LIBRARY
MichIgan State
1 University

 

 

 

PLACE II RETURN BOX to romavotnio chockout from your noord.
To AVOID FINES rotum on or baton doto duo.

DATE DUE DATE DUE DATE DUE

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU is An Aifinnotivo Action/Emu! Opportunity Institution
WM!

THE E

 

THE DEVELOPMENT OF MULTI-CHANNEL TV AND TV NEWS VIEWING:
A CROSS-SECTIONAL ANALYSIS OF TV NEWS VIEWING
AND NEWS LEARNING IN A BROADCAST-CHANNEL-ONLY
AND MULTI-CHANNEL SITUATION

BY

Sug—min Youn -

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Mass Media

1993

THE

 

multi-
from 1
gratit
ChOiCt
liter:
news ‘
Then
facto

due t

ABSTRACT
THE DEVELOPMENT OF MULTI-CHANNEL TV AND TV NEWS VIEWING:
A CROSS-SECTIONAL ANALYSIS OF TV NEWS VIEWING

AND NEWS LEARNING IN A BROADCAST-CHANNEL-ONLY
' AND MULTI-CHANNEL SITUATION

BY

Sug-min Youn

This dissertation investigated how the development of
multi-channel TV affects TV news viewing and news learning
from TV. Based on knowledge gap studies, uses and
gratification studies, expectancy value theory, program
choice studies, and relevant cognitive psychology
literature, the theoretical factors which are related to TV
news viewing and news learning from it were identified.
Then, the changing nature of the relationships between these
factors and TV news viewing and news learning from TV news
due to multi-channel TV development was examined.

Since multi-channel TV has been introduced all over the
U.S. and multi-channel TV subscription cannot be
manipulated, the major analysis of this research was limited
to finding cross-sectional differences between broadcast-
channel-only TV viewers and multi-channel TV viewers. A
cross-sectional telephone survey was executed and obtained a
sample size of 208 multi-channel TV subscribers and 95 non-
subscribers.

With regard to the amount of TV news viewing, the
results indicate that (a) viewer availability and content

gratification expectancy of TV (CGE-TV) which are major

~._ r_.fiw__,.u._,-

facto‘

i

situat

 

situat
ofTV

news C

. _-
u..-

vievin

 

develd

 

maximi

I
Percei
only 1
and t]

grati:

0f rue
AlSo,
tel1ms
news

OtheI

factors of TV news viewing in a broadcast-channel—only
situation become less important in a multi-channel
situation, and (b) relative content gratification expectancy

of TV (RCGE-TV), channel repertoire, and 33123,2359993192.9§

news channels emerge as new important factors of TV news
GIZEI§§TI—;;;;, the results showed that multi-channel TV
development tends to facilitate "(program type) interest
maximization."

The results indicate that (a) multi-channel TV viewers
perceive TV as a better news medium than broadcast—channel-
only TV viewers in terms of news variety, depth of coverage,
and the expectancy of news learning motivation
gratification, but contrary to the original expectation (b)
TV news viewing has a stronger relationship with the level
of news learning in a broadcast-channel-only situation.
Also, this study provides substantial evidence that, in
terms of the baseline awareness of major news events, TV

news viewing is as important as, or more important than,

other media exposure.

I
the ma
First,
Disser
(Telec
assist
Disser
Dr. Hc
( Comm

the in

 

T
Univeg
the ph
SOCial
assist

nembeI

ACKNOWLEDGEMENTS

I owe many people special thanks for helping me through
the mass media Ph. D. program at Michigan State University.
First, I would like to thank the chairperson of the
Dissertation Committee, Dr. Thomas F. Baldwin
(Telecommunication), for his valuable guidance and
assistance. I am very thankful to the other members of the
Dissertation Committee, Dr. Straubhaar (Telecommunication),
Dr. McCarty (Telecommunication), and Dr. Dearing
(Communication). The superb expertise of each member became
the indispensable nourishment for my dissertation.

The Department of Telecommunication at Michigan State
University provided the financial assistantship throughout
the Ph. D. program. The Institute for Public Policy and
Social Research (IPPSR) at Michigan State University
assisted me with data gathering. I am grateful to the staff
members in these two institutes.

Many other people helped me with my Ph.D work.
Specially, there are two persons to whom I always wanted to
express special thanks; Kil-ho Kang and Soc-won Lee.

Without Kil-ho Kang (and his wife Keong-sin), my university

senior both at Seoul National University in Korea and

Michigan State University, my study and life in the U.S.

would have been much difficult and lonely. Particularly,
iv

when I was tackling the tough first year courses, his help
was absolute. Soc-won Lee, my longtime college friend, has
always been beside me with full of encouragement, jokes, and
wisdom. Last year, he went through the harshest moment in
his life and we lost contact temporarily. Recently, he came
back with sincerer encouragement, funnier jokes, and
shrewder wisdom.

Finally, I would like to thank my family members. For
my parents, Kyo-jung Youn and Yong-jeong Kim, I simply
cannot find any adequate expression of appreciation. Like
most Korean parents, they devoted their entire lives to
their four sons. Without their endless faith, love, and
support for me, their second son, this work would never have
been completed. My daughter, Yeon. How happy I was when
you were born two and half years ago. How difficult it was
to bring you up as both your mother and father being full—
time students. Nevertheless, Yeon, it was you who filled my
days and nights with pure bliss. And, there is my wife,
Hye—ok. She was in the center of everything. She has
carried the heavy burden of a full-time foreign student,
housekeeper, mother, and wife with remarkable patience.

From the never-sufficient time and money, she showed the
marvelous skill of maintaining our happy little home. As a
conventional oriental man, I hate to say it but I don't see
any other adequate word of appreciation for my wife. "I

think, Yeon's mother, . . . I love you."

 

 

Page
LIST q-
LIST 0

cnmd

 

II.

III

Page

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O O O O O O O O O 0

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

CHAPTER

I. INTRODUCTION . . . . . . . . . . . . . . .

Research Background. . . . . . . . . .
Research Purposes. . . . . . . . . . .
Research Questions and Organization of
Dissertation . . . . . . . . . . . . .

II. LITERATURE REVIEW. . . . . . . . . . . . .

TV News Viewing. . . . . . . . . .
Knowledge Gap Studies . . .
Uses and Gratification Studies
Program Choice Studies. . . .
TV News Viewing Models. . . .

TV News Viewing and News Learning.
Socio-Economic Status (SES) .
News Learning Motivations .
TV News Viewing . . . . . .
Other Factors . . . . . . .
News Learning Model . . . .

III. RESEARCH METHODOLOGY . . . . . . . . . . .

Introduction . . . . . . . . . . . .
Measurement of Key Variables . . . . .
Socio-Economic Status . . .

Content Seeking and Process Seeking

CGE-TV and RCGE-TV. . . . . . . .
Viewer Availability . . .
News Channel Awareness. . . . . .
Channel Repertoire. . . .
Value Perception of News Channels
Amount of TV News Viewing . . . .
Amount of Other Media Uses. . . .
Level of News Learning. . . . . .

vi

viii

16
32
37
44
45
52
54
61
62

64

64
64
65
65
67
67
68
68
68
69
70
70

TABLE

Chapte
Page

IV.

APPENi
APPENI
BIBLIc

 

TABLE OF CONTENTS (Continued)

Chapter
Page

Research Design. . . . . . . . . . . .
Selection of Sample . . . . . . .
Administration of the Survey. . .
Composition of the Sample . . . .

Analytical Techniques Used in Hypothess

Testing 0 O O O O C O O C O O O C
IV. RESEARCH RESULTS AND HYPOTHESES TESTING. .

Introduction . . . . . . . . . .
Amount of TV News Viewing. . . . . . .

Average Amount of TV Viewing and TV News

Viewing . . . . . . . . . . . . .
TV News Viewing in a
Broadcast-Channel-Only Situation.
Broadcast News Viewing in a
Multi-Channel Situation . . . . .

lS

24-Hour News Viewing in a Multi-Channel

Situation . . . . . . . . . . .

Hypotheses Testing. . . . . . . .
TV News Viewing and News Learning . . .

News Learning Level of

Four News Events. . . . . . . . .

News Learning in a

Broadcast-Channel-Only Situation.

News Learning in a

Multi-Channel Situation . . . . .

Hypotheses Testing. . . . . . . .

V. DISCUSSION AND CONCLUSION

Introduction . . . . . . . .
Interpretation of Key Findings . . . .

Development of Multi-Channel TV and

TV News Viewing . . . . . . . .

Development of Multi-Channel TV and

News Learning . . . . . . . . . .

Contributions of This Research . . . .
Suggestions for Future Studies . . . .

APPENDIX I. THE DESCRIPTIONS OF 6 NEWS EVENTS .

APPENDIX II. TELEPHONE SURVEY QUESTIONNAIRE. .

BIBLIWRAPHY O O O O O O O O O O O O O O O O O 0

vii

72
72
76
77
80
81

81
81

82
85
91
94
97
105
105
106

110
116

122
122

123
132
137
139
142
144

164

 

 

Table
Page

5.
6.

7.

10.

11.

12.

13.

LIST OF TABLES

Program Choice Options of Broadcast-Channel-Only
TV Viewers . . . . . . . . . . . . . . . . . . . .

Program Choice Options of Multi-Channel TV Viewers
Channel Available for Broadcast-Channel-Only
Viewers and Multi-Channel TV Subscribers in the
Sampling Areas . . . . . . . . . . . . . . . . . .

The Major Production Statistics of the Sample. . .

,Sample Composition . . . . . . . . . . . . . . . .

Average Amount of TV Viewing a Day . . . . . . . .
Average Number of days of TV News Viewing. . . . .

Correlation Matrix of Variables Related to
the Amount of TV News Viewing in a
Broadcast-Channel-Only Situation . . . . . . . . .

Multiple Regression Analysis Results of
the Amount of Broadcast News Viewing in
a Broadcast-Channel-only Situation . . . . . . . .

Correlation Matrix of Variables Related to
the Amount of TV News Viewing in a Multi-Channel
Situation 0 O I O O O O O I O I C O O O O O O O I 0

Multiple Regression Analysis Results of
the Amount of Broadcast News Viewing in
a Multi-Channel Situation. . . . . . . . . . . . .

The Correlation between Independent Variables and
the Amount of 24-Hour News Viewing . . . . . . . .

Multiple Regression Analysis Results of

the Amount of 24-Hour News Viewing
in a Multi-Channel Situation . . . . . . . . . . .

viii

17

18

74

78

79

84

84

86

88

93

95

98

98

17.

 

20.

21.

22.

23,

24

25.

26.

 

Table

Page

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

The Relationships between SES and the Amount of
TV News Viewing. . . . . . . . . . . . . . . . . .

The Relationships between CGE-TV and RCGE-TV and
the Amount of TV News Viewing. . . . . . . . . . .

The Relationships between Viewer Availability and
the Amount of TV News Viewing. . . . . . . . . . .

News Learning Level of 4 News Events . . . . . . .

Correlation Matrix of Variables Related to
the Level of News Learning in a
Broadcast-Channel-Only Situation . . . . . . . . .

Multiple Regression Analysis Results of
the Level of News Learning in a
Broadcast-Channel-only Situation . . . . . . . . .

Correlation Matrix of Variables Related to
the Level of News Learning in a Multi-Channel
Situation 0 O O O O O O O O O O O O O O O O O O O 0

Multiple Regression Analysis Results of
the Level of News Learning in a Multi-Channel
Situation 0 O O O O O O O O O O O 0 I O I O O O O O

The Relationships between SBS and the Level of
News Learning. . . . . . . . . . . . . . . . . . .

The
and

Relationships between News Learning Motivation
the Level of News Learning . . . . . . . . . .

The
and
the

Comparison of Broadcast-Channel-Only Viewers
Multi-Channel TV Viewers Regarding
Evaluation of TV as a News Medium. . . . . . .

The
the

Relationships between TV News Viewing and
Level of News Learning . . . . . . . . . . . .

Primary News Sources among Different SES Groups. .

ix

.100

.102

.103

O 106

.107

.109

.112

.114

.116

.117

.118

.120

.126

LIST OF FIGURES

Figure
Page
1. The Theoretical Model of TV News Viewing In a
Broadcast-Channel-Only TV Viewing Situation. . . .
2. The Theoretical Model of TV News Viewing In a
Multi-Channel TV Viewing Situation . . . . . . . .
3. The Theoretical Model of 24-Hour News Viewing. . .
4. The Theoretical Model of News Learning . . . . . .
5. TV News Viewing in a Broadcast-Channel-Only
Situation 0 O O O O O O O O O O O O O O I O O O O O
6. Broadcast News Viewing in a Multi-Channel
Situation 0 O O O O O O O O O O O O O O O O O I O O
7. 24-Hour News Viewing in a Multi-Channel Situation.
8. News Learning in a Broadcast—Channel-Only
Situation 0 O I O O O O O O O O O O O O O O O O O O
9. News Learning in a Multi-Channel Situation . . . .

.111

.115

CHAPTER I.

INTRODUCTION

Research Background

For the last decade, one factor which could transform
the whole viewership and social effects of TV, has evolved
continuously. That factor is the development of
multi-channel TV.1 The phenomenal advancement in video
compression and fiber optic communication technology is
expected to bring as large as 500 channel TV in the future,
even though the expectations on the timing vary due to the
rapidly changing industrial and regulative environment
(Brown, March 8, 1993). The enormously expanded channel
capacity has brought a variety of new TV channels which are
geared to the narrowcast strategy such as news, sports,

music, and entertainment, etc.2 In this situation, the

 

1 In this dissertation research, the term 'multi-
channel TV' will be used to denote the new generation of TV
which provides many channels besides traditional over-the-
air broadcast channels. This term is preferred to ’cable
TV’ since cable is just one means of delivering multi-
channels. Also, from the viewer's point of view, there is
no physical distinction between over-the-air channels and
cable channels. In this vein, too, multi-channel TV is a
better term than cable TV which usually refers only to the
new channels available through cable subscription.

2For example, among the multi-channel networks planning
to launch within 1994 are "Recovery Net/The Wellness

1

 

 

CDDCEx

expos

netwo

 

this
4! m...
1' Stan:
sitx
Bab
(Ere
' a1“
Yc>

lbr

I?

 

 

-.¢Q’£!I7 /

 

 

2
concept of TV viewing characterized by the universal
exposure to a set of limited programs provided by a few TV
networks is bound to change in various aspects.

