

If we continue our high levels of
energy use, future generations will
.293 be able to have a level of
living like ours.

Consumers should make every effort
to conserve on energy use.

Consumers are largely responsible
for the current energy shortage.

We will have plenty of energy sources,
if we just invent new processes for
finding and developing them.

The "energy crisis" is just a hoax
produced by the oil companies.

If an industry is to provide jobs in
certain areas, we must be ready to.
put up with some energy wastage.

We are in an "energy crisis."

Government is doing nothing to
alleviate the nation's energy
problem.

If the consumers tried to conserve
energy, it would really make a
difference.

SA

SA

SA

SA

SA

SA

SA

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

27.

28.

29.

30.

31.

32.

80

Society must ultimately control what
citizens do with the nation's energy
resources.

More emphasis should be placed on
individual's economic rights than
on society's energy rights.

Industry will act in the society's
energy interest, if the government
will let them.

Where energy sources are privately
owned, the society should have pp
control over what the owner does
with them.

Most energy wastage is caused by'
business. ‘

If we do nothing about it now, energy
shortage will soon become the major
national problem.

SA

SA

SA

SA

SA

SA

SD

SD ‘

_ SD

SD

SD

SD

INSTRUCTIONS:

81

 

Beer Containers.

 

your own opinion:

SA = Strongly Agree

D = Disagree

A = Agree SD = Strongly

N 8 Neither Agree nor Disagree'

Below you will find a list of statements about
We would like to know your opinions in that area.
Against each statement, circle the response that is closest to

Disagree

 

The production of metal containers
uses much more energy than the
production of glass containers.

"Certified bottles" (on which we pay

a cash deposit of 5¢ each) use much
more energy than "uncertified bottles"
(on which we pay a cash deposit of

10¢ each). -

Metal containers should be replaced by
glass containers wherever possible.

Beer companies will not promote glass
containers unless the government
forces them to.

Beer companies would rather promote
metal cans than glass bottles, so
they can make more profit.

I would not buy beer in metal
containers if glass containers
are available.

Tbo much of my money is locked up in
container deposits.

Forced deposit has resulted in
needless disruptions all around.

The amount of metal used in making
beer cans is so insignificant that
I would not worry about it.

The production of metal containers
uses scarcer raw materials than the
production of glass containers.

SA

SA

SA

SA

SA

SA

SA

SA

SD

SD

SD

SD

SD

SD'

SD

SD

SD

SD

C 43.

C 44.

C 45.

C 46.

82

The amount of energy resources used in
beer cans is so insignificant that I
would not worry about it.

I should have the freedom to buy the
beer container I like.

I hate the thought of returning empty
beer containers to the store.

"Certified bottles" (on which we pay
a cash deposit of 5¢ each) use much,
more natural resources than
"uncertified bottles" (on which we.
pay a cash deposit of 10¢ each).

SA

SA

SA

SD

SD

SD

SD

83

BEER PURCHASE BEHAVIOR

 

In this section are listed a number of questions regarding

yourself as beer purchase/consumer.

Please respond to them as asked.

 

C 47.

C 48.

C 49-54.

Do you drink beer?
YES
NO

Do you buy beer?
YES
NO

If YES to Question 48 and NO to Question 47, answer
Question 49-55 and then STOP.

If YES to Question 47 and NO to Question 48, SKIP
Questions 49—55. GO to Question 56 and continue.

If YES to BOTH Questions 47 and 48, please GO right
AHEAD and complete all the questions.

Listed below are the features generally considered in
purchasing beer. Rank order them in terms of their
importance to you.

(1 a Most Important . . . . . . . 6 = Least Important)

Price of beer

Brand name

Taste

Package type (glass or can)
Package size

Container deposit

84

C 55. Check one of the following:

I would buy any brand if it were the cheapest
I would buy only certain brand(s) .
I would buy that beer which has the taste I like.
None of the above
C 56-86. For gash brand below, mark 1 beside those that are yppg
favorites: 2 beside those you have tried and liked:‘

'3 beside those you have either not tried or indifferent to;
4 beside those you have tried and disliked: and S beside

 

those you have tried and really hate.

Heineken Dark

 

Schlitz Malt

_____Blatz _____Mdchelob Light
Braumeister ____ Miller
Budweiser _p___Molson
Busch ____ Natural Light
Colt 45 ____ Old Milwaukee
Coors __ Olympia
Falstaff _____ Pabst
Guiness ‘____ Pabst Light
Heineken Light _____Schlitz

Any brand left out? Please specify.

Labatts Stroh
Lite Stroh Light
Michelob

C 87-117.

C 118.

For each brand below, mark 1 beside those you would
always buy if available in the store and you could
afford it; 2 beside those that you would also buy;

3 beside those you are indifferent to: 4 beside those
you would buy only if you had to; and 5 beside those

you would never buy.

 

Blatz Michelob Light
Braumeister Miller
Budweiser Molson

Busch Natural Light
Colt 45 Old Milwaukee
Coors Olympia
Falstaff Pabst

Guiness Pabst Light

Heineken Light

Heineken Dark

Schlitz
Schlitz Malt

Labatts Stroh
Lite Stroh Light
Michelob

Any brand left out? Please specify.

0............. COOOOOOOOOOOOO

How often do you generally return beer containers to
the store?

Almost always

Generally

Infrequently

Almost never

C 119-121.

86

After consuming the beer you buy, how often do you
return the empty containers to the store? Check for

each use condition separately.