Among the various changes in TV viewing, the focus of
this dissertation is to investigate how TV news viewing is
affected by the development of multi-channel TV. Past
studies which explored TV news viewing in a multi-channel
situation (Baldwin, Barrett, & Bates, 1992a and 1992b:
Baldwin et. al., 1988: Becker et. al., 1983: Gelman, 1983:
Grotta & Newsom, 1983: Heeter & Baldwin, 1988: Henke et.
al., 1984: Jeffres, 1978a: Reagan, 1984: Webster, 1984:
Youn, 1993b) mostly focused on the changes in the amount of
broadcast TV news viewing due to multi-channel development.
The findings from these studies are far from being
consistent. Some researchers (Henke et. al., 1984: Jeffres,
1978a) found that in a multi-channel situation, the amount
of broadcast network (ABC, CBS, NBC) news viewing decreased.
On the contrary, other researchers (Baldwin et. al., 1988:
Becker et. al., 1983: Heeter & Baldwin, 1988: Reagan, 1984:
Youn, 1993b) found that TV viewers in a multi-channel
situation watched as much as or more network news than TV
viewers in a broadcast-channel-only situation. Underlying

these inconsistent findings are the differences in research

 

Channel" for recovering drug addicts, "The Golden American
Network" for older Americans and "The Golf Channel" for
golfers (Stern, June 7, 1993).

 

 

setti

sampl

mini}
long-
audi«
conc
deve
hypo
mult
audi
divE
of t
the
Balc
rat:
broe
con.
so 3
tha
(:10

CUR]

(:19
Qui
Hea
att
mu}

3
settings and designs (e.g., data gathering time points,
samples, and measurements).

In this vein, Baldwin's et. al. study (1992a), which
minimizes the problem of external validity by analyzing
long-term changes of TV news viewing based on national TV
audience rating data, provides the most comprehensive
conclusions about the changes in TV news viewing due to the
development of multi-channel TV. Baldwin et. a1.
hypothesized that in a multi—channel situation (a) the
multichannel offerings of non-news programs could divert
audiences from network and local broadcast news (the
diversion hypothesis) and (b) the head-to-head competition
of the 24-hour news networks in national news would erode
the broadcast news audience (the competition hypothesis).
Baldwin et. al. found that while broadcast network news
ratings have not changed much over the years among
broadcast-channel—only TV viewers, they have declined
continuously among multi-channel TV viewers and dramatically
so among premium channel subscribers. This study also found
that the decrease of broadcast network news share has a
close inverse relationship with the increase of the

cumulative 24-hour news viewing share.3 Based on this

 

’To interpret this as evidence of the competition
hypothesis is still not without question. As their study
clearly shows, average ratings for 24-hour news has remained
quite constant by year (about 1.5 including both CNN and CNN
Headline News). Therefore, it is more reasonable to
attribute network news viewing erosion to the growth of
multi-channel TV penetration (diversion), rather than to the

4
study, we can draw the following conclusions regarding the
influence of multi—channel TV development on the amount of
TV news viewing:

(a) Mainly due to diversion effect and probably due to
competition effect, broadcast news viewing has been

decreasing over the years.‘

(b) 24—hour news viewing has been increasing mainly due
to the steady growth of multi-channel TV penetration.

(c) Therefore, TV news viewing is changing in the

direction of watching more diverse news programs,

including both broadcast news and 24-hour news

programs.5

These trends provide a useful picture of the overall
changes. However, these trends are just a broad overview of
the complicated changes in TV news viewing due to the
development of multi-channel TV. Under these general

trends, the directions and degrees of changes can vary among

TV viewers: some TV viewers may increase the amount of TV

 

viewership increase of 24-hour news (competition).
Furthermore, as Baldwin et. a1. themselves mention (1992(a),
p.654), without knowing how much duplication is in the CNN
audience or what proportion of people actually substitute
CNN viewing for broadcast network news, it is difficult to
estimate the true competition between network news and 24-
hour news.

‘Since local news is not in a direct competition with
24-hour news, it is less likely to be affected by
competition effect.

9It is unclear from the past studies, however, whether
the overall amount of TV news viewing (including both
broadcast news and 24-hour news) has been increasing or
decreasing. It is likely that the total amount of TV news
viewing has been somewhat constant assuming that the
constant growth in 24-hour news viewing has made up the
decrease in broadcast news viewing.

5
news viewing, while some others decrease it. From the past
studies, however, it is difficult to find any elaborated
discussions about these individual variations in changes.
For example, will highly educated and less educated people
show different directions and degrees of changes? What
would be the case for those who differ in news learning
motivations?

Also uninvestigated from the past studies are the
changes in news learning from TV news viewing due to the
development of multi-channel TV. News learning from TV news
viewing in a broadcast-channel-only situation has been
extensively investigated by past studies (discussed in
Chapter II). However, we do not know what cognitive effect
TV news viewing has in a multi-channel situation. For
example, will the development of multi-channel TV make TV
play a more important role in informing people of important

social issues? How will it change the roles of other media?

Res a e
This dissertation will investigate how the development
of multi-channel TV affects TV news viewing and news
learning from it. First, with regard to the changes in TV
news viewing, going beyond describing the overall changes,
this study will examine how the TV news viewing of people is
affected differently due to the development of multi-channel

TV. For this we need to identify the factors which bring

6

(or sometimes suppres) the differences in TV news viewing.
Then, we need to examine how the roles of these factors are
different in broadcast-channel-only and multi—channel
situations, resulting in the variations in the directions
and degrees of changes among different social groups. With
regard to the changes in news learning from TV news viewing,
too, we need to identify the factors which are related to
the cognitive effect of TV news and to examine how the roles
of these factors are affected by the development of multi-
channel TV.

Since multi-channel TV has been introduced all over the
U.S. and multi-channel TV subscription cannot be
manipulated, the use of repetitive group analysis (the
comparison of the same group before and after multi-channel
TV subscription) is practically impossible. Therefore, the
major analysis of this research is limited to finding cross-
sectional differences between broadcast-channel-only TV

viewers and multi-channel TV viewers.

e rc es ' n nd Or niza 0 0'55 0
The major research questions investigated by this

dissertation are:

(a) How does the development of multi-channel TV change
TV news viewing? (In terms of cross-sectional
comparison, what are the major differences between a
broadcast-channel-only situation and a multi-channel
situation with regard to TV news viewing?)

7
(b) How does the development of multi-channel TV change
news learning from TV news viewing? (In terms of cross-
sectional comparison, what are the major differences
between a broadcast-channel—only situation and a multi-
channel situation with regard to news learning from TV
news viewing?)

In Chapter II, theoretical factors which are relevant
to TV news viewing and news learning from it will be
explicated and their relationships with these two variables
in both broadcast-channel-only and multi-channel situations
will be proposed as empirically testable hypotheses. In
Chapter III, methodological issues, including the research
design and the measurement of major variables, will be
elaborated. In Chapter IV, through data analysis, the
actual findings and hypothesis test results will be
presented. Finally, in Chapter V, the implication of the
research findings and hypothesis test results will be

discussed, and the directions for improvements through

future studies will be suggested.

 

 

 

 

 

contr

.
m." «Iv-mu!" —' _ -

W
rm?

three

V Jaw—vu-

USES

The q

 

Struc
(3281
it.

tale:
and n

SOme

 

gap r
alrea

 

CHAPTER II

LITERATURE REVIEW

TV Neiniejirlq

In the mass media field, major theoretical
contributions regarding TV news viewing have been made from
three different lines of studies: knowledge gap studies,
uses and gratification studies, and program choice studies.
The central focus of knowledge gap studies is on how socio-
structural variables, particularly socio-economic status
(SES), affect the amount of media exposure and learning from
it. Uses and gratification studies explicate the
relationship between audience needs or motivational factors
and media uses. Meanwhile, prggram_ghgig§_studies highlight
some other important factors such as programming schedule or

programming awareness in an actual program choice.

Knowledge Gap Studies
In 1970, Tichenor et. a1 (1970) proposed the knowledge

gap hypothesis: that is mass media tend to increase the
already existing knowledge level differences between the

more informed (high SES) and the less informed (low SES).

9

As the infusion of mass media information into a social

system increases, segments of the population with

higher socioeconomic status tend to acquire this
information at a faster rate than the lower status
segments, so that the gap in knowledge between these
segments tends to increase rather than decrease

(Tichenor, Donohue and Olien, 1970: 159—160).

The issue of knowledge gap has brought broad academic
attention: according to Gaziano (1983), 58 studies were
published by 1983. Underlying this academic interest is the
concern that even the ubiquitous and easily accessible mass
media news sources tend to aggravate the existing social
inequalities.

In knowledge gap studies, "knowledge" denotes a very
specific aspect of mass media's cognitive effects -- the
recall or understanding of public affairs or science news,
which has more or less general appeal, delivered by mass
media (Tichenor et. al., 1970).1 It is not necessarily
applied to more audience-specific topics, such as "stock
market quotations, sports, and lawn and garden care
(Tichenor et.al., 1970, p.160)." Also, "knowledge” concerns
the news information originated by mass media such as
newspaper or TV, even though it does not exclude

interpersonal communication (personal contact) as an

important mediator or messenger of this information. In

 

1To discuss the more general term "knowledge" is beyond
the scope of this study. Also, it is not strongly relevant
to the knowledge gap concept which denotes a very specific
aspect of mass media's cognitive effects. However, for an
extensive discussion on the concept of knowledge, see
Machlup, F. (1980 & 1982).

10
actual measurement level, "knowledge" means the degree to
which this news information is recalled or learned.

The knowledge gap, therefore, means the difference in
the degree to which the news information of general appeal
originated by mass media is recalled or understood.
Differences can be found individual by individual. However,
the general focus of interest, particularly in terms of the
original conceptualization of the knowledge gap hypothesis
by Tichenor, et. al., is on the structural gap among
different social groups, particularly among groups with
different socio-economic status (SES),’*which is considered
as the fundamental organizational structure of modern
society (Eichar, 1989: Leptrato & Lewis, 1974: Rossides,
1976: Vanneman & Cannon, 1987). Among various measures of
SES, knowledge gap studies have adopted education level as
the primary measure.

ack ound o ies of t now e e e .
In discussing the knowledge gap hypothesis, the term

"hypothesis" requires special attention. It means that

 

2Several studies have criticized the basic orientation
of knowledge gap studies focusing on SES. Genova and
Greenberg (1978) proposed that knowledge gap comes not from
SES but from interest differences. In the same vein, Ettma
and Kline (1977) proposed reformulation of the knowledge gap
hypothesis based on differences in motivation. Currently,
Reagan (1993), raised the same argument by saying that
knowledge gap is no more than "Knowledge difference" due to
the differences in interest. In this study, the role of
these interest or motivational factors will be discussed
more in detail with relation to the uses and gratifications
studies.

11
there are theories on which this hypothesis is based. The
theories suggested by Tichenor et. al. (1970, pp. 161-162)

are as follows:3

(a) communication skill. People with more formal
education are expected to have the higher reading and
comprehension abilities necessary to acquire public
affairs or science knowledge.

(b) stored information or existing knowledge. People
who are already better informed from prior exposure to
the topic through mass media or from formal education
itself are more likely to be aware of a topic when it
appears in the mass media and are better prepared to
understand it.

(c) relevant social contact. Education generally
indicates a broader sphere of everyday activity, a
greater number of reference groups, and more
interpersonal contacts, which increase the likelihood
of discussing public affairs topics with others.

(d) Selective exposure, acceptance, and retention of
information. Voluntary exposure is often more closely
related to education than to any other set of
variables. What appears to be selective exposure
according to attitudes might often more appropriately
be called "de facto" selectivity resulting from
educational differences

These original theoretical explanations suggest that
there are two basic conditions of the knowledge gap, the

differences in the amount of news information exposure and

the differences in the amount of learning from actual news

 

3Besides the four theories presented here, Tichenor et.
al (1970) also included the nature of mass media system.
They saw that mass media are geared to the interests and
tastes of the higher-status segment and, therefore, tend to
increase the knowledge gap. This explanation was excluded
here, since this system-level factor is hard to elaborate
empirically and is less applicable to the focal medium of
this dissertation research, TV, which is targeted to the
largest common denominator of audience.

12
information exposure. The main argument of the knowledge
gap hypothesis is that SES tends to satisfy these two
conditions of gap creation. First, the amount of news
information exposure is likely to be different among
different SES groups due to the differences in media content
selectivity and social contact. As noted by Samuelson et.
al. (1963), since increased formal education indicates an
expanded and more differentiated life space, including a
greater number of reference groups and more interest/
selectivity in science and other public issues, high SES
people are likely to be exposed to more news information.
Second, the amount of learning from media exposure is likely
to be different among different SES groups due to different
communication skill and stored knowledge. Since education
increases the communication skill and stored knowledge which
facilitate the understanding and retention of news
information, a high SES group tends to learn more from the
same amount of news exposure.

S and ount TV s v'ew'n . Contrary to the
original knowledge gap hypothesis, past studies which
focused on news learning from TV news have revealed with
remarkable consistency that TV news viewing tends to reduce
knowledge gap among different SES groups (Atkin et. al.,
1976: Becker & Whitney, 1980: Becker et. a1, 1978: Galloway,
1977: Israel & Robinson, 1972: Katz et. a1, 1977: McClure &

Patterson, 1976: Miller & MacKuen, 1979: Nordenstreng, 1972:

13

Neuman, 1976: Robinson, 1972 and 1974: Robinson 8 Levy,
1986: Roper, 1985: Shingi 8 Mody, 1976). Particularly, two
aspects of the findings from these studies need special
attention: (a) the universal exposure to TV news and (b) the
knowledge leveling effect of TV news viewing. First,
television news viewing is a universal phenomenon and the
amount of TV news viewing is not correlated with education.
Both the more educated and the less educated people in
society are equally likely to turn on TV news (Israel 8
Robinson, 1972: Neuman, 1976: Roper, 1985: Tunstall, 1983).
Secondly, heavy dependence on TV news raises the knowledge
level of the less-educated segments of society closer to the
overall average, but at the same time depresses the
knowledge level of the highly educated closer to the
average.‘ Newman (1976, p.122) noted these two aspects of
TV news viewing (universal exposure and knowledge leveling
effect) as follows:

In comparing the college professor and his

construction-worker counter—part, it may be not only
that they are equally likely to turn on the news but

 

‘For example, according to the study by the Center for
Political Studies of the University of Michigan in 1980
(cited from Robinson 8 Levy, 1986), for the political
information questions, regular news viewers with less than a
high school education scored a higher average point than
non-viewers, but of those with a college education, heavy
news viewers scored a lower average point. A news awareness
survey by Survey Research Center of the University of
Maryland in 1983 (cited also from Robinson 8 Levy, 1986)
found similar results. For the questions about news
figures, TV news viewers among the less educated scored
above average, but among college graduates, TV news viewers
again scored below average.