A. Small Indoor Parties:

 

Almost always
About three-quarters of the time
About half of the time

About one-quarter of the time

Almost never

 

 

B. Large Indoor Parties:

Almost always

About three-quarters of the time
About half of the time

About one-quarter of the time

Almost never

 

C. Picnics or Other Outdoor Parties:

_____Almost always
____ About three-quarters cf the time
_____About half of the time

About one-quarter of the time

Almost never

APPENDIX B

GROUPING OF QUESTIONNAIRE ITEMS

The questionnaire items were grouped into environmental

belief and behavior scales as follows:

Scale 501, Society Must Control Resource Use

 

 

5. More emphasis should be placed on individual's economic rights
than on society's natural resource rights.

28. More emphasis should be placed on individual's economic rights
than on society's energy rights.

30. Where energy sources are privately owned, the society should
have no control over what the owner does with them.

3. Where natural resources are privately owned, the society should
have no control over what the owner does with them.

6. Society must ultimately control what citizens do with the
nation's natural resources.

27. Society must ultimately control what citizens do with the
nation's energy resources.

Scale SOZgAResources More Important than Jobs

23. If an industry is to provide jobs in certain areas, we must be
ready to put up with some energy wastage.

10. If an industry is to provide jobs in certain areas, we must be
ready to put up with some natural resource wastage.

Scale 503ggIndustry_Will Not Act on Their Own

4. Industry will act in the public interest if the government
will let them. (Re: Natural Resources)

29.

Industry will act in the society's energy interest, if the
government will let them.

87

88

Scale 504, Consumers Must Conserve Resources

 

We must enjoy life with the natural resources we now have,
Fear of natural resources shortage should not discourage us '

We should turn to conserving natural resources only if it

 

The "scarcity of natural resources".is just a hoax played on

We will have plenty of natural resources, if we just invent

We will have plenty of energy sources, if we just invent

 

We are entering a period of scarcity and shortage of most

If we do nothing about it now, energy shortage will soon

 

Consumers should make every effort to conserve on energy use.

If the consumers tried to conserve energy, it would really

Consumers are largely responsible for the current energy

7.
and let the future take care of itself.

15.
from using natural resources and enjoying life today.

14.
does not change our life style.

Scale 505, Resource Shortage is Real

12.
the consumers.

22. The "energy crisis" is just a hoax produced by the oil
companies.

9.
new processes for finding and developing them.

21.
new processes for finding and developing them.

Scale 506, Resource Shortage is Serious

1.
natural resources.

18. If we continue our high levels of energy use, future
generations will not be able to have a level of living
like ours.

24. We are in an "energy crisis".

32.
become the major national problem.

Scale 507, Consumers Can Help

19.

26.
make a difference.

20.
shortage.

l3.

Consumers can do much to alleviate the natural resource shortage.

89

 

 

 

 

Scale 508, Industry Not Doing Their Best

16. Many companies are working together to conserve natural
resources.

17. Many companies are working together to conserve energy
resources.

31. Most energy wastage is caused by business.

2. Industry is largely responsible for the natural resource
problem.

11. Industry will never do anything in the society's interest
if it reduces their profits.

Scale 509, Resources Used in Cans Significant

41. The amount of metal used in making beer cans is so insignificant
that I would not worry about it.

43. The amount of energy resources used in beer cans is so insigb
nificant that I would not worry about it.

Scale 510, Prefer Glass Over Metal Containers

35. Metal containers should be replaced by glass containers
wherever possible.

33. The production of metal containers uses much more energy than
the production of glass containers.

38. I would not buy beer in metal containers if glass containers
are available.

42. The production of metal containers uses scarcer raw materials
than the production of glass containers.

Scale 511, Beer Companies Prefer Can Containers

36. Beer companies will not promote glass containers unless the
government forces them to.

37. Beer companies would rather promote metal cans than glass

.bottles, so they can make more profit.

90

Scale 512, Favor Container Return

 

 

 

 

40. Forced deposit has resulted in needless disruptions all around.

39. Too much of my money is locked up in container deposits.

45. I hate the thought of returning empty beer containers to the
store.

44. I should have the freedom to buy the beer container I like.

Scale 513, Certified Bottles Conserve Resources

34. "Certified bottles" (on which we pay a cash deposit of 5¢ each)
use much more energy than "uncertified bottles" (on which we
pay a cash deposit of 10¢ each).

46. "Certified bottles" (on which we pay a cash deposit of 5¢ each)
use much more natural resources than "uncertified bottles"
(on which we pay a cash deposit of 10¢ each).

Scale 514, Government is Doing Nothing

8. Government is doing nothing to alleviate the nation's natural
resources.

25. Government is doing nothing to alleviate the nation's energy
problem.

Scale 516, Frequency of Container Return

118. Frequency with which beer containers are returned to the store.

119. Frequency with which empty beer containers are returned to the
store after small indoor parties.

120. Frequency with which empty beer containers are returned to the
store after large indoor parties.

121.

Frequency with which empty beer containers are returned to the
store after picnics or other outdoor parties.

APPENDIX C

FACTOR INTERCORRELATIONS AND LOADING MATRIX

91

OH
OH
OH
ON
HO
ON
OO
OV
NO
OO
OH
OH
NH
OH
NH
HH

HH
NH
ON
OH
NH
VN

OH
ON
ON
NO

mv..

ON
OO
ON
NO
HV
OO
ON
HH
HH

NH
OI
OH
OH
NH
HH

NN

NN
HN
ON
OO
ON
OO
NN
OO
OO
OO
OO
OH
NN

OH
OH

VH
NH

NN
ON

OH
OH

NH

OH.

NN
OO
OO
ON
OO
NH

OH
ON
OO
NO
NN
NV
ON
NH
OH
HN
NH
NH
ON
OO
OO
ON

VH

OH
OH
OH
HN
VN
OH

OH

HH.

OH
NN
VN
NO
VH
NH
OH
VH
VI

NH
ON

NN
VH

OH

OH
ON
OH
OH
OH
HH

NH
HH
NN
NV
NO
OV
OH
VN
OH
OH
VH
ON
ON
NN
NO
NN

ml
OHI

OH
OH
HN
VH

OH

ON
VH
OH
HO
OO
OH
ON

NN

HN
VN

ON
ON

ON

HHI
ml
MI

HI
NH
OI

NH
NH

VN
OO
HO
OH
ON

OH
ON
NN
ON

HH
NI
OH
OH
OH

OH
OH
OH
ON
HN

OH
NH

ON
ON

OH

OH
OH
OH
HH

OH
ON

NN
NO

ON

ON
OH
NN
ON
OH
OO
ON
NN
HH
NH
VH
NH

VH

NO
ON
OO

HN
VN

NN

HmcommHO on» cH OuHHmcSEEOUO

NH
O

OH
O

HI
ON
NH
OH
NH
OI
NH
NH
O

ON
NN
VH
O

N

NO
NO

HO
ON

ON
ON

O

O
O

NN
NH
OH
ON
OH
NN
ON
NH
ON
HN
OH
OH
OH
ON
HO
HV
HO
OO
OV

O

NN
NH
ON
HN
ON
OH
O

OH
OH
ON
OO
ON
NN
VN
ON
NH
ON
OO
ON
HO
OO
OO
NO

OO

O

OH
ON
HN
VH
ON
N

OH
NH
NH
OH
OO
NN
NO
ON
NN
ON
NN
HN
ON
OO
OO
NO
HO

ON

O

N

HN
OH
NH
HO
OH
OH
VN
HH
OH
ON
VH
NN
ON
ON
ON
NO
VN
ON
OV
NO
HO
OO

O

ON
ON
OH
NO
VN
OH

HN

NN
NH
VH
OH

ON

OH
ON
NN

OO
ON

92

OI
OH
NN
ON
VO

HH

NH
OH
ON
ON

OH

OH
NH

H-
ON-
OH-
OH.

HH
OH

vH
HN
MI

ml-
OH
ON

ONI
NHI

NH

NI.
MI-
Nu...

NH
VH
ON

NI
OH

H

ml
Ml-
ml

OH
ON

ON.

NI.
NI

HI.
ml

tnLOOQ

OI
OI
OI
OHI

OH
NH

OH

O
mnl
HH

OH
ON

NH
NH

VH

OH
VN
ON
ON
HO
ON
VI
ON

O
NHI

HI
OH

OH
OH
HH
VH
ON

OH
OH
OH
OH
OH
NN
ON
NN
NN

OH
OH

OH

H

O‘O‘MOHO‘LnI-IIMI‘
I

OH
NH

VN
VH
NH
ON

HI-
NH

VH
ON
ON
HH
NH

HI-

OH

NH

VH

OH
OH

ON
HN

NN
ON

HI-

OH
NH

Nil
NH

ON
NH
ON
ON

ON
OH
OH

OH
OH

OO

OH

{\OHHMGOH

[‘O
NH

VH

OH
ON

OHI
HHI

OH
VH

VH
OH

VH
OH

NH

NH

ON

OO
OV

ON

OV
VO

VV
OV
OO
OV
NO
OO
NV
OO
OO
OO
OV
HV
HH

HO
NH
OH
OH

93

HV
NH
OI
ON
OH
OH

OO
OO

ON
ON
OH
ON

N

mvwa‘mhmm
I I

6‘

OH
ON

MI

HH.

VO

VO
VO

ON
ml

(‘0
H

OHMNMVO‘M

07’

ON

NH
HI
NO

NO
OO

NV

NN
NO

NI
NH

HI

NH

V' .
ON

NO
OHI
OV

NV
HO

OO
VN
HN
HV

O- O
NO ON
O- N
O- O
HH ON
OH O
OH ON
ON ON
OH ON
O OO
OH . ON
OH HN
NN OO
O O
N H
N NH-
N- O-
N- OH
O O
H O
O O
O OH
O N-
O- O
OH O

O

ON
OH
ON
HN
HO

ON

ON
NN
NN

v5...
NI

H-
H-
O-
7

ON

N

OO
NN
ON
HO

OHI

O
OH

OH
VH
OH
HH
OH
OH

VN
VN
OO
NN

mI

“I

NHI

NI

HH
OH

ml

OH

NH
OH

HH

HN
OH

ON
ON
OO

MI

ON

O

HN
OH
NH
NH

OO
VN

OH

ON

HH
OH
NN
OH

ON
NH

VN
ON

OO

OO

NN
OH
OH

ON
HO

HV
VN
OH
VO

OH
OH

OH

OO

O
NN
NH

ON
OH
VN

NN
NN

OV
HO
OO

NO
NH
VI
OI

OH

ON

O
ON

OH
OH
ON

NO
ON

OO
OO
OO
OO
HH

NH

NH

OHO
NHO
HHO
OHO
OOO

OOO
NOO

OOO
OOO

VOO
OOO
NOO
HOO
OHH
VNH
ONH
NNH
HNH
ONH
OHH
VO

OO

NO

HO

94

NI
ON
NH
NH
ON

HN

OH
OH

NH

OH
ON
NH
VH

OV

OI
OHI

HI
ON

ON..