14
that they remember the same amount of what they see ...
not only on the lead story of the day and on human
interest stories of general appeal but on the rank and

file news items ranging from foreign affairs to
national politics and issues of ecology.

Interest naxinizatign theory. Therefore, contrary to
the basic notion of the knowledge gap hypothesis, past

studies show that SES is not related to the amount of TV
news viewing (the relationship between SES and news learning
will be discussed in the TV News Viewing and News Learning
section). Why is this the case? Also, would it be the same
in a multi-channel situation?

Jeffres (1978a) suggests answers for these questions by
proposing "interest maximization theory." In the context of
the knowledge gap hypothesis, Jeffres asSumes that high SES
and low SES differ in news interest/selectivity. In a
broadcast—channel-only situation, however, the differences
in news interest/ selectivity between high SES and low SES
cannot be linked to the different amount of TV news viewing
due to the structurally confining factors such as news
program availability and programming schedule. Jeffres
proposes that in a multi-channel situation, where people can
maximize their program type interests free from these
structurally confining factors, high SE8 TV viewers would
increase the amount of TV news viewing while low SES TV
viewers would decrease it. The result, he predicted, would
be a widening gap in the amount of TV news viewing between

high SES and low SES groups.

15

Jeffres put this idea to a test by comparing the
relationships between SES and the amount of TV news viewing
before and after multi—channel TV introduction. The result,
however, did not support his idea. No significant
differences were found in the amount of TV news viewing
between different SES groups after multi-channel TV
introduction. Therefore, as far as a formal empirical test
is concerned, we are left with the conclusion that the
development of multi-channel TV does not affect the amount
of TV news viewing among different SES groups.

Still, several problems in this study make it difficult
to accept this conclusion. Methodological problems such as
confounding of the seasonal variation of ratings and control
of other extraneous variables weaken the validity of this
study (for more detailed critique of this study, see Youn,
1993a). The most serious problem in Jeffres' study,
however, is that it was carried out when the development of
multi-channel TV was in its infant stage. More
specifically, the multi-channel TV in Jeffres' study means
eight channel TV including fairly overlapping distant
channels. It is questionable whether this small scale
channel increase due to the introduction of the infant-stage
multi-channel TV really satisfies the basic conditions of
interest maximization.

The fully developed multi-channel TV, nowadays, is

evaluated as approaching the basic condition of interest

l6
maximization more closely. According to Youn’s study
(1993a), in a multi-channel situation, a large number of
programs in all major program types are available during
prime time, compared to a broadcast-channel-only situation
(see Tables 1 and 2). During traditional TV news hours (6-7
PM and 11-11:30 PM), multi-channel TV provides plenty of
non-news program options, and 24-hour news channels provide
abundant news program options during non-broadcast-news
hours.

In this situation, with plenty of news and non-news
program choice options available around—the-clock, the role
of SES in determining the amount of TV news viewing as is
predicted by the interest maximization theory is likely to
be realized. Put differently, high SES viewers will turn to
news programs more often while low SES groups divert to non-
news programs. From this discussion, the following
hypothesis regarding the relationship between SES and TV
news viewing is proposed:

H1: SES will be positively related to the amount of TV

news viewing in a multi-channel situation, while it

will not be related to the amount of TV news viewing
significantly in a broadcast-channel-only situation.
8 ' o e

Through uses and gratification studies, the
relationship between audience needs or motivations,
including news/information needs, and actual media uses has
been extensively investigated. As a result, past uses and

gratifications studies have identified quite diverse

17

.330 H
3.83— 938 owns: on uncaognAHHBHHoHéHHH. 2 $223

 

 

 

 

 

 

 

 

 

 

 

 

 

Hrs 23 28- SH? #8.. :8- 38- “a? :8- :8- H28 H38
3825 Sea 28 :8 Has 28 28 38 can H38 H88 H58
803 . . . . . . H H H H
mean u H H H . . . . . .
mHHHHgHHHoSHDHHmouEfigHH . . . . H H H H H H
H.232 €358” . . . . . . . . . .
32530 macaw . . . . . . H H H H
83% H H H . ~ N H H H H
animals . . . . . . . . . .
Haunts :55 . . . . . . . . . .
23.33 . . . . . . . . . .
9:32.. . . N H . . . . . .
£3956 . . . H H N H H H H
09o». H . H . . . . . . .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

18

25o H
Romans 938 8325 on EHEAHHBRH .3 3286

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.28 $8 8- Ho- :8- :3. 8.. Ho- 8- SH? H38 H38 .65
3.823 £3 Ho 8 :3 38 Ho 8 Ho H28 HSHH :30
82.8 H H H H H H H H H H H.H
H8” 9 H H H H H H H H H H.H
HacomnHoHHuHhsmoaflHoasH H H H H H H H H H H H.H
H.832 23883.. H . H H . H H H H H H...
casino 963 H H . . H H H H H H H.H
88% H H H H H H H H H H H.H
HHHEmoHHH . . . . . . . . H H .H
32H?" 350 H H H H H H H H H H H.H
2338 . H H . . . . . . H .H
858% H H H H . . . . . . H.H
awakens H H H H H H H H H H H .H
952. H H H H H H H H H H H.H

 

 

 

 

 

 

 

 

 

 

 

 

 

19
(somewhat overlapping) media use needs or motivations
(Atkin, 1990: Bantz, 1982: Becker, 1979: Palmgreen 8Rayburn,
1979: Palmgreen, Wenner 8 Rosengren, 1985: Rubin, 1981 and
1983: Rubin, Perse, 8 Powell, 1985). From these studies a
group of motivation factors, more or less overlapping over
various studies, have been identified as follows:

(a) Becker (1979): surveillance, vote guidance,

excitement, reinforcement, communication, relaxation,

alienation, bias of media, and partisanship.

(b) Palmgreen 8 Rayburn (1979): relaxation, learning

about things, communicatory utility, forgetting,

passing time, companionship, and entertainment.

(c) Rubin (1983): relaxation, companionship, habit,

passing time, entertainment, social interaction,

information, arousal, and escaping.

(d) Atkin (1990): enjoyment, killing time, relaxation,

general information, companionship, specific guidance

for decision making and behavior, reinforcement of
attitudes, escaping, interpersonal communication
facilitation, and social acceptance.

However, one fundamental distinction of media uses
which has very important implications with regard to news
content exposure including TV news viewing has appeared with
remarkable regularity. In uses and gratifications studies'
parlance, this distinction can be summarized as one between
content seeking v process seeking.

C te t d o e 5 see ' . In uses and
gratification studies, content gratification and process
gratification (Cutler and Danowski, 1980, pp. 269-270),

content seeking and media seeking (Jeffres, 1978b),

instrumental media use and ritualized media use (Rubin,

20

1981, 1983, 8 1984), and Katz's et. al. (1974) suggestion
that audience members can receive gratification from both
media content and the exposure situation, all denote the
same distinction of two sets of media use motivations.
These distinctions are between (a) use of messages for
intrinsic values that bear a direct link to particular
substantive characteristics of the messages and (b) the use
of messages for extrinsic values that do not bear a direct
link to particular substantive characteristics of the
messages. In this study, the former set of media use
motivations will be named as content seeking (motivations)
and the later, process seeking (motivations).

Characterizing content seeking is message uses to gain
knowledge, increase or reduce uncertainty in personal and
social situations, or to support existing predispositions.
0n the contrary, process seeking, the gratification for
which comes mainly from being involved in the process of
communication behavior rather than the message content per
se, includes a myriad of escape uses and stimulation uses
that often involve engagement in entertainment and uses
combatting social isolation through connections with
mediated culture and its actors.

Past studies (Gantz, 1978: Jeffres, 1978b: Youn, 1993b)
found that TV news viewing is related to process seeking as
well as content seeking since people tend to watch TV news

in the course of TV viewing, as well as for the specific

21
purpose of news information seeking. However, since the
primary objective of TV news programs compared to other non—
news/entertainment programs is to deliver news/information,
and since content seeking is the need to obtain news/
information content, it is likely that TV news program
viewing is more strongly related to content seeking,
compared to other TV programs.

This discussion tends to lead us to the conclusion that
those who pursue strong content seeking would watch more TV
news. However, strong content seeking would not necessarily
lead to TV news viewing, if the evaluation of TV news in
gratifying content seeking is low. In this case, they will
turn to another medium such as newspaper for which they have
better evaluation. Put differently, content seeking denotes
"news/information seeking" or "news learning motivations" in
general rather than "TV news seeking." Therefore, in order
to get at a more elaborate understanding of TV news viewing
intention, we need to consider a viewer's evaluation of TV
in gratifying content seeking as well as the strength of
content seeking per se.

t v o . By considering this
evaluation factor, expectancy value theory (Atkinson, 1957:
Feather, 1959: Fishbein, 1963: Fishbein 8 Ajzen, 1975:
Rotter, 1954: Tolman, 1932) provides a more elaborate
theoretical definition and measurement of TV news viewing

intention. Although the various theories under this label

22
differ somewhat in their emphases, all view either behavior,
behavioral intentions, or attitudes (or all three) as a
function of (a) expectancy (or belief), that is, the
probability that an attitude object possesses a particular
attribute or that a behavior will have a particular
consequence and (b) evaluation, that is, the degree of
affect, positive or negative, toward an attribute or

behavioral outcome.5

 

5The various discussions about expectancy value theory
include:
(a) Tolman (1932) - According to Tolman, people learn
"expectations," i.e., beliefs that a given response will be
followed by some event. Since these "events" can be either
positive or negative "reinforcers" (i.e.,can have positive
or negative valence), his argument essentially is that
people will learn to perform (or increase their probability
of performing) behavior that they "expect" to lead to
positively valenced events.
(b) Edwards' (1954) Subjective Expected Utility (SEU) model
- According to this theory, when a person makes a
behavioral choice, he will select that alternative which has
the highest subjective expected utility. The subjective
expected utility of a given alternative is defined in the
following equation.

a
330:2 SPiUi,

14
where SEU is the subjective expected utility associated with
a given alternative: SP5 is the subjective probability that
the choice of this alternative will lead to some outcome 1:
[h is the subjective value or utility of outcome i: and n is
the number of relevant outcomes.
(c) Rosenberg's (1956) Instrumentality Value Model -
Rosenberg defined attitude as a "relatively stable affective
response to an object" and argued that this attitude is
"accompanied by a cognitive structure made up of beliefs
about the potentialities of that object for attaining or
blocking the realization of valued states" (p. 367).
According to him, the more a given "object" was instrumental
in obtaining positively valued goals and in blocking
negatively valued goals, the more favorable the person's
attitude toward the object. This hypothesis is expressed in

23

Fishbein (1963) has made this relationship an explicit
part of his theory of attitude which can be described as
follows: (a) An individual holds many beliefs about a given
object: i.e., the object may be seen as related to various
attributes, such as other objects, characteristics, goals,
etc. (b) Associated with each of the attributes is an
implicit evaluative response, i.e., an attitude. (c) Through
conditioning, the evaluative responses are associated with
the attitude object. (d) The conditioned evaluative
responses summate, and thus (e) on future occasions the
attitude object will elicit this summated evaluative
response, i.e., the overall attitude.

According to the theory, a person's attitude toward any
object is a function of his beliefs about the object and the
implicit evaluative responses associated with those beliefs.
The central equation of the theory can be expressed as

follows:

I)
110:2;pr

 

the following equation:

I:
110:2; IiVi,

where I1 is instrumentality, i.e., the probability that 0
would lead to or block the attainment of a goal or value 1:
\h is value importance, i.e., the degree of satisfaction or
dissatisfaction the

person would experience if he obtained value 1: and n is the
of goals or value states.

 

 

 

 

wheré
belie
relaw
and :
varie
nonl1
Howe}

view¢

 

241)

numb
or t
Exal'

lxin

3!)
ha

rrmrrmH—Hmrdrf

24
where A0 is the attitude toward some object, o: tn is the
belief about 0, i.e., the subjective probability that o is
related to attribute i: en is the valuation of attribute i:
and n is the number of beliefs. Because both b and e are
variables in the expectancy value equation, attitude is a
nonlinear function of b, or of e, if each is taken singly.
However, the model is linear when the b x e product is
viewed as a variable (Fishbein and Ajzen, 1975, pp. 237-
241).

A major assumption of the theory is that only a limited
number of beliefs are relevant or salient to a given object
or behavior at a given time. The totality of a person's
beliefs represents the information that a person has about
him(her)self and his/her social and physical environment.‘

Appligntions 9f ernggtnncy-valne tneory. Two different
applications of expectancy-value theory to mass media uses

have been provided up to now. One is Galloway and Meek's

 

‘Fishbein and Ajzen provide a person's attitude toward
the supersonic transport (SST) as an example. If this
person holds the following beliefs such as (a) SST is an
airplane, (b) SST is noisy, (c) SST is not economical, and
(d) SST is a pollutant, his/her attitude toward the SST is a
function of the strength with which (s)he holds these
beliefs and of his/her evaluations of each attribute. The
subjective probabilities and evaluations that might have
been obtained can be presented as follows.

Belief b e be
airplane .90 +2 1.80
Noisy .80 -2 —1.60
Not economical .60 -1 - .60

Pollutant .50 -3 -1.50

 

 

 

 

(1981

Palmq

Meek

the f

where

1th 9]

 

grati

equa
usin
medi
the
“hi

V11

es.
s:
ex
19
th
ex}

991

25
(1981) expectancy model of media exposure, and the other is
Palmgreen and Rayburn's (1982, 1983) summation model.
(a) Expectancy model of media exposure: Galloway and
Meek applied expectancy-value theory to explain the amount
of a specific medium exposure. Their idea is expressed by
the following equation:

Exposure = L‘E1 x V,,

where 1?.1 is the strength of expectancy associated with the
in: gratification, and V, is the value of the in,
gratification.

The problem in this equation is that media exposure is
equated with the attitude toward a medium or intention of
using a medium (since the product of expeCtancy value of a
medium is the attitude toward a medium or intention of using
the medium). As a result, this model ignores other factors
which would affect the actual media exposure besides
viewer’s attitude or intention.