ON

OH
OH

NH

HH
NH

NH
OH
ON
OH

HV

HI

VNI

ON

NI

ON
NN

OH

HHI

VH

v-IINO‘OI-Ilnomm

tom
NN

mkoNI-IV
m

In
r-I

OH

NI
NH

HO
NN

HH

OH

OH

VH
NH

NH

OHI
OH

OHI

NH
NH

ON
NN

OH
OH

OH
NH
OH

OH
ON
OH

OH
ON

OH

NI
HH

OH

OH

OH

O!
HHI

ON
NH

ON

NHI

OH

OH
OH
ON
NN

NN
OH
OH
OHI
OI

OH
OH

NN
OH

ON

OHI

ON

NH
OH

ON
HN

OH
ON

MI
NH
OH

NN
OH

OH
NH

ON
HN

OH

MI

NHI

NN

OH
OH

OH
OH
ON

ON
HN
OH

OH
ON

OV
OO

ON
HN

OH

HN

NH
VN

ON
HO
ON
ON
OO
VN

OH

HN
OH

HH
OO
ON

NN
ON
ON
HO

OH

VH

OH

mI

OO

ON
ON
OO
OO

NH
OH

HH
VH

ON
NH

NH
OH

OH

ON

ON
OO
ON
NN

NI
NN
OH

OH

OH
OH

HN

HH

OH

HHI

HN

NN
NH

VH
OH

ON

OH
ON
NN

OO
ON

OHO
NHO
OHO
OHO
VHO

95

OH
VN
ON

ON
OH
OH
NH
OO
OO

OI
OH

OH
ON

OH

NH
OH

OV

OH
VN
HO

HH
OH
ON

NN

OH
NN
OO
OO

HV

NI

OH

OH

MI

OH
NN
OV
VN
HN

ON

O‘I‘Nfl'
(DNIOH

OH

MI

MI

ON

VH

OH
OH
VH
OH

N

OH
ON

HI
OH

OH
OO

OO

OV
NO

HO

NO

OH

OH
OH
HH
OH

OH
NH
OH

HI
OH
OH

VN
NO

OV

OV
HV

NV

VN

OH
OH
ON

OH

OH
OH
VH
OH

OH
HH

OH
OH
OH
VV
NO

HV
NV

VV

OH

VH
OH
OH

OH
ON
OH

HH
NH
ON

HN
HO

NV
VV

NO

VH
OH

NI
HI
NI

NH

OH

OH
NH

OH
ON

ON

HN

OV
OO
OV

NO
OO
NV
OO
OO
OO
OV
HV
HH

HO
NH
OH
OH
ON
ON
OH
NO

VN
OH

96

NN . VH HH VH OH O VN ON VN OO OO ON VO OO ON OOO

OH . OI HI N NH HH VH OH ON OO VO OO OV OV OO OOO
HO OI H VI N N ON O VN OV NV OO OV ON V VOO
NH NN ON OO OO NO O OHI OI HI HH NH ON HH O OOO
NH: O O H HH VH OH V OH OH NH O HH O N NOO
NN HH VH O HN NH OH ON OH OO ON OH NV ON NN HOO
NH O O OI NH N O O O OH OI N HH O O OHH
NH O OH OH O O OI NI HI V O H N V O VNH
NI NI O NH O OI OHI O OI VHI HI OI OI O HI ONH
H OI O . NH. N NI OHI N OI NHI OI OI VI O O NNH
VH O H OI O OI VH NI O NN O V VH O OI HNH
N V N VI O HI O OI H VH O H VH O O ONH
OH N O N O O O NI N VH HI OI OH N HI OHH
OHI VH NI OI O HI N H N OH O H O OI NHI VO
O N H VI NI H O OI O V HI O NI VI NI OO
N OI OI OHI OI OI O O OH N OI OI VI OHI O NO
OI N OI N H O OI OI NHI O HH O O N H HO
OI O V HI O V HI NI NI OI OI N VI OI OI OO
OI OI HI O OI VI OI O V OI N O O O OI OV
O OH OI O ON ON VI HHI HHI NI VI OHI OI OI HHI ON
O ON O OH OH VH VI OI OHI H N OI O V NI O
OH H OI N O O V OI O HI O O V O O OV
OH HI OI VI OI OI H O OH OI VI N H HI NI VO
HN OH OI N O O O OI V O NH O HH O OH VV

HV HH N _ HO NH OH OH ON ON OH NO VN OH H HN

97

N NI, HH HH H N NH N NH ON N NH NH N N «H NN
N NHI. N N N HI NH O NH OH HH «H ON N NH NH «

N OI H N N OH NH N N NH HI N O N N N OH
O NI N N . N NH NH NH N NH H N O HI O N NN
«I «I N NI N « OH NH «N .«H NH ON N N HI NH NN
NI NI OI O N O O «H NN NN HH NH N N HI NH O
HI N OHI «I N O ON HN NN NN O N N O HI O N

« O NI NI NH N NN NH NN NN N N HH H O N ON
NH OI N HI O O NH NH NH HN N N N H H O NN
NI OHI HHI NI . N N NH «H «H NH N O O «I O N N
ON N« NN N O« «N «« N« NN O« NN ON N« NN NN NN

NH NH O N NI NH N N O _ N OH NI N HH N O NHN
HH N NI NH NH O NI HHI O «I HHI N NI NI N O NHN
NH NH N N NI N HI HH NI N ON N O NH « NI OHN
HHI HHI N «I NI HI NI O OI « « HI N N NI OHI NHN
«I N NN H NH NN NN NI NI NHI HI HI OHI H NI NHI «HN
NN NH . O NI N NI NI N NI N «I NI OH N N OI NHN
NN ON « NI N N N N O N NH NH NH «N NH ON NHN
N ON «N HH HH NH OH NI N NI « NH «H OH NH « HHN
NN N« NN NH NN NH OH HI HI N N NH OH OH HH N OHN
NN NN NI O HH N «I HH N NH «N NN NH ON N NI NON
NO NH H« NN N« NO «N NHI NI NNI NI N N NH ON N NON
OHI NH NI HI NN «HI ONI O« NN NO HO NN H« N« «N NH NON

OV HV HH N HO NH OH OH ON ON OH NO VN OH H HN

98

OI
OH
VH

N-
N.
N-
N.