(b) Summation model of media use intention: By
equating expectancy value of a medium with gratification
sought (i.e., media use motivation) instead of actual media
exposure, Palmgreen and Rayburn (1982, 1983, 1985a, and
1985b) provide an advanced application of value expectancy
theory. The relation between a gratification sought and
expectancy value of a medium is expressed as the following
equation:

GS; = bieil

 

 

 

wherfa
PTOQLh

probe
behax

affej

song)

 

Wher

vari

Pal
an
3w

ar

26
where GS,== the ith gratification sought from X (some medium,
program, content type, etc.): tn=the belief (subjective
probability) that x possesses some attribute or that a
behavior related to X will have a particular outcome: e,=the
affective evaluation of the particular attribute or outcome.
Based on this equation of a single gratification

sought, they proposed the following summation model:
268, = Elble1 ,

where 2G3, indicates a generalized orientation to seek
various gratifications from a particular source.

Congenr grarificnrign expecrancy of Ty (ggg-Ty).
Palmgreen and Rayburn's model (1982, 1983, 1985a and 1985b:
an application of this model is also found in Babrow 8
Swanson, 1988) provides an important implication in defining
and measuring the strength of TV news viewing intention
(defined as the sum of various TV news viewing motivations).
According to their model, the intention of using TV to
satisfy one's news/information needs can be expressed as the
sum of the product of each news/information motivation
(value) and evaluation (expectancy) of TV in gratifying this
motivation.

For a valid application of expectancy value theory to
TV news viewing, however, instead of expectancy, value needs
to be considered first, even though this does not make any

difference in actual computation. Expectancy value theory

 

was 6
obje:
certs
not <
put (
towa:

choi1

 

In 01

 

set (
how ‘

eval

int.
rel
see
vi.

Se

> 1'“

Se

 

27
was developed to explain one’s attitude toward a person,
object, etc. Therefore, expectancy comes first. That is,
certain characteristics of a person or a thing, which are
not changeable, are noticed first. Then, based on the value
put on each characteristic, the comprehensive attitude
toward that person or thing is decided. In actual program
choice process, however, we can expect the opposite process.
In other words, it is assumed that a viewer has a certain
set of values (motivations) to be satisfied. After that,
how well various TV programs can satisfy these values is
evaluated.

Then, the first issue with regard to TV news viewing
intention is to identify a certain set of values which are
related to TV news viewing. From the discussion on content
seeking and process seeking, it is likely that TV news
viewing is mainly related to content seeking while process
seeking is related to non-news program viewing. Then, based
on Palmgreen and Rayburn's model, TV news viewing intention
will become the sum of the product between the strength of
each content seeking motivation and the expectancy of TV in

gratifying this motivation.
CGE-TV = 2 C; X Ell

where CGE-TV (content gratification expectancy of TV)
indicates a generalized orientation to gratify content

seeking from TV: cg is the strength of each content seeking

 

 

moti'

moti‘

can‘

wher
indi
seek

moti

 

moti

con:

Vht
Va

Qe

SE

28
motivation: E5 is the expectancy of TV in gratifying this
motivation.
In the same vein, the process seeking expectancy of TV

can be expressed as the following equation.
PGE-TV = 5.: P1 x E“

where PGE-TV (process gratification expectancy of TV)
indicates a generalized orientation to gratify process
seeking from TV: P3.is the strength of each process seeking
motivation: Eh is the expectancy of TV in gratifying this
motivation.

The total gratification expectancy of TV can be

conceptualized as the sum of CGE—TV and PGE—TV.
TGE-TV = CGE-TV + PGE-TV = 2 G, x E,,

where TGE-TV indicates a generalized orientation to gratify
various media use motivations from TV: G, indicates a
general media use motivation item (both content and process
seeking): Eh is the expectancy of TV in gratifying this
motivation.

CGE-TV is evaluated as a good indicator of TV news
viewing intention in a broadcast-channel—only situation
where available program types are relatively limited
(especially during traditional TV news viewing hours when
few non-news programs are available). In other words, when

a viewer has only two choices (i.e., watching TV news or

29
turning off TV), the stronger content seeking and the better
the evaluation of TV in gratifying content seeking, the more
likely he/she will turn to TV news. From this the following
hypothesis is proposed:

H2: In a broadcast-channel-only situation, CGE—TV will

be positively related to the amount of TV news viewing.

ela 'v t nt G at'f c t'o ct n 0 TV C -
TV). In a multi-channel situation, where a number of
program types are available besides news programs, a viewer
will have three different choices (i.e., watching TV news,
watching other programs, and turning off TV). Therefore,
CGE-TV may not be a good indicator of TV news viewing
motivation. Put differently, in a multiéchannel TV
situation, even though a viewer persues strong content
seeking and the evaluation of TV is good (high degree of
CGE—TV), if (s)he persues stronger process seeking and the
evaluation of TV in gratifying process seeking is better,
(s)he will turn to non—news programs.

More basically, in a multi-channel situation where a
variety of program types are available at the same time, a
viewer will turn to a program by which he/she can maximize
total gratification expectancy. In other words, the
intention to watch TV news will depend not on the absolute

size but on therelative_size of the expectancy value of TV

30
news compared to those of other programs.7 Therefore, to
estimate the amount of TV news viewing, we need to obtain
the expectancy values of all the other programs, which is
practically unfeasible.

A more practical and theory-oriented solution regarding
this problem, however, is suggested from the conceptual
distinction of content seeking and process seeking. As was
discussed before, among TV programs, TV news viewing will
mainly be related to content seeking, while process seeking
will lead to more non-news program viewing in which news/
information delivery is not the primary goal.

Therefore, the strength of TV news viewing intention,
that is the relative size of the value expectancy of TV news
programs compared to the value expectancy of other programs,
can be estimated by computing how large a potion of TGE-TV
of a viewer is taken by CGE-TV. If a large portion of TGE-
TV is taken by CGE—TV in a viewer's case, then it is
expected that he/she has relatively strong expectancy value
of TV news programs. On the contrary, if only a small
portion of TGE-TV is taken by CGE-TV in another viewer's
case, then it is likely that (s)he has relatively weak

expectancy value of TV news programs.

 

’It can be compared to the situation of choosing the
best friend among one's friends. Expectancy value equation
just provides the attitude toward each friend. To choose
the best friend, expectancy values for all the friends need
to be measured.

31
Thus, the relative strength of gratification expectancy
of TV news can be replaced by the relative content
gratification expectancy of TV (RCGE—TV), which is computed

by dividing CGE-TV by TGE-TV.

k
2 C13,

HUGE—TV: 1' 1

;
g 615.1

1

: CGE-TV : CGE-TV
TGE-TV CGE-TV+PGE-TV ’

 

 

where RCGE-TV indicates the relative strength of
gratification expectancy of TV news: Ch is the strength of
the in, content seeking motivation: E, is the evaluation of
TV in gratifying content seeking item i:_k is the number of
content seeking items: G, is the strength of the in, media
use motivation: Eh is the expectancy of TV with regard to
the motivation item i: n is the total number of motivation
items.

In a multi-channel situation, where plenty of news and
non-news programs are available, RCGE-TV, as an indicator of
the relative strength of the total gratification expectancy
of TV news compared to other TV programs rather than CGE-TV
as an indicator of the absolute strength of content
gratification expectancy of TV, is estimated as a better
predictor of the amount of TV news viewing. Especially, the
amount of 24-hour news viewing is expected to be related to

RCGE-TV more strongly, since the viewing of 24-hour news

32
channels as specialized news channels is likely to depend
more heavily on the relative strength of the news/
information seeking motivations through TV. Therefore,

NM
HH..____

regarding the relationship between RCGE-TV and the amount of

WVHH

TV news viewing, the following hypothesis is proposed:

H3: In a multi-channel TV situation, the amount of TV
news (particularly, 24-hour news) viewing will be
positively related to RCGE-TV.

Program Choicerstudies

So far we have attempted to specify the relationship of
SES and news seeking motivations with TV news viewing in
both broadcast-channel-only and multi-channel situations
based on knowledge gap studies and uses and gratification
studies. Besides this, we can identify several other
important theoretical factors which are related to TV news
viewing from program choice studies and other multi—channel
TV audience studies (Baldwin et. al., 1992(b): Heeter, 1988:
Heeter 8 Greenberg, 1988: Webster, 1983: Webster 8 Agostino,
1982). These factors include viewer availability, news
programming (channel) awareness, channel repertoire and
value perception of news programming (channels).

Viewer availnnility. Since TV news is available only
a certain portion of time in a broadcast-channel-only
situation, its viewing will heavily depend on whether a
viewer can watch TV during TV news hours (Webster 8

Agostino, 1982). For multi-channel TV viewers, also, viewer

 

 

 

avai
View
this
case
when

rela

W

 

chaq
haV1
mul
hox

DI

33

availability will be a crucial factor of deciding TV news
viewing with regard to broadcast news viewing. However,
this availability will not matter at all in 24-hour news'
case (Heeter 8 Greenberg, 1988), since a viewer can watch it
whenever he/she wants to. From this, regarding the
relationship between viewer availability and the amount of
TV news viewing, the following hypothesis is proposed:

H4: Viewer availability will be positively related to

the amount of broadcast news viewing in both broadcast-

channel—only and multi—channel situation, while it is
not related with the amount of 24-hour news viewing.

Hens programming (channel) awareness. For broadcast-

channel—only TV viewers, news programming awareness will not
have strong influence on TV news viewing. However, for
multi-channel TV viewers, especially with regard to the 24-
hour multi-channel TV news, the importance of news
programming(channel) awareness in TV news viewing will
become significant. With multi-channel TV, the assumption
that each time viewers select a program they are aware of
and weigh all program alternatives to select a most
preferred option is untenable (Heeter 8 Greenberg, 1988).
There are simply too many options. Even on a general basis,
multi-channel TV subscribers are not very aware of the
different channels available to them, let alone the

different programs.‘ It is likely that viewers

 

'According to Arbitron Report (1983) cable subscribers
are not aware of all the services available to them or even

 

 

 

 

(par
PTO?
prog
news
the

awar'

PTOP

 

 

917055....Hn I

Q.

34
(particularly multi-channel TV viewers) who have high news
programming (channel) awareness are likely to watch news
programs more often, and, in turn, this will improve their
news programming (channel) awareness. From this, regarding
the relationship between news programming (channel)
awareness and TV news viewing, the following hypothesis is
proposed:

H5: News programming (channel) awareness will be

positively related to the amount of 24—hour news

Viewing.

Qnennei repertoire. Studies about the multi-channel TV
audience (Heeter, 1988: Nielson, 1983: Reagan, 1993:
Television Audience Assessment, 1983: Webster, 1983: Webster
8 Agostino, 1982) have found that multi-channel TV viewers
have a limited set of channels that they check and watch
regularly and intentionally. This limited set of channels
is called a channel repertoire (Heeter, 1988).

According to Webster (1983), who analyzed the viewing
shares with and without cable across 24 markets, networks
still attract the plurality of viewing share. Pay cable

channels attract a 14 to 20 percent share in pay cable

 

of what service they're watching at any given time (p.13).
In door-to-door interviews, viewers were able to correctly
identify an average of 9 of 35 available channels by channel
number or location on the channel selector and were aware of
one additional service of uncertain location. Twenty-three
percent of respondents were able to identify only 0—3
channels. Another 23 percent correctly named 14-27
channels, and the remainder identified between 4 and 13.

35

homes. Distant stations draw a 10 percent share in the
largest market and as much as 46 percent in smaller markets.
The remaining cable-only channels account for very little
viewing time (less than 10 percent). According to Heeter’s
study (1988), of 35 channels carried, only the three local
network affiliates were regularly watched by 50 percent or
more of the cable subscribers surveyed. HBO, WTBS, and a
local independent were watched by 40-50 percent. Nine of
the 22 other channels available only with cable (ESPN, MTV,
CNN, USA, etc.) were watched by one-tenth to one-third of
viewers. Thirteen cable channels were watched by less than
one-tenth of subscribers.

Channel repertoireh’ as a type of predisposition toward
TV channels, is likely to affect TV news viewing of multi-
channel TV viewers. More specifically, the inclusion of 24-
hour news channels in channel repertoire is likely to
increase the amount of 24-hour TV news viewing. From this,
regarding the relationship between channel repertoire and
the amount of TV news viewing, the following hypothesis is
proposed:

H6: The inclusion of 24-hour news channels in channel

repertoire will be positively related to the amount of
24-hour news viewing.

 

’In this study, channel repertoire will be defined as
channels turned to intentionally. A channel unintentionally
viewed in the course of channel flipping cannot be
considered as a part of channel repertoire no matter how
frequent and regular that kind of viewing happens, since
that type of viewing is purely matter of chances affected
heavily by channel placement.

36

Value perception of news channele. Multi-channel TV
viewers pay a certain amount of money to get extra channels
which are not available to broadcast—only-channel viewers.
In other words, multi-channel TV viewers seek extra values
in return for their subscription fee. Since multi-channel
TV provides a package of channels rather than sells each
channel, various value perceptions can be associated with
multi-channel subscription. For example, for some
subscribers, to receive extra entertainment will constitute
the main value, while for some others, to receive extra
news/information will be perceived as the main value. In
the former’s case, the subscribers will increase the amount
of entertainment program viewing. In the latter’s case, the
subscribers will turn to more news programs.

Therefore, the values associated with multi-channel TV
will dictate the types of programs mainly viewed from it.
Put differently, 24-hour news viewing will be closely
related to the value which a subscriber puts on news/
information channels. According to Baldwin et. al. (1992b),
from an economic point of view, value for a service should
translate into monetary terms. Thus, those who are willing
to pay an extra subscription fee for the news/information
channels of multi—channel TV will watch 24-hour news more.
From this, regarding the relationship between the value
perception of news channels and the amount of 24—hour news

viewing, the following hypothesis is prOposed:

37
H7: The value perception of news/information channels

will be positively related to the amount of 24-hour
news viewing.

TV News Viewing Models

 

From the discussion so far, we can propose theoretical
models regarding TV news viewing in broadcast—channel-only
and multi-channel situations respectively (in the latter’s
case, broadcast news and 24-hour news, respectively).

Amount of TV news viewing is necessarily contingent upon
several other factors, such as the amount of TV viewing
hours, the amount of other media uses, and other demographic
factors (age, gender, number of children), besides the major
theoretical factors discussed previously. Therefore, these
variables will be included as well in the models.

Besides these variables, in a multi-channel situation,
broadcast news and 24-hour news can affect each other in
both competitive and complementary fashions. Therefore,
factors specifically related to 24-hour news viewing (such
as 24-hour news channel awareness), including the amount of
24-hour news viewing will be included when we discuss the
broadcast news viewing model in a multi-channel situation.
In the same vein, the amount of broadcast news viewing will
be included in the 24-hour news viewing model.