VI

MI

ml.

MI

MI

OO

MI

OV

vi.

O
HH

ON

OH
OH
VH

MI
OHI

VH

OH
OH

MI.

OV

ml
VI

HI
FI
Ml.

VO

HN

NH
NN

HN

OH
ml

NH

HH

OH
OH

HH
VH

OH
OH

VV

NN
OH
OH
NH
ON
NH

OV

VH
VN

VN

OH
OH

NH
NH
«H
NN
«H
NH
«H

OH

OO

OH

ON
Hm
NI

NO‘OI‘ID

OH
OH
ON
OH
OH
VO
HN
ON
ON
ON
NN

OV

ON
OO

HH
NH

HH
mI

MI

OH
OH

MI
MI

NI

NO

ON
OO

OH

OH
OH
OH
VH
OH
OI

ml

VH
HH

OH
NH

HH
MI

NI
NI

OO

MI
MI

OH
OH

NV

\DVI—IMNOROFHO

0M”?

N
I

m
('1

ON

OO
OH

OH

OO

OO

NV
NH

NN
ON

OH
NI

ml

OO

OO
OO
OV
HV
HH

HO
NH
OH
OH
ON
ON
OH
NO
VN
OH

HN

NN
NH
VH
OH

99

VI OI
NI OI

NI O

HH ON

VHI OI

OOH NNI
NNI OOH

O OI
N HI

NI OI
OI OI

OHI NI

OO OV

NI

I‘NI-I

VI

OI
HO
OO
NI

HI

VI

HI

“I

OH

ON

HHI
NI

NI
OO
HO
OI

ml
MI

VH

HH

NI
OI
VO
HO

HH
OH

mI

OH

OV

@l
OI
HO

VO
HI

OH
NH

VI
OH
HH

VO

NI
HHI
OHI
VH
HH
ON

HI
OH
ON
OO
ON
NH
HN

VV

NH

NI
ON
OH
HN

HI
NHI
VI

HH

ON
OH

OHI

NI

NI

HI

mI
OH

OO

HH
OH
OHI

NH
ON
VO
NO
OO
NH

HH

OV

I"

NO
OO

HH
HH

NO

NH

OH
VH

OO
NO

HH

OO

HmONmm

ml

MI
OH
OH

OH
HH
HH

NH
OH

N«.

OH

HH
HH

OH
ON

OO

HI
NI

mI
GI

vI

VH

HN

OH
ON
ON
NN
HN

OO

I-I

NmmmmI-IHNMLO

@030
'7

N .4
NI

NH
OH

VH

OO
OO

NN
OO

OO

VNH
ONH
NNH
HNH
ONH
OHH
VO
OO
NO

HO

OV
ON

100

HI

HI
NN

ml
VH

NN

NHI
HHI
ONI
NHI

MI

NN.

OH
HI
NN
ON

OH
NI

OH

OV

VH

NHI

OH

OO
ON
ON

NH

OH
VN
NN
OH
NN
ON
VH

VV

ON

VN
Ml
ml

OO

OH
ON

OH
VN
ON
ON

NH
ON
ON

OV

NN

NN
MI

OH
NO
OH
OH
OO
NI
OH
OH
HO
NN
NH

OO
NN

OO

VH

NI
OH
ON
NH
NH
NO

OH
ON
OV
OO
ON
OH
OO

OV

OH

OH
ON
NO
OH
ON
OI
NN

OO

OHI

NI
OH
VH
HH
VO
NN
OH
OI
OH

HH
VN

HH

OHI

NV

OH

mI
HH

OH
NV

VN

HN

OH

VN

NH

OH

OO

OHO
NHO
OHO
OHO
VHO
OHO
NHO
HHO
OHO
OOO
OOO
NOO
OOO
OOO
VOO
OOO
NOO
HOO
OHH

101

ON
ml

VN
NH
ON
NN
NN
VN
ON

ON
VN
OO
HO
OO

OOO

V
OH
OH

NH

..HH

OH

NN
HN
VH
HN
NN
ON
OO
OO

OH

OH
OO
OO

NOO

ON
OH
OO
ON
OH
NV
ON
NN
ON
OH
NO
HV
NN
OV
OO
HO
NN
NO
NO
OO
VO
HO
NO
OO

HOO

QBMO'NVVQB.I\\OOM

H
r-I

03m
r-I

VH
NH
HH

OHH

NI
HI

MMHNFMMMOVI‘I—Im

0‘ LG
I I

OH

MI
VI

VNH

O

ONH

N
MI
NHI

MI
vI

N
H
I

MI
HI

NNH

HH
OH
N.I
O H

OH
OH
OH
OH
NH

HNH

NI

NI
NI

OH
OH
NH

ONH

NI

VH
HI

OH

HI

MI

NH

HH

VH
OH
OH

NH

OHH

ml
NHI

HH

ml
VHI

OH

VO

VH
HI

OH

OHI
HHI

NOW)