Broadcaet-cnannel-oniy eiruetion. In a broadcast-
channel-only situation, TV news is available only during
limited time. Also, during TV news viewing hours only a few

program type options are available except news programs,

38
which makes TV news viewing a compulsive behavior unless a
viewer decides to turn off TV. Therefore, as was
hypothesized before, the most important factor with regard
to TV news viewing will be the structural constraining
factor of viewer availability. Put differently, the amount
of TV news viewing in a broadcast—channel-only situation
will mainly depend on whether a viewer can watch TV during
TV news hours.

Also, since TV news viewing is almost compulsory once a
viewer decides to watch TV, the amount of TV news viewing
will depend on TV news viewing intention measured by the
absolute strength of content gratification expectancy of TV,
rather than the relative strength of gratification
expectancy of TV news programs compared to those of other TV
programs. Therefore, CGE-TV will be positively related to
TV news viewing.

From this discussion, the theoretical model regarding
the amount of TV news viewing in a broadcast-channel-only
situation is proposed as Figure 1.

finiti-enennei siruation. In a multi—channel situation
compared to a broadcast-channel—only situation, a different
set of theoretical factors are likely to be related to TV
news viewing. First, with regard to broadcast news viewing,
it is important to consider two major differences of a
multi-channel situation compared to a broadcast-channel-only

situation: (a) other non—news program types are widely

 

News Program
Availability

 

 

 

 

TV News
Viewing TV News
Intention Viewing
(CGE-TV)

 

 

 

 

 

 

 

 

 

Other
Variables

 

 

 

Fi e . The Theoretical Model of TV News Viewing in a
Broadcast-Channel-Only TV Viewing Situation

 

ai
ws
ca
.11

'al
{ct
m;
u
=11:

 

 

teal

 

40
available during traditional TV news hours and (b) 24-hour
news is available as an alternative or complement to
broadcast news. Since the availability of broadcast news is
still limited to a certain portion of hours, viewer
availability is likely to be one of the most influential
factors. However, since broadcast news viewing is no longer
compulsory, as is the case in a broadcast-channel-only
situation due to the two points above, a weaker relationship
between news availability and the amount of broadcast news
viewing is likely to be observed.

Meanwhile, the decrease of the influence of viewer
availability factor is likely to introduce changes in the
influences of audience factors such as SES and TV news
viewing intention. As was hypothesized before, with regard
to TV news viewing intention, since broadcast news viewing
is likely to happen after comparing various TV program
options, the amount of broadcast news viewing will depend on
how strong the gratification expectancy of TV news is
compared to those of other TV programs, rather than the
absolute strength of content gratification expectancy of TV.
Therefore, RCGE-TV as a better indicator of TV news viewing
intention than TGE-TV is expected to be positively related
to the amount of broadcast news viewing in a multi—channel
situation. Also in a multi-channel situation, according to
interest maximization theory, high SES viewers are likely to

stay with broadcast news, while low SES viewers will divert

 

 

 

to r
relz
Fror
bro:

pm]

con:
Fir:
Whi<
bro
lon
vie‘
fac

wil

t0

Wit

Chi
Chi
rel
the

24.

 

41
to non-news programs. Therefore, SES is likely to be
related positively to the amount of broadcast news viewing.
From this discussion, the theoretical model regarding the
broadcast news viewing in a multi—channel situation is
proposed as Figure 2.

With regard to 24-hour news, the roles of the factors
considered in broadcast news viewing are expected to change.
First of all, as was hypothesized before, news availability
which is a dominant factor of broadcast news viewing in both
broadcast-channel-only and multi-channel situations is no
longer likely to be a significant factor, since 24-hour news
viewing is available around—the-clock. Meanwhile, audience
factors such as SES and TV news viewing intention (RCGE-TV)
will become more significant.

Besides, new theoretical factors which are not relevant
to broadcast news viewing are expected to become important
with regard to 24-hour news viewing. As was discussed
before, these factors include news channel awareness,
channel repertoire, and the value perception of news
channels. These factors are expected to be positively
related to the amount of 24-hour news viewing. From this,
the following model (Figure 3) is proposed with regard to

24-hour news viewing.

42

 

SE8

 

 

 

 

News Program

Availability "““-~HHHrrH*~mmm~
Broadcast

News
TV News ”’,,H»”r,HHHHr”"”" viewing
Viewing

Intention
(RCGE-TV)

 

 

 

 

 

 

 

 

 

 

 

Other
Variables

 

 

 

Fignre z. The Theoretical Model of Broadcast News Viewing in
a Multi-Channel Situation

 

 

 

 

 

RCGE-TV

 

 

 

 

News Channel

Awareness \
Channel 24-Hour News

Repertoire Viewing

Value /
Perception of

 

 

 

 

 

 

 

 

 

 

News Channels

 

 

 

Other
Variables

 

 

 

Fignre 3. The Theoretical Model of 24—hour News Viewing

44
TV News Viewing and News Learning

So far, the relationship between various theoretical
factors and TV news viewing in both broadcast-channel—only
and multi-channel situations has been discussed. In this
section, the discussion will focus on the changing nature of
news learning from TV news viewing due to the development of
multi—channel TV.

If TV news were the only source of news learning, then
a simple cross-sectional comparison of news learning from TV
news viewing (measured by the relationship between the
amount of TV news viewing and news knowledge level) among
multi-channel TV viewers and broadcast-channel-only TV
viewers would show us exactly how multi-channel TV
development would affect news learning from TV news viewing.
However, to make the comparison more complicated, there are
many other factors which affect news learning (e.g., SE8,
news/information seeking motivation, and exposure to other
news sources, etc.). Furthermore, those who subscribe to
multi-channel TV and those who<d§ not/Cannot be equated to
the experimental and control groups which differ only in
this multi-channel subscription factor. Actually, past
studies which compare multi-channel TV subscribers and non-
subscribers (Becker et. al, 1983: Collins et. al, 1983:
LaRose 8 Atkin, 1988: Metzger, 1983: Rothe et. al, 1983:
Sparkes, 1983: Sparkes 8 Kang, 1986) suggest that these two

groups might differ in various demographic and media use

45
aspects, even if the findings are somewhat inconsistent
across studies.

Therefore, in examining the relationship between TV
news viewing and news learning in broadcast-channel-only and
multi-channel situations, we need to consider these factors
of variation in news learning. In this vein, this study
will include SES, news learning motivation, other media
exposure, and other demographic variables besides TV news

viewing.

Socio-Eeonomie Statne (SES)

One of the most extensively discussed factors (mainly
by knowledge gap studies) of news learning is SES, which is
primarily measured by education level. As was discussed
before, Tichenor et. a1. (1970) hinted that communication
(cognitive) skill and prior knowledge are major theoretical
factors which link SES to news learning. More elaborated
investigation on these factors, however, is provided from
cognitive psychology literature related to human information
processing rather than knowledge gap studies.

S and i ve ski . The most important
characteristic of human information processing system is
that it is limited in capacity (Broadbent, 1958: Navon 8
Gopher, 1979: Norman 8 Bobrow, 1975: Posner, 1978).

Processing capacity, or alternatively, cognitive capacity,

is defined as "the limited pool of energy, resources, or

 

 

 

fuej

mai:

J oh:
D00
Kal.
Bob
197
imp-
sem
to
str
cor
Phi
ch<

 

WI:

Dr I:

 

 

46
fuel by which cognitive processes are mobilized and
maintained" (Johnston 8 Heinz, 1978, p.422).

Studies on human information processing (Bransford 8
Johnson, 1972: Britton et al., 1979: Britton 8 Price, 1981:
Dooling 8 Lachman, 1971: Gardner 8 Schumacher, 1977:
Kalsbeek 8 Sykes, 1967: Navon 8 Gopher, 1979: Norman 8
Bobrow, 1975: Sperling 8 Melchner, 1978 : Tyler et al.,
1979) have identified two classes of message variables which
impose demands on the cognitive processing capacity:
semantic and structural.10 Semantic variables are related
to the meaning or interpretation of the text. In contrast,
structural variables are largely independent of text
content. For example, the arrangement of text ideas into
phrases and sentences (i.e., syntax) and the kinds of words

chosen to express content (i.e., diction) are structural

 

‘mPerfetti (1979) provides a more elaborated
structural—semantic processing level distinction as follows.

Level Information to trample Characteristic
he Represented lenories
The old man blocked his path.
prelinqusitic Acoustic patterns 2 3 2 2 2 3(-) vocal quality
en + II + ms + mm + mm + mm e.q., voice
ntensity or
pitch)
phonological phonemic seq- /diy cwld men blakt his peO/ phoneme
mentsrhierar- sequences
chically includes (e.g.,
supraseqmentals consonant-
cf 1 vowel
sequences)
syntactic superficial con- 8(IP(The(old men)))(VP(blocked constituent
stituentsrhier (his path))) sequences
archically (e.g., verba-
includes 2 tin phrases)
propositional basic semantic CAUSI(Agent: X(PLACI local seman-
relations (Instrument: I, Object:o, tic relations
Location: Path of Y))) (e.g., actions,
states)
referential identification of ax [(x,men)-(x, old)] references
referential 3V [Y,male] (e.g.,names,
constituents places)
thematic thematic: role a) The old man (blocked an connections
of referential intruder's path) with text
constituents b) Siegfried (had his path blocked (e.g., theses)
by an old man)
functional intentions, a) A confrontation existed connections
motivations, b) Siegfried faced possible injury with text and
deleted and nontext
generalised knowledge

propositions (e.g., scripts)

47
aspects of text. Structural variables specify the forms
that express the propositional content.

Past studies (Bransford 8 Johnson, 1972: Britton 8
Price, 1981: Briton et. al, 1979: Dooling 8 Lachman, 1971:
Gardner 8 Schumacher, 1977: Kalsbeek 8 Sykes, 1967: Navon 8
Gopher, 1979: Norman 8 Bobrow, 1975: Sperling 8 Melchner,
1978: Tyler, et al., 1979) found that cognitive capacity
allocated to the structural aspects of a message does not
have an effect on learning, while cognitive capacity
allocated to the semantic aspects of a message does. More
specifically, when messages vary in structure (but not in
semantic content), the learners allocate more capacity to
the text with more complex structure. Since this capacity
is allocated to the structural aspects of the text, learning
of the semantic content is less likely (Britton et. al.,
1979: Britton 8 Price, 1981). On the other hand, when texts
vary in semantic content (but not in structure), the
learners allocate more capacity to the text with more
semantic content. In this case, since the capacity is
allocated to the meaning of the text, increased learning of
the semantic content would be expected and has been found
(Bransford 8 Johnson, 1972: Dooling 8 Lachman, 1971: Gardner
8 Schumacher, 1977: Tyler et al., 1979).

Therefore, the amount of learning from a message
depends on the level to which a person can process the

message (level of processing). As Greeno (1977, p.12)

48
mentions, difficulty in decoding at the level of words and
small phrases can cause a great deal of an individual's
processing capacity to be used there. When that occurs,
there is less capacity left for the essential task of
integrating the concepts of sentences into meaningful
representations, and the propositions of a story into a
meaningful structure.

By means of instruction and practice (repetition),
however, the capacity demand for structural aspects of the
text can be kept within acceptable limits. It has been
suggested that some component reading skills can approach
automatic levels of performance (LaBerge 8 Samuels, 1974:
Schneider 8 Shiffrin, 1977: Shiffrin 8 Schneider, 1977). As
Huey (1968) remarked, repetition progressively frees the
mind from attention to details and reduces the extent to
which consciousness must concern itself with the process.11

These discussions on cognitive capacity, levels of
processing, and the role of instruction and practice provide
the baseline explanation on the relationship between SES and
news learning. Education (and other factors such as broad
social contacts) can be compared to instruction and practice

which enhance the levels of processing. In other words,

 

11For example, phonological analysis is automatic.
This does not mean that phonological information is easy to
remember. It means that it can be done without disrupting
other processing. Thus, a subject attending to syntax will
automatically process phonology. But a subject attending to
phonology will not necessarily automatically process syntax
(Perfetti, 1979, p.173).

   

 

 

49

education will facilitate the processing of structural
aspects of mass media messages and therefore reduce the
capacity demand for them. Therefore, it is likely that
those who get more education (high SES) will be able to
allocate more cognitive capacity to the semantic content of
messages and get more learning from them. On the contrary,
for the low SES people, structural aspects of messages are
likely to demand more allocation of cognitive capacity,
which would result in less capacity allocation for semantic
content of the message. Therefore, it is generally expected
that they get less learning than high SES people do from the
same message.

§E§_end_prier_knpniedge. Another group of human
information processing studies (Anderson et. al., 1977:
Bever, 1970: Kintch, 1979: Salomon, 1979: Thorndyke, 1977)
have identified that in the acquisition of new knowledge

prior schemata or internal representations12 play an

 

‘”Schemata or internal representation (mental
representation) means symbols, schemas, images, or ideas
which are the outcomes of the active interactions between
built-in mental structure and the outside nature. The most
important aspect of modern cognitive sciences is the
configuration of mental representation as the main
domain and agenda for systematic cognitive investigation.
This concept is hardly without controversies. No consensus
has been reached regarding the best ways of conceptualizing
mental representations. Behaviorists and neuro-scientists
question this concept itself. However, current cognitive
psychologists widely accept this concept, with the belief
that the investigation of organic mechanism or algorithm
related to mental phenomena can be rightly directed and
interpreted only when the ultimate product and goal of these
‘mechanisms or algorithms, is configured as mental
representation (Gardner, 1985).

50
important role. According to Salomon (1979, p. 90), a coded
message that corresponds less well to one’s schemata, or
internal representations, requires more mental effort for
its recoding. In other words, such a message requires more,
and less well-mastered, mental elaborations. Kintsch (1977)
notes that coding elements that deviate from one's
anticipatory schemata require skills of translation in which
one is not proficient, thus leading to difficulties, errors,
and variations among individuals. For example, with regard
to the comprehension of stories, usually people tell a story
in its natural order, and assuming they do so, the listener
can disregard the cues indicating the order of events.
However, when the story deviates from the natural order of
events, the listener must use the linguistic structure to
recover the natural order (Kintch, 1977, p.315).