NO

OI
NHI

NI
OHI
OI
OI

ON

OH

NO

VN
OH

HN

NN
NH

OH

ON

OH

ON
NN

OO
ON

102

OHI
OH
OH

OH

NH
ON
NH
ON
VN
NH
VH
NH
ON
NH
NN
ON
OO
OO
NO

OOO

N N :6 N-
H N .4

0mm
v-I

r4 u» cw N (V an «I u» O IN
p4 ~I

Ofi'r-l
I-II-II-I

NOO

ON
ON
OO
OO
OH
OH
HH

OH
ON
NN
HH
VH

HN
NH
OH

HOO

HI
NI

vH
ON
NN

MI

OHI

mI

WI

OH

NH

Ml
NH

OHH

OI OI
NI OI

HI H

NI HHI

NH O

VH N

OH NI
NH NI

OH O

OH O

OH O

NH NI

OH O
OH NH

VNH ONH

NI

HHI
OHI

OHI

NNH

OI
vH
ON
ON

«H
VH

mI

MI
VH

HNH

ONH

OHI

ON

HN
ON
OH

OH
HI

HH
ON

OH

OHI
OI

ml

OHI
OHI

VO

\9
I
(D
I

H
O

mfl'NLnCrr-Imom
6‘

m
I
H
H

I-INOInlnm
,4

OO NO

H
0|

0|

MI

HO

OO
Ov
ON

Ov
VO

vv

Ov
OO
Ov
NO
OO
Nv
OO
OO
OO
Ov
Hv
HH

HO
NH
OH
OH

103

ON
NN
OO
ON

NO
OHI
HN

NO

OOH
ON
Ov

HH

NI

OH

NI
HHI

OOO

OH

.ON

OH

.3

OH
OH
HO

ON
OOH
OO

NH
OH
HH

ml

NOO

HH
Ov
HN
VH
NO
ON
Hv
Ov

,3

OO

Ov
OO
OOH
NN

OI
ON
vH
ON
NI

VI

HOO

OOH
NH

.-OHI

ml
mv
fim

OHH

HH
mI

NHI
OH
HO
ON

OH

«NH

mI

OHI
HO

NO

OHI
«HI
NHI
OHI
HHI

ONH

NNH

NI

mm
mv
2
3-
3..

.3

OO

HOMM

HNH

ml

NI
OH

Lnr-IINN

ONH

HI
ON
OH

OH

HH
NH

NH

OH
NH
ON.
HO

NHI
OHI
OO
vN

OHH

NI
OHI
OHI

OOH
OH
HH

«O

NH

OHI
HHI
ml

OH
OOH

OO

VHI

OHI

NI

V N
l

O‘HOI-IOv-IFCDNO

OOH

NO

mI

OHI
OH

OHO
NHO
HHO
OHO
OOO
OOO
NOO
OOO
OOO
VOO
OOO
NOO
HOO
OHH
VNH
ONH
NNH
HNH
ONH
OHH
«O

OO

NO

HO

104

OOONOONOV'ONNOO

HNV
HHH

OHO

NHO

NH

OOO

NI

HN
OH

OH
OHO

OHI

NOO

OHI

.OI

NHI
NHI
HI

NI
OHO

NHI
OHI
VNI

HHI
OI
OHI
VH
HH

HHI
VHO

OOH

\D

OHO

NH
NO
OH

HI

VNH

ON
HV
ON
ON
ON

NN
ON

ON
OH
OH

OH
VN
ON
OO
OO
NN
ON
NHO
OHI
OO
NHI
HHI
NHI

ONH

V
NH

mI

MI

OH
NH

HN
OH

OH
NH

HHO

NNH

N
ON
MI

ON
ON
OH

OH
HH

HH

OHO

OV

ON

HNH

NI
ON
HH
NH
NO

HH
ON

HN
VH
VH

HH
NH
OH
NN
VN
ON
OH

OOO

VO
HHI
NO

ONH

HO
VV
OH

2
mm
3
m:
3
2

OOO

HO

VO

OHH

OH
OH
ON
VN
ON

NOO

OHI

NO
VH

VO

ON
OO
VO
NO
NV

ON
NN
ON

OH
OH
OO
ON
ON
OO
NN
NO

OOO

OHI

HO

OO

3
mo
3
oo
8
S...
mm

OH
NH
VN
NH
HO
ON
NN
ON

OOO

ON
ON
NV
OO
NV
OO
ON
OO
VN
ON
OH
VN
HV
OV
OV
OO

VOO

HO

HN

NN
NH
VH
OH

ON

OH
ON
NN

OO
ON

OHO
NHO
OHO
OHO

VHO

105

ON O A VN OI OI O OO O OH ON V OH VN ON ON

NN o NN N- «- oH Nm 0H NH om N- mH mH Hm NN
m _N «H o- N- mH mN NH NH Nm o mH mN o« mm
m- «- m o m oH- mH mN Nm m «H o «H H m
N m- m N mH N «H mN Nm oH mN m- NN « o
.OH- ..m N «- N- «H «H HH «m NN OH m- mH N HH
m- mH H HH- m NH m mH «« «N NN m m H- H
m- H m- m- H HH NH Nm «m NH «H HH- « NH o
N oH « m- HH m mH N« mo «N HN NH «N N H
NH HH mH HH-. «- NN mm m mm Nm N- NN NN NH mm.
NH m NH HH- .m «H om ON N« Nm « NH NN mH on
o m- m m «N o « «N NN m- H« m- «H m- m-
m NH m «- H m- m- HH NH o Nm H- HH H- H
m- NH m- m- mH N m HH mN HH N« NN- «H N «-
NH o m .H- mN m- m mH NH m N@ «H- 0H NH N
N N- H- m- NN m- N «H «H «- «m oN- m HH N
m HH- HH 0 m- m m m- H- HH mH- «« «N «H mN
o o m- o- m- N- o m H- m m- mm «N NH m
m «- m « NH- m m m- N mH NN- mm «N mN «N
oH HH- oN « H- «- NH « m «N m- Ho mm om mm
m- N N H- H- m- NH «H «H NN m mm mm «m N«
N m-. o N oH- oH NH «H «H mH m H« oN om mm
HH m- «H N H m «N «H «H oN «H N« «o m« m«
m N « m- m- m NH mH HH m «N «N ow m« NN