Making a similar point, Anderson and others (1977,
p.377) mention that distortions and intrusions will appear
only where there is a lack of correspondence between the
schemata embodied in the text and the schemata by which the
reader assimilates the text. Bever (1970) found that in
order to derive meaning from verbal messages, the listener
uses mainly semantic strategies and performs syntactic
analyses only when necessary. And such analyses are
necessary when a discrepancy between external and internal
modes is experienced. Thorndyke (1977) reports a study in

which the effects of deleting the grammatical orderliness of

51

a story were investigated. When the theme line was deleted
from the story, recall dropped from 80 percent to 58
percent. So did comprehension. Perfetti and Goldman (1974)
also found that recall of a sentence was higher when it was
part of a paragraph than when it was in isolation. Other
studies also found that greater context produces less
processing time unless the subject is explicitly required to
make context judgments during encoding (Dooling, 1972:
Mistler 8 Lachman, 1972).13

These discussions on prior knowledge provide another
important explanation on the relationship between SES and
news learning. The gap in prior knowledge among different
SES groups (Tichenor et. al., 1970) can be interpreted more
exactly as the differences in schemata or internal
representations (symbols, schemas, images, and ideas) rather
than the differences in the amount of issue-specific
knowledge, since education would affect the whole
configuration of these mental phenomena rather than increase
the level of a specific knowledge (public affairs
knowledge). It is likely that higher education is likely to
enrich the prior internal representation level (richer
symbols, schemas, images and ideas about the society).

Therefore, for high SES people, news information will have

 

”Another term which denotes the role of prior schemata
or internal representation in receiving new knowledge
studies is "top-down" processing (Treisman, 1979, p.303). In
this vein, the role played by the communication (cognitive)
skill is called, the "bottom-up" processing.

52
fewer coded messages that correspond less well to their
schemata, or internal representations. Thus, they are
likely to process news information in higher levels.
However, for low SES people, news information will
correspond less well to their internal representation.
Therefore, they will get less learning from it compared to
their higher SES counterparts.

From this discussion, it is expected that SES is
significantly related to news learning. Put differently,
high SES people are likely to learn more about news stories
even when they are exposed to the same amount of news
content through the same media due to the function of
cognitive skill and prior knowledge (internal

representation).

H8: SES will be positively related to news learning.

e ° '0 5
Besides SES, another factor which has been extensively

discussed by past studies (Becker, 1979: Gantz, 1978: Genova
8 Greenberg, 1979: Jeffres, 1978b: Palmgreen 8 Rayburn,
1979: Palmgreen, Wenner 8 Rosengren, 1985: Rubin, 1981,
1983, and 1984: Rubin, Perse, 8 Powell, 1985) with regard to
news learning is motivations. From these studies, a
significant relationship between motivations and actual news
learning has been found. For example, Genova and Greenberg

(1979) found that news interest (i.e., individualized news

53
seeking motivation) rather than SES causes a bigger gap in
news knowledge. Gantz (1978) found that there is a
significant relationship between types of TV news viewing
motivations and amount of news recall. According to his
study, those who mainly sought information from TV news
viewing showed significantly higher recall than those who
mainly sought entertainment.

In deciding the motivations which are closely related
to news learning (news learning motivations), the
distinction of content seeking and process seeking again
provides a very useful conceptual guide. As was discussed
before, content seeking is a set of motivations to gain
knowledge, increase or reduce uncertainty in personal and
social situations, or support existing predispositions. On
the contrary, process seeking is a set of motivations to
escape from reality, to get stimulus or entertainment, and
to overcome social isolation through mediated culture and
its actors.

Therefore, from these two sets of motivations, it is
expected that content seeking motivations are directly
related to news learning.“ Put differently, the stronger a

person is motivated to gain knowledge, reduce uncertainty,

 

“Content seeking should not be confused with
(relative) content gratification expectancy of TV. Content
seeking denotes news learning (seeking) motivations, the
gratification of which is not necessarily confined to a
specific medium use, while (relative) content gratification
expectancy of TV means the intention of using TV (a specific
medium) to gratify these motivations.

 

 

54
or support existing predispositions, the more he/she is
likely to learn about important social issues. From this
discussion, the following hypothesis is proposed:

H9: Content seeking (news learning motivation) will be
positively related to news learning.

TV News Viewing

The dominant conclusion from past TV news studies
(Becker 8 Whitney, 1980: Booth, 1970: Browne, 1978: DeFleur
et. al., 1992: Findahl 8 Hoijer, 1975 and 1985: Katz, Adoni,
8 Parness, 1977: McClure 8 Patterson, 1976: Newman, 1976:
Nordenstreng, 1972: Robinson et. al., 1980: Robinson 8 Levy,
1986: Stauffer, Frost, 8 Rybolt, 1981: Stern, 1973: Wilson,
1974) is that TV news viewing has little effect in news
learning. Two major findings from past studies support this
conclusion.

The first evidence of little news learning from TV news
viewing comes from those studies which investigated the
degree of news recall just after TV news viewing (Katz,
Adoni, 8 Parness, 1977: Neuman, 1976: Nordenstreng, 1972:
Robinson et. al., 1980: Stern, 1973: Wilson, 1974).
Throughout these studies, researchers found that the
audience could recall only a limited amount of information,
or none at all, after viewing television news programs. For
example, Neuman (1976) discovered that half of his subjects
could not recall a specific news story unaided. Katz,

Adoni, and Parness (1977) found that one fifth of television

55

viewers could not spontaneously recall even one item. Even
in an experimental setting, Wilson (1974) found that the
average viewer failed to retain 79 percent of the
information contained in a fictional television news story.

The second evidence of little news learning from TV
news viewing comes from studies which compared news
knowledge level of TV news dependents and newspaper
dependents (Becker 8 Whitney, 1980: Brinton 8 McKown, 1961:
Browne, 1978: McClure 8 Patterson, 1976: Robinson, 1974,
1975). For example, Robinson (1974) found that those who
depended on television were less knowledgeable about
Watergate scandal than those who depended on the newspaper.
In another study (1975), Robinson found that those who
relied on television for political information were more
confused and cynical about politics than those who relied on
other media. Becker and Whitney (1980) who studied the
effectiveness of newspapers and television to inform
residents of central Ohio about political news found that
newspaper dependent people knew more about local issues,
local government officials, and national affairs, while the
findings for television dependent people were just the
reverse.

- te o o v' e e 0 st s ' .

These findings from past studies tend to lead us to conclude
that TV news viewing has little effect on news learning.

However, when we examine the findings from past studies more

56
closely, we can arrive at quite a different interpretation
for them.

(a) Weak recall from TV news viewing: Cognitive
psychology research shows that a rapid loss of information
about surface features of presentation occurs when the
context is similar to that of previous learning episodes
(Weisberg, 1980). This loss seems closely related to the
assimilation of the text information into existing knowledge
structures (Clifton 8 Slowiaczek, 1981: Kintsch, 1976).
News learning can be viewed primarily as the same cognitive
process -- semantic abstraction of news information into
existing knowledge structures. Then, it is expected that
TV news viewers will show a poor recall level of surface or
episodic features. For example, requests to recall material
seen on a particular occasion (e.g., "What was on the news
tonight?" is expected to be relatively ineffective (Berry,
1983). In this vein, it may be quite misleading to
attribute the the poor recall level of TV news found from
past studies to the little cognitive effect of TV news.15

(b) Knowledge level differences between TV news

dependents and newspaper dependents: According to media

 

15This provides one explanation why different SES
groups do not show differences in the amount of news
learning. Since the measurement of news learning in these
studies did not tap into the differences in semantic
learning which are closely related to SES (different
cognitive skill and prior knowledge), but the differences in
recall of episodic features, which have little theoretical
link to SES, it is rather natural that no significant
learning gap between SES groups were found.

57
dependency studies (McDonald 8 Reese, 1987: McLeod 8
McDonald, 1985: McDonald, 1983), media dependency or
reliance is an indicator of the qualitative orientation of
audience members toward current events news and information.
In this vein, television news dependents are generally those
who have a low level of orientation toward current news
events. In other words, television may become the chief
source of news for TV news dependents, because there are
often times when it is difficult to avoid television news if
one is watching television.

Therefore, it is problematic to interpret the findings
from past studies that TV news dependents show less
knowledge level than newspaper dependents as the evidence of
little news learning from from TV news viewing. Put
differently, the differences in news knowledge level between
TV news dependents and newspaper dependents can be
attributed to the differences in general orientation toward

news/information rather than the medium factor itself.“

 

“This provides an important re-interpretation of
another major evidence of the little knowledge gap effect
of TV news. The difference in news interest/ selectivity is
one of the crucial factors which brings the variability of
news knowledge among different SES groups. To examine the
relationship between SES and news knowledge among TV news
dependents is to examine the effect of SES on the news
knowledge level, controlling this very factor of variability
due to SES difference. Furthermore, it is possible that
high SES TV news dependents might include the segment of
high SES people who have particularly little interest in
news information while low SES TV news dependents represent
those who have average or above-average news interest among
low SES people. This may be why low SES TV news dependents
show a higher level of news knowledge compared to a low SES

58

This re—interpretation of the findings from past
studies, leads us to question the common belief regarding
little news learning from TV news viewing. It might be true
that TV news is not as effective as other media such as
newspaper in news learning (DeFleur et. al., 1992). Still,
the actual news learning effect from TV news viewing seems
to be too much underestimated.

Role pf TV news in news ieerning in broegcast-ghennel-
onl nd m '- h nne s tua ons. Considering the fact
that news learning from TV news viewing may not be so
limited as is generally perceived, we can now turn to the
discussion of how the relative role of TV news viewing in
news learning will be different in broadcast-channel-only
and multi-channel situations.

In a broadcast-channel-only situation, TV news is
composed of a relatively limited amount of news information,
the major content of which is significantly overlapped with
other news media content. Therefore, by the time viewers
watch evening TV news, they may be quite familiar with most
of TV news stories already. Still, since TV news viewing
during TV news hours is almost compulsory and there exists
little selectivity among news stories, no matter how

redundant the news information is, viewers have ilStle

 

group average, while high SES TV news dependents show a
lower level of news knowledge compared to a high SES group
average in the 1983 University of Maryland survey and 1980
University of Michigan survey (Robinson 8 Levy, 1986).

59
choicerbut'to watch TV news unless they decide to turn off
TV.H*NW “I“

Therefore, the relationship between the amount of TV
news viewing and news learning in a broadcast-channel-only
situation is likely to be affected (reduced) by ceiling
effect. Ceiling effect means that variability of a variable
(for example, learning from TV news) is not limitless but
has an upper or lower limit (ceiling)."7

However, in a multi-channel situation, viewers have
many different news program options at the same time, and
news viewing during TV news hours is not even compulsory.
Therefore, if they find a TV news program quite redundant or
superficial, viewers can always turn to other news programs
which they perceive as better, or non—news programs if no
news programs are satisfactory. Therefore, in a multi—
channel situation, news learning from TV news viewing is not

likely to be affected (reduced) by ceiling effect so much as

is the case in a broadcast-channel-only situation.

 

‘"Shingy and Mody's study (1976) on the effect of
educational TV in rural Indian villages provides a good
example of ceiling effect. They found that the more
frequent a farmer’s contact with village organizations and
change agents, and the larger his ownership of agricultural
implements, the less information was gained from the low-
ceiling educational TV programs.

It should be noted that the notion of ceiling effect
here is mainly applied to the cognitive dimension (i.e.,
news information content) of TV news. It is not applied to
the affective dimension (i.e., vivid and rich visual
content) of TV news.

60

The fact that ceiling effect is minimal in a multi—
channel situation is likely to affect both (a) viewers'
evaluation of TV news, and (b) the role of TV news in news
learning. First, with regard to viewers evaluation of TV
news, TV viewers in a multi-channel situation will conceive
TV news as providing more variety and in-depth news stories
compared to TV viewers in a broadcast-channel-only
situation. Also, TV viewers in a multi-channel situation
are likely to evaluate TV as gratifying various news viewing
motivations better (expectancy of TV to gratificy content
seeking) than TV viewers in a broadcast-channel—only
situation. From this, the following set of hypotheses
regarding the perception of TV as a news medium are
proposed:

H10: Multi—channel TV subscribers will perceive TV news

as providing more variety in news stories than non-

subscribers.

Hll: Multi-channel TV subscribers will perceive TV news

as providing more in-depth news stories than non-

subscribers.

H12: Multi-channel TV subscribers' evaluation of TV in

gratifying news learning motivations (content seeking)

will be higher than non-subscribers.

Secondly, since ceiling effect does not affect news
learning from TV news viewing in a multi-channel situation
as much, the amount of TV news viewing in a multi-channel

situation is expected to have a stronger effect on news

learning. Especially, 24-hour news viewing is likely to

61

have a stronger relationship with news learning than
broadcast news viewing, since 24-hour news will be less
affected by the ceiling effect. Therefore, with regard to
the relationship between the amount of TV news viewing and
news learning, the following hypotheses are proposed:

H13: The amount of TV news viewing will be related to

news learning more strongly in a multi-channel

situation than in a broadcast—channel-only situation.

H14: In a multi-channel situation, the amount of 24-

hour news viewing will be related to news learning more

strongly than the amount of broadcast news viewing.
0t e actor

So far, the theoretical relationships of SES, news
learning motivations, and the amount of TV news viewing with
news learning have been discussed. Besides these, news
learning is contingent upon many other factors. This study
will include the amount of other news media uses and other
demographic variables as these factors.

aner_nene_negie_n§e§. Other news media such as
newspaper, radio, and interpersonal communication is likely
to affect news learning. It is likely that the more
frequently a person uses these media, the more news learning
(s)he will achieve.

a ' v ' e . This study will include two

major demographic factors, age, gender, and household income
which have been frequently reported as having a significant

relationship with news learning (e.g., Gantz, 1978).

62

Newe Leerning Model

From the discussion so far, we can propose a
theoretical model of news learning (Figure 4) which includes
TV news viewing as a factor. The role of TV news in news
learning, symbolized by the link between these two
variables, is likely to be affected by the other factors
included in this model as well as the development of multi—
channel TV. However, if the influences of other factors of
news learning are assumed to be constant in broadcast—
channel—only and multi-channel situations, the relationship
between TV news viewing and news learning is likely to

become stronger in a multi-channel situation.

63

 

Amount of
TV news
viewing

News

Other ’fprlHaer”" Learning

Media
Exposure

 

 

 

 

 

 

 

 

 

 

 

 

 

News Learning
Motivation

 

 

 

Demographic
Factors

 

 

 

Fignre 4. The Theoretical Model of News Learning

CHAPTER III.