OHO NHO OHO OHO VHO OHO NHO HHO OHO OOO OOO NOO OOO OOO VOO

106

m O OH H ml m ON ON ON ON HN Om OOH Oh Om OOO

O _O N ,O- ON- O NO N NH ON N OO ON OOH HO NON
OH N- NH O . OH- N- OO O N ON N NO ON HO OOH OON
N N N O- NH O NN NN NN NN NO OH- HN O NN NON
O OH- N O N NH ON N O . NH NH OH NH . NH HN NON
NN ;O- NN O- O- HH OO HN OH NN ON HO OO NO OO HON
OOH H- NN H- N- O ON N O- NH OH NH N O OH OHH
NH NN . OH N- H- O OH NH ON OH HH H- O O O ONH
OH- NO NH- HH- NH- N N- O- O H- N NH- N- O O- NNH
N- ON NH- O- O N O- OH- O O N OH- N- H- O- NNH
OO O- ON N. .N O NN O N- NH O- NH HH N OH HNH
ON HH- NO N O N- OH O N- OH H N O O NH ONH
HN O- OO N O H- ON OH N NH HH NH O O NH OHH
H OH- N NO OH OH- HH- N O- NH- O O N O- NH ON
N NH- O HN N- NH N O N N H- N O O NH NN
N O O NN O- N O HH- O H OH- O OH- O N- NN
N- O O NN O N- O NH N- N- O OH- NH H H- HN
H N- O NN H H N- H NH- OH- O N- N- O N- ON
N- O- N- ON O- O- HH- N O- O- O- O H O- N- OO
H- O- O H- NN N- N- OH N O NN NH- HH- ON- , NH- NN
N- N- O H NN O- N- NH NH H- ON O- N OH- NH- O
N H H O- N- ON NH H- NN NN N O N OH N- OO
O N N- O OH- ON O O- OH NN O- O H H- N- ON
OH NH- OH O H N ON NN NN ON NH N OH ON NN OO

OHm hHm OHO mHm va MHm NHm HHm OHm mom OOm hOm OOO mom vOm

107

OOH

mH
OH
NH

OHm

HI
OOH
HHI
NHI

5H
OH
NH
NHI

hHm

mm
HHI
OOH

NI

ON

HI:
NH

vH

OHm

NHI

OOH

NHI
MH-
NI

mHm

Nil

OOH

mI
OH
OH
HI
hm
VHI

OHm

OHm

ON

OH

NHm

N OI mH OH

VI bH OH NH

0 HI 5H. m

m NHI MHI NI

OH OH HI hmI
OI mN ON VI

HN 5N MO 5
OOH ,Om hH Om
Om OOH Ov vm
5H Ov OOH N
Om vm N OOH
NI N vN mNI

HHm OHm OOm OOm

MH
NHI
vH

VHI

OH

QN

mNI
OOH

hOm

ON
mN
ON
ON
HN
Om

OOm

OmI

NV

MH
ON

0v

mOm

OH
ml
mH

OHI

NI

0v

Om

mv

vom

OHm
bHO
OHm
mHm
va
mHm
NHm
HHm
OHm
OOm
OOm
hOm

BIBLIOGRAPHY

BIBLIOGRAPHY

Ajzen, Icek and Fishbein, Martin. "Attitude-behavior Relations:
A Theoretical Analysis and Review of Empirical Research."
Psychological Bulletin 15 (1972): 888-918.

/Albrecht, S.L.; DeFleur, M.L.; and Warner, L.G. "Attitude-behavior
Relationships: A Reexamination of the Postulate of
Contingent Consistency." Pacific Sociological Review 15
(1972): 149-168.

Allport, G.W. Personality: A Psychological Interpretation.
New York: Holt, 1937.

,/Arbuthnot, Jack. ”The Roles of Attitudinal and Personality variables
in the Prediction of Environmental Behavior and Knowledge."
Environment and Behavior 9 (1977): 217-232.

 

Arbuthnot, Jack and Lingg, Sandra. ”A Comparison of French and
American Environmental Behaviors, Knowledge, and Attitudes."
International Journal of Psychology 10 (1975): 275-281.

 

Bogardus, 8.8. Fundamentals of Social Psychology. 2nd ed.
New York: Century, 1931.

 

Brookfield, H.C. "0n the Environment as Perceived."
In Progress in Geography: International Reviews of
Current Research. London: Edward Arnold (1969).

 

 

I/Bruvold, William H. ”Belief and Behavior as Determinants of
Environmental Attitudes." Environment and Behavior
5 (1973): 202-218. '

 

Canter, D.V. "Architectural Psychology." Proceedings of the
Conference held at Dalandui, University of Strathclyde
(February-March 1969). London: RIBA Publications (1971).

Collins, A. and Loftus, E. "A Spreading Activation Theory of
Semantic Processing." Psychological Review 82 (1975):
407-428

 

Collins, A. and Quillian, R. "HDw to Make a Language User."
In Organization of Memory. Edited by E. Tulving and
W. Donaldson. New York: Academic Press, 1972.

108

109

Craig, 8. "Consumer Reactions to Price Changes: An Experimental
Investigation." An unpublished Doctoral Dissertation,
Ohio State University, 1971.

Crespi, I. "What Kinds of Attitude Measures are Predictive of
Behavior?" Public Opinion Quarterly 35 (1971): 327-334.

Dispoto, Raymond G. ”Interrelationship Among Measures of Environ-
mental Activity, Emotionality, and Knowledge."
Educational and Psychological Measurement 37 (1977):
451-459.

/’Fishbein, M. "The Prediction of Behaviors from Attitudinal Variables."
In Advances in Communication Research. Edited by
C.D. Mortensen and K.K. Sereno. New York: Harper and Row,
1973.