RESEARCH METHODOLOGY

Introduetion

This chapter provides an overview of the data
collection and analytical procedures employed in the present
study. Briefly stated, after the measures of key variables
were constructed, a telephone survey was executed to collect
data. All interviews were conducted from a central
supervised location by highly trained interviewers for a
three week period: 303 interviews were completed. The
survey data were analyzed using the Statistical Package for
Social Sciences (SPSS) program. A series of analytical
techniques, including Pearson's correlation, multiple
regression analysis, chi-square test, and t—test were

utilized.

Measnremen; of Key Veriapiee
For this study, the following variables need to be

measured: SES (socio—economic status), content seeking and
process seeking, content gratification expectancy of TV
(CGE-TV) and relative content gratification expectancy of TV
(RCGE—TV), viewer availability, news channel awareness,

channel repertoire, value perception of 24-hour news

64

65
channels, amount of TV news viewing, amount of other media

exposure, and level of news learning.

Socio-Economic Statue (SES)

As is the case in most mass media studies, education
level was adopted as the primary indicator of SES. To
measure education level, eight different hierarchical levels
of education were constructed: (a) grade school only, (b)
high school not completed, (c) high school graduate, (d)
vocational or technical school, (e) some college, (f)
college graduate (BA, BS), (g) post graduate work, and (i)

graduate degree (MA, MS, PhD).

0 t ee 'n n o e See

In order to construct the measures of content seeking
(motivations) and process seeking (motivations), various
media use motivation statements, which were constructed by
past uses and gratification studies, were reviewed. From
this, two opposite-direction scales, composed of four 5-
point items, were constructed for content seeking (Ch) and
process seeking (P,), respectively. First, content seeking
was measured by the following four items:

(a) I try to make myself a more knowledgeable person

from what I read or view.

strongly agree 1 2 3 4 5 strongly disagree

(b) I am the type of person who feels left out if I'm
not familiar with what's going on in society.

(c) It is important to me to have well-informed
opinions about controversial issues.

66

(d) I like to hear the ideas of others in order to have

more objective ideas of my own.

In this study, content seeking was equated conceptually
to news learning motivation. Therefore, the sum of these
four item scores (EC,) is interpreted as a measure of the
strength of news learning motivations. Process seeking was
measured by the following four-item scale:

(a) If I can be entertained from what I read or view, I

don't care much about the informational content.

strongly agree 1 2 3 4 5 strongly disagree

(b) I tend to turn to what I read or view just to pass
the time.

(c) I tend to avoid complicated material that I have to
try hard to understand.

(d) I tend to read or view material to relax rather

than to gather new information.

Once these scales were constructed, the evaluation of
TV in gratifying each content seeking and process seeking
motivation was measured by asking respondents to rate TV
with relation to each item above (e.g., How would you rate
television as a way to make you more knowledgeable?). The
sum of the evaluation scores for four different content
seeking items (2B,) indicates the overall evaluation of TV
as a news/information medium. In the same vein, the sum of
the evaluation scores for four different process seeking
statements shows the evaluation of TV as an entertainment/

escape medium.

67

CGE-TV and RCGE-Ty

Once content seeking and process seeking, and
evaluation of TV were measured, CGE—TV and RCGE—TV can be
computed by the equations presented in Chapter II. For
example, CGE-TV of a respondent is computed by the following
procedure: (a) the score for a content seeking statement is
multiplied by the score of a corresponding TV evaluation
statement (C,:{Ih), and then (b) the four product scores
are summed (2C,:(IL). Relative content gratification
expectancy of TV is computed by dividing content
gratification expectancy of TV (2C, x 13,) by total

gratification expectancy of TV (TGE-TV).

Viewer eveilepility

Viewer availability was measured by asking respondents
how many days a week they watch TV during traditional TV
news hours. There are two major TV news hours in the
evening -— 6-7 PM and 11—11:30 PM. The former is the
primary news hour for both local news and network news. All
three major network (ABC, CBS, NBC) affiliates broadcast
news during this hour. From 11 to 11:30, all three major
network affiliates broadcast local news. Therefore, the
number of days a week a respondent watches TV during 6-7 PM
was designated as viewer availability 1, and the number of
days a week a respondent watches TV during 11-12 PM (to make
the purpose of the measure less obvious, 11—11:30 PM was

replaced by 11-12 PM) as viewer availability 2.

68

News Channel Awareness

News channel awareness was measured by asking
respondents whether they can describe the major programs
provided by C—SPAN and C-NBC. When a respondent described
both channels correctly, 2 points were given. When only one
channel was correctly recognized, 1 point, and when neither
of them were correctly described, 0 point was assigned,

respectively.

Channei Repertoire

Channel repertoire is defined as a set of channels
turned on or watched intentionally. From this, the
inclusion of news channels in the channel repertoire was
tapped in by asking the following question.

Pe0ple tend to turn to a set of cable channels

intentionally, and other channels, unintentionally by

channel hopping. Would you say you turn to CNN or
other cable news channels intentionally or

unintentionally?
Veine Pereeptien of gene Qnenneis

Value perception of news channels is defined as the
monetary value a viewer put on news channels. To get at
this value perception, the following question was asked.

If you had to, would you pay an additional fee to

receive these news channels (CNN, CNN-Headline News,

C-SPAN, and CNBC)?

When the answer was clearly affirmative, 2 points were

given. when the answer was firmly negative, 0 points were

69
given. For those who said neither 'yes' nor ’no’ clearly, I

point was assigned.

nnount of TV News Viening

The first major dependent variable is the amount of TV
news viewing. In this study, two major news types were
distinguished: broadcast news and 24-hour news. Broadcast
news means both local and network news programs, which
deliver daily news events covered by news reporters or fed
by other news agencies with the traditional format of
newscasting. In this study, tabloid—type news shows,
celebrity interviews, news dramas, and news clips included
as a part of other programs, etc. were excluded from
broadcast news.1

Network news covers mostly national and international
level news stories. Local news tends to focus on state (the
state of Michigan) and regional level (Lansing and Jackson
areas) news events. However, it also covers major national
and international level news events from a local

perspective.” In this study, the amount of local news

 

1The distinction between traditional news programs and
these entertainment oriented news shows, however, has become
somewhat unclear, as the former (especially, network news)
has gradually taken the format of the latter for the last
few years. The ideas such as "USA Todaying of news"
(Broadcasting, Sep. 24, 1990, pp.34-39) or "network evening
newscasts at times are indistinguishable from such
syndicated tabloid fare as Hard Copy (Broadcast, Feb. 3,
1992)" reflect this trend.

”For example, the national and international—level news
events (e.g., the flood, G-7 Summit Meeting, and Gays in the

70

viewing and network news viewing were measured by the number
of days in a week a view watches local and network news
programs. Then, the amount of broadcast news viewing was
obtained by computing the mean of these two.

Twenty-four—hour news is defined as news programs
delivered by specialized news channels such as CNN, CNN-
Headline News, C-SPAN, and CNBC. In these channels it is
very difficult to distinguish news programs and non-news
programs. Therefore the amount of 24-hour news viewing was
measured by the number of days a week a viewer turns to

these channels.

nnount pf Other Newe Megie U§e§

In this study, as other news media besides TV news,
newspaper, radio news, and interpersonal communication were
included. As was the case in the measurement of the amount
of TV news viewing, the amount of these news media uses was
measured by the number of days in a week a respondent is

engaged in these media use activities, respectively.

Leyel of Newe Leerning

In this study, the level of news learning (knowledge)
is defined as the degree to which a respondent is aware of

important news events in society. In specifying what

 

Military) which were used in this study to measure viewers’
news knowledge level were covered by local news, even though
the major focus of news coverage was somewhat limited on the
state level and regional level aspects/implications of those
events.

71
important news events in society mean, two conditions need
to be carefully considered: (a) they should have a
significant meaning for most social members rather than a
specific segment of social members in order to ensure that
the individual-specific interest differences do not affect
the variability in news learning, and (b) these events
should be covered extensively by all mass media, and the
coverage for them should last for a while in order to ensure
that the variability in news learning is directly related to
normal news media exposure rather than other special
activities or occasions (e.g., library search).

Four news events during these periods were evaluated as
satisfying these two conditions. These stories include "The
War in Bosnia," "G7 Summit Meeting in Tokyo" (international
news events), "The Flood in the Midwest," and "Gays in the
Military" (national news stories). These news events had
general interest for most society members. All these news
events were running news stories for more than one month by
the time the survey began and were covered extensively by
various news media including TV and newspapers. A brief
description of each news event is presented in Appendix I.

For the actual measurement of the level of news
learning (knowledge), the respondents were asked whether
they were aware of each of these news events. When the
answer was affirmative, then a probe question in a multiple—

choice question form was given to check the accuracy of

72
their knowledge more closely. For example, the questions
used for "The War in Bosnia" were as follows:

Are you aware of the war in Bosnia (formerly
Yugoslavia) ?

1 Yes

2 No

(If yes) From what you heard, is the war a result of a
conflict between students and the government: a
conflict among countries in Eastern Europe: or a
conflict among different ethnic groups?

1 the conflict between students and the government
2 the conflict among countries in Eastern Europe

3 the conflict among different ethnic groups

4 other

5 don't know

Research Deeign

The objective of this research is to investigate the
changes in TV news viewing and news learning from it due to
the development of multi—channel TV. For this it is
necessary to compare the amount of TV news viewing and the
news knowledge of broadcast-channel-only TV viewers with
those of multi~channel TV viewers. To obtain data which
enable this cross-sectional comparison of these two groups,

a telephone survey was designed and administered.

Seiecrion pf Senpie

Channel offerings and channel arrangements which are
different among multi-channel TV systems can affect overall
program choice and TV news viewing seriously. In order to
control these factors, it is necessary to select the sample

from the areas covered by one multi-channel TV system.

73

In this study, the sample was selected from City of
Lansing and the adjacent areas3 (total population 206,425)
in mid-Michigan which are covered by Continental
Cablevision. In these areas, multi-channel TV subscription
rate (10/31/91) is 66.7 percent (59,376 subscribers/ 89,000
households): this is somewhat higher than the national
average, 60.6 percent (Aug, 1993). Premium channel
subscription rate (11/26/91) is 54.0 percent (32,089 premium
channel subscribers/ 59,376 basic subscribers).

Broadcast-channel-only TV viewers in these areas
(Lansing/Jackson market) receive five TV channels (ABC, CBS,
NBC affiliates, a Fox affiliate, and a local PBS channel).
In addition to these channels, multi-channel TV subscribers
in these areas receive a variety of cable network channels
including major 24-hour news channels such as CNN, CNN
Headline News, C—SPAN, and C-NBC, off-the-air distant
channels originating from other TV markets, and other access
and information channels originating from the cable system.
Table 3 shows the channels available for multi-channel TV

subscribers and broadcast-channel-only viewers.

----*-—--‘--~------‘

 

3These areas include Clinton County, Eaton County,
Ingham County, Alaiedon Township, De Witt, De Witt Township,
Delhi Township, Delta Township, Eaton Rapids, Grand Ledge,
Lansing Township, Oneida Township, and Watertown Township.

Table 3

74

Channels Available for Broadcast-Channel—Only Viewers and
Multi-Channel TV Subscribers in the Sampling Areas.

 

Broadcast TV

WLAJ (ABC), WLNS (CBS), WILX (NBC)
WSYM (Fox), WKAR (PBS)

 

Multi-
Channel
TV

 

 

 

 

Distant WKBD (Fox), WTVS (PBS) Detroit: WJRT

Channels (ABC) Flint: WOOD (NBC) Grand Rapids:
WWMT (CBS) Kalamazoo

Cable WTBS, WGN, A&E, BET, C-SPAN, CNBC,CNN,

Networks Discovery, ESPN, Family, CNN-Headline
News, Home Shopping, Lifetime, MTV,
TNN, Nickelodeon, Weather, TNT, USA,
Univision

Origi- time-weather, bulletin board, public

nations access, educational access, government

 

access, religious access, library
access, emergency alert, public
service announcements, local sports.

 

Source: Ieievieion 8 geple Eacrpeok, 1993.

 

75

After sampling areas were decided, a list of working
phone exchanges (the first 3 numbers of a phone number,
e.g., 355) for those areas were generated using information
provided by the phone company and phone directories. The
sample size of the project determines the number of working
phone banks (first five digits of a phone number, e.g., 355-
12) that need to be identified to randomly generate the
sample. It is standard practice at the Institute for Social
Research at the University of Michigan that six interviews
be taken from working phone banks (Waksberg, 1978). Since
the intended sample size of this study was 300, at least 50
working household phone banks needed to be interviewed.

Once the 50 banks were identified, using a formula of
(a) a 75 percent hit rate (meaning that of all the phone
numbers that have been identified, 75 percent will be
working and the other 25 percent will be disconnected,
businesses, non-working numbers, etc), (b) a 95 percent
eligibility rate (meaning that 95 percent of the households
contacted would be eligible: those ineligible would be non-
English speaking households, no adults over 18, etc.), and
(c) a 75 percent completion rate, a total of 561 numbers
(about 11 numbers per bank) were randomly generated to

achieve 300 completed interviewse‘ Phone numbers were

 

‘The actual computation of phone numbers is:
300 / (.75) x (.95) x (.75) = 561

 

 

76
replaced as necessary to ensure at least six completed

interviews from each phone bank.

Adminietration of the Suryey

Telephone interviews were conducted by highly trained
interviewers of the Institute for Public Policy and Social
Research (IPPSR) at Michigan State University over a three-
week interval (August 13,1993 to August 28, 1993).5 Before

HH...

few ambiguous questions were revised. The finalized version

_. -.....H-H urn—IHAflm» a...
W—H. .

of the survey instrument is reproduced in Appendix II.

The interviewing occurred in a supervised, central
location. Calling sessions were scheduled from 10 AM to
9:30 PM, seven days a week. Interviewers were instructed to
let the telephone ring at least 10 times before coding the
attempt as "no answer." All telephone numbers yielding
results of "no answer" or "busy" were called back for the
maximum number of 20 times. For randomi-zation, the one who
had the most recent birthday at the home numbers answered

was asked to respond to the survey.

 

”Since a three week survey period is fairly long, the
differences in interview dates can become a confounding
factor affecting on the level of news learning. For
example, the respondents who were interviewed last would
have had more exposure to the running news stories than the
earlier ones. Therefore, in order to examine the
differences in the level of news learning due to the
differences in interview dates, interview date was included
as an independent variable in the analysis.