 

a/Fishbein, Martin, ed. Readings in Attitude Theory and Measurement.
New York: John Wiley and Sons, Inc., 1967.

Fishbein, M. and Ajzen, I. "Attitudes Towards Objects as Predictors
of Single and Multiple Behavioral Criteria.”
Psychological Review 81 (1974): 59-74.

Fishbein, M. and Coombs, F.S. "Basis for Decision: An Attitudinal
Analysis of Voting Behavior." Journal of Applied Social
Psychology 4 (1974): 95-124.

 

Folsom, J. Social Psychology. New York: Harper, 1931.

 

Heberlein, Thomas A. and Black, J. Stanley. ”Attitudinal Specificity
and the Prediction of Behavior in a Field Setting." Journal
of Personality and Social Psychology 33 (1976): 474-479.

Heise, D.R. Causal Analysis. New York: Wiley-Interscience, 1975..

 

Hunter, J.E. and Cohen, 5.8. "Package: A System of Computer
Routines for the Analysis of Correlational Data."
Educational and Psychological Measurement 29 (1969):

Hunter, J.E. and Gerbing, D.W. "Unidimensional Measurement, Second
Order Factor Analysis and Causal MOdelSC' In Research in
Organizational Behavior, Vol. IV. Edited by Barry M. Staw
and L.L. Cummings. Greenwich, Conn.: JAI Press, Inc.

(In Press).

110

HUnter, J.E. and Hunter, R.F. "PATHPAC: A program for path analysis
for recursive systems using either given path coefficients or
using calculated OLS path coefficients from given causal
influence indicators or using path coefficients calculated
from given causal influence indicators by a newJProcrustes
least squares method.” East Lansing: Michigan State '
University, 1977.

Hunter, J.E.; Levine, Ralph L.; and Sayers, Scott E. "Attitude
Change in Hierarchical Belief Systems and Its Relationship
to Persuasibility, Dogmatism, and Rigidity." Human
Communications Research 3 (1976): '3-28.

 

Janis, I.L. and Hoffman, D. "Facilitating Effects of Daily Contact
between Partners who Make a Decision to Cut Down on Smoking."
Journal of Personality and Social Psychology 17 (1970):
25-35. - ‘

Kelman, Herbert C. "Attitudes are Alive and Well and Gainfully
Employed in the Sphere of Action." American Psychologist
29 (1974): 310-324.

Kothandapani, V. "Validation of Feeling, Belief, and Intention to
Act as Three Components of Attitude and Their Contribution
to Prediction of Contraceptive Behavior." Journal of
Personality and Social Psychology 19 (1971): 321-333.

Lindsay, P. and Norman, D. Human Information Processing: An
Introduction to Psychology. New York: Academic Press,
1972. ’

Liska, Allen E. ”Attitude-behavior Consistency as a Function of
Generality Equivalence between Attitude and Behavior Objects."
Journal of Psychology 86 (1974): 217-228.

McGuire, W.J. "A Syllogistic Analysis of Cognitive Relationships."
In Attitude Organization and Change. Edited by-
M.J. Rosenberg, et al. New Haven: Yale University Press,
1960.

McGuire, W.J. "Theory of the Structure of Human Thought." In
Theories of Cognitive Consistengy. Edited by R.P. Abelson,
et al. Chicago: Rand McNally, 1968.

Maloney, M.P. and Ward, M.P. "Ecology: Let's Hear from the People."
American Psychologist 28 (1973): 583-586.

Nunnally, Jum C. Psychometric Theory. 2nd ed.
New YOrk: McGraw—Hill Book Company, 1978.

111

//Perry, Ronald W. "Attitudinal Variables as Estimates of Behavior:
A Theoretical Examination of the Attitude-Action Controversy."
European Journal of Social Psychology 6 (1976): 227-243.

 

/’Poole, Marshall Scott and Hunter, J.E. "Behavior and Hierarchies_
of Attitudes: A Deterministic Model." In Messsge-Attitude
Behavior Relationshipi Theory, Methodology, and Applications.-
Academic Press, Inc., 1980.

 

 

Rokeach, M. Beliefs, Attitudes, and Values. San Francisco:
Jossey-Bass, 1969.

 

I/Schuman, Howard and Michael P. Johnson. ”Attitudes and Behavior."
Annual Review of Sociology 2 (1976): 161-207.

Thomas, W.I. and Znaniecki, E. The Polish Peasant in Europe and
America. Vol. 1. Boston: Badger, 1918.

 

/Tittle, Charles R. Hill, Richard J. "Attitude Measurement and
Prediction of Behavior: An Evaluation of Conditions and
Measurement Techniques." Sociometry 30 (1967): 199-213.

 

Udell, Jon G. "Can Attitude Measurement Predict Consumers Behavior?"
Journal of Marketing 29 (1965): 46-50.

 

Veevers, J.E. "Drinking Attitudes and Drinking Behavior: An
Exploratory Study." Journal of Social Psychology
85 (1971): '103-109. -

Weigel, Russell H. and Newman, Lee S. "Increasing Attitude-Behavior
Correspondence by Broadening the Scope of the Behavioral
Measure." Journal of Personality and Social Psychology
33 (1976): 793-802.

Weigel, R.H.; Vernon, D.T.A.; and Tognacci, L.N. "The Specificity
of the Attitude as a Determinant of Attitude-Behavior
Congruence." Journal of Personalitygand Social Psychology
30 (1974): 724-728.

WOod, G. "Organization Processes and Free Recall." In
The Organization of Memory. Edited by E. Tulring and
W. Donaldson. New York: Academic Press, 1972.

"IIIIWMIHIIWIITI‘IIW