77

The actual survey yielded 303 completed interviews.
The major production statistics of sample after the survey
is presented in Table 4. The completion rate and refusal
rate are important measures of data quality. Usually, if
the completion rate is close to 65 to 70 percent, it is
considered desirable. The completion rate of this study (79
percent) is substantially higher than this criterion value.
Moreover, the refusal rate of 13 percent is substantially
below the median refusal rate of 28 percent (Wiseman 8
McDonald, 1979). Therefore, the administrative attempts to

control data quality were successfully achieved.

't' f a

The composition of the sample is presented in Table 5.
Of the total 303 cases, 208 cases (68.6 percent) were cable
subscribers and 95 cases (31.4 percent), non-subscribers.
This result is quite consistent with the 66.7 percent multi—
channel TV subscription rate in the sampling areas. The
sample included a very wide range of age groups which
spanned 18 to 92 (mean=44.40, median=40, mode=33). In terms
of gender, females (184: 60.7 percent) were somewhat more
sampled than males (119: 39.3 percent). The most common
family size was 2 (mean 2.543, median=2, mode=2). The
education level of the sample showed a wide range of
variations from grade school only to graduate degrees

(median, mode = some college).

Table 4

78

The Major Production Statistics of the Sample

 

Interview dates

August 10, 1993 to August 28,1993

 

Average interviews/day

16.8 interviews

 

Average length/interview

12.71 minutes: SD 4.51, minimum 7
minutes, maximum 51 minutes

 

Average number of call
attempts

3.51 call attempts: SD 3.21:
minimum 1, maximum 20 attempts

 

Completion rate

79 percent: Completed interviews
(303)/(completed interviews [303]
+ refusals [51] + non-interviews*

[30])

 

Refusal rate

 

 

13 percent (Refusals /[completed
interviews + refusals +
non-interviewSJ)

 

* Non-interviews include the following: (a) potential
respondents who are incapable of answering due to
illness, physical limitations, language barriers, or

respondents gone for

the duration of the study: (b)

respondents "left over" at the end of the calling
period (cases that were contacted and determined to be

eligible households,
a result of reaching

but an interview did not result as
the desired number of interviews):

(c) cases that reached the call limit (20) with at

least one contact to
household.

determine that the number was a

 

79

Table 5.
Sample Composition

 

 

 

 

 

 

 

 

 

 

 

 

Valid Cum
Variable Value Pregnancy Percent Percent Percent
888
18-20 12 4.0 4.0 4.0
21-30 64 21.1 21.3 25.2
31-40 78 25.7 25.9 51.2
41-50 48 15.8 15.9 67.1
51-60 35 11.6 11.6 78.7
61-70 36 11.9 12.0 90.7
71-92 28 9.2 9.3 100.0
2 .7 Missing
Total 303 100.0 100.0
Gender
MALI 119 39.3 39.3 39.3
IRMALB 184 60.7 60.7 100.0
Number of
ramily Members 1 69 22.8 22.8 22.8
2 109 36.0 36.1 58.9
3 45 14.9 14.9 73.8
4 54 17.8 17.9 91.7
5 20 6.6 6.6 98.3
6 5 1.7 1.7 100.0
1 .3 Missing
Education Level
GRAD! SCHOOL ONLY 2 .7 .7 .7
HIGH SCHOOL NOT COMPLETED 18 5.9 6.0 6.6
RIG! SCHOOL GRADUATE 81 26.7 26.9 33.6
VOCATIONAL/TECHNICAL SCROOL 8 2.6 2.7 36.2
80MB COLLEGE 112 37.0 37.2 73.4
COLLEGE GRADUATB (BA, BB) 49 16.2 16.3 89.7
POST GRADUATE 6 2.0 2.0 91.7
GRADUATE DRORIR (MA,M8,Pb.D) 25 8.3 8.3 100.0
2 .7 Missing
Race lthnicity
BLACK OR AFRICAN AMERICAN 23 7.6 7.8 7.8
WHITE OR CAUCABIAM 251 82.8 84.8 92.6
ASIAN OR PACIFIC ISLANDS! 4 1.3 1.4 93.9
NATIVE AMERICA! 5 1.7 1.7 95.6
HISPANIC 13 4.3 4.4 100.0
7 2.3 Missing
Occupation
UPPER WHITE COLLAR 49 16.2 16.3 16.3
LOWER WHITE COLLAR 62 20.5 20.6 36.9
UPPER BLUE COLLAR 23 7.6 7.6 44.5
LOWER BLUE COLLAR 39 12.9 13.0 57.5
GOVERNMENTAL INPLOYBB 16 5.3 5.3 62.8
PARMIR 2 .7 .7 63.5
ROUSRWIFR-ROUSBRUSBAMD 36 11.9 12.0 75.4
STUDENT 20 6.6 6.6 82.1
WWWYID 4 1.3 1.3 83.4
RRTIRID 50 16.6 16.6 100.0
2 .7 Missing
Income
L888 TEAM 810,000 16 5.3 5.9 5.9
$10,000 - $14,999 17 5.6 6.2 12.1
$15,000 - 919,999 27 8.9 9.9 22.0
$20,000 - $24,999 24 7.9 8.8 30.8
$25,000 - $29,999 21 6.9 7.7 38.5
$30,000 - $34,999 35 11.6 12.s 51.3
$35,000 - $39,999 17 5.6 6.2 57.5
$40,000 - 844,999 22 7.3 8.1 65.6
$45,000 - $49,999 7 2.3 2.6 68.1
850,000 - $54,999 2 .7 .7 68.9
855,000 - $59,999 29 9.6 10.6 79.5
OVRR 60,000 56 18.5 20.5 100.0
30 9.9 Missing
Total 303 100.0 100.0

 

 

 

80
Therefore, the actual survey obtained representative
numbers of broadcast-channel-only TV viewers and multi-
channel TV viewers as cases, and a wide range of variations
in demographic variables such as age and education levels.

In this vein, too, the survey results were very successful.

Analyticai Tecnnignes Used in Hyporhesie Testing

Three inferential statistical procedures were used for
the testing of the research hypotheses: Chi-square test, t-
test, Pearson’s product-moment correlation, and multiple
regression analysis. The alpha level was set at .05 for all
tests. Although the hypotheses were directional, they were
tested by more conservative two-tailed statistical tests.
SPSSPC+ (version 5.0) was used for the various statistical
tests. The results will be provided in tables in the next

chapter.

CHAPTER IV

RESEARCH RESULTS AND HYPOTHESES TESTING

Introduction

The purpose of this chapter is to present the results
of the statistical analyses testing the hypotheses proposed
in Chapter II. Major hypotheses testing in this research
involves the cross-sectional comparison of TV news viewing
and news learning from it between broadcast-channel-only TV
viewers and multi-channel TV viewers. Therefore, in
composing this chapter, the statistical analysis results of
amount of TV news viewing and news learning from it in each
TV viewing situation will first be presented. Then the

results of the hypothesis test will be addressed.

eneunr of TV News Viewing
The first objective of this study is to identify the

various theoretical factors of TV news viewing and to
investigate their relationships with the amount of TV news
viewing in both broadcast-channel-only and multi-channel
situations. The theoretical predictors explicated in this
study include SES, TV news viewing intention (CGE-TV and
RCGE—TV), viewer availability, news programming(channel)

awareness, channel repertoire, and value perception of news

81

82

channels. Among these variables, SES, TV news viewing
intention, viewer availability are relevant to TV news
viewing in both broadcast—channel-only and multi-channel
situations. The other variables such as news channel
awareness, channel repertoire, value perception of news
channels are only relevant to a multi-channel situation.

Besides these variables, TV news viewing is necessarily
contingent upon other factors, such as the overall amount of
TV viewing, the amount of other media exposure (the amount
of newspaper reading, the amount of radio listening, and the
amount of interpersonal communication) and other demographic
factors (age, gender, number of children, and household
income). These variables were also included in the

analysis.

Av Am u t o V'ew‘ d TV N ws V' 'n

The average amount of TV viewing for the whole sample
(298 cases, 5 cases missing) is close to 3 hours, ean =
2.92, SD = 1.88, nedien = 2 hours, npge = 2 hours. The
average amount of TV viewing in a broadcast-channel—only
situation (94 cases) is a little less than two and half
hours, nenn = 2.36, S9 = 1.71, while the average amount of
TV viewing in a multi-channel situation (204 cases) is a
little more than 3 hours, neen = 3.17, SQ = 1.91.
Therefore, these two groups reveal substantial differences

in the amount of TV viewing (.81 hour: about fifty minutes),

83
r = 3.52, 23 = 296, p < .001. This result coincides with
the findings from past TV audience studies (Becker,
Dunwoody, 8 Rafaeli, 1983: Gelman, 1983: Grotta 8 Newsom,
1983: Henke et. al., 1984: Reagan, 1984: Webster, 1984) that
multi-channel subscribers tend to watch more TV than
broadcast-channel-only TV viewers. Table 6 summarizes these

findings regarding the average amount of TV viewing.

With regard to the amount of TV news viewing,
broadcast-channel-only TV viewers watch broadcast news about

four days a week, mean = 3.99, SQ = 2.42, while multi-

 

channel TV viewers watch broadcast news a little more than
four days a week, neen = 4.30, S9 = 2.22. The results
demonstrate that no significant differences exist between
these two groups, r = 1.09, 2: = 300, p =.274. Multi-
channel TV viewers also turn to 24-hour news about three to
four days a week, neen = 3.61, S2 = 2.50, and about one hour
in daily average, neen = 1.04 hour, SQ = .96. Therefore,
multi-channel subscribers watch as much broadcast news as
the broadcast-channel-only viewers and also watch 24-hour
news. Thus, on average, we can conclude that multi-channel
TV viewers watch TV news more than broadcast-channel-only

TV viewers. Table 7 summarizes these findings.

Table 6.

84

Average Amount of TV Viewing a Day

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Number Mean SD t- DF 2tail
of cases value Sign.
Broadcast TV Viewers 94 2.36 1.71
3.52 296 .001
Multichannel TV Viewers 204 3.17 1.91
Total 298 2.92 1.88
Table 7.
Average Number of Days of TV News Viewing
Number Mean SD T- DF 2—tail
of cases value Sign.
Broadcast News* 95 3.99 2.42
1.09 300 .274
Broadcast News** 207 4.30 2.22
24-Hour News 201 3.61 2.50

 

 

 

 

 

 

*
**

broadcast-channel-only situation
multi-channel situation

 

 

85
TV Nens Viewing in a Broadcast-Channel-Only Situation

In a broadcast-channel-only situation, the major
theoretical factors such as the amount of viewing, SES, TV
news viewing intention (CGE—TV), and viewer availability
were proposed. In addition, other related -actors such as
RCGE-TV, the amount of TV viewing, the amount of other media
exposure, and other demographic factors (age, gender, the
number of family members, and household income) were
included in the analysis.

Cerreiation matrix. In order to determine (a) how the
amount of TV news viewing is related to the independent
variables and (b) how independent variables are related to
one another (which has a crucial meaning with regard to the
following multiple regression analysis), all possible sets
of correlations among all the variables were performed.

Table 8 shows the resulting correlation matrix.

According to this correlation matrix table, the amount
of broadcast news viewing is significantly correlated with
the two major theoretical factors, CGE-TV, r:.3217, p <.01,
and viewer availability 1, r = .6844, p < .001. It is also
substantially correlated with the amount of TV viewing, r =
.4793, p < .001 and age, r = .4719, p < .001.

Examination of the correlation matrix shows a complex

pattern of relationships among independent variables (CGE-TV

E36

HasHo H.

888ng HERE on 33.38 aoHHnoa 8 .96 En OH 2 HHS; 32:5 5 a Roman-fiAHHHEoHSHHH $8.38

Ha HHS; mum

38:5
mum -.HHHH
conned .HHHH.H .oaoo
wooH-Ha .OHHH -.OHHH
HesHHpuHHHwH H .HHHH...-.HHHH
paHHHmvHHHnH H .HHHH .oHHo
Ha «HmeHam .HHHH...-.HHoH
zmcmvanon .Hon -.HHHH
xHaHo amen -.ooaH -.HHHH
HenoneonuossH .HHHH -.HHHH
Hmo .HHHH...-.HHHH
ooaaou .HHHH -.HHHH
HHBHHH HHHo -.HHoH -.HHoo
Hsoosm .OHHH .oHHH

m cm apnea” HH

8?? ”8?: EHHHH

.Homme
.HHHH.
.Hims
.Hmso.
.OHH.
.HHHH
.Humo
.HHHH

-.HHHH

-.OHH.
.HHHH

.HHHH
u.HmmH
u.HoHH

.oHHH

.oHHH

.HHHH

.HHHH
u.oHHH

.oHHH

.HoHH

HaHHHH

-.oHHo

.HHHH... .HHOH

.OHHH -.OHHH

.ooeH -.HHHH

.oHHH -.oHHH

.HHHH...-.HHHH
-.OHHH -.oHHH
-.HHHH .oHHH

.HHHH -.HHHH

.2 H2? x88 H32..- So 838.. HEHHH
«HacHsm Haven amen voumosuH HHHo
-.OHHH

.OHHH .HHHH

.HHHH .HHHH... .HHHo...

.HHHH .HHHH. -.oHuH .ooHH

.HHHH -.oHHH -.HHHH .oHHo -.OHHO
-.HHHo -.OHHH .oooH .OHHH -.HHoH .HHHH
-.OHHH .HHHH. .HHHH. .HHoo... .HHHH -.HHHH .HOOH

H-HHHHoa mHHSHHHoHSoo . HA .ou. .. HA .oH. ... HA .ooH

87
and RCGE-TV, CGE—TV and viewer availability 1, CGE-TV and
amount of TV viewing, viewer availability and the amount of
TV viewing, viewer availability and age, newspaper reading
and interpersonal communication, newspaper reading and age,
newspaper reading and household income, radio news listening
and interpersonal communication, radio news listening and
household income, and interpersonal communication and
household income). These results show that independent
variables in this study are not independent but are closely
inter-related to one another (the implications of these
correlations among independent variables will be discussed
in Chapter V).

M t' s 'o n ' . In order to determine
the relationship between all the independent variables
together and the amount of broadcast news viewing in a
broadcast-channel-only situation, a multiple regression
analysis was carried out. The examination of the
scatterplot and residual plot of each independent variable
by dependent variables did not show any strong evidence of
non-linearity, heteroscadacity and the existence of
significant outliers. Table 9 shows the results of the

analysis.

The multiple correlation coefficient (R) is .82 which

is interpreted as the correlation between the entire

88

Table 9.
Multiple Regression Analysis Results of the Amount of
Broadcast News Viewing in a Broadcast—Channel-Only Situation

 

 

 

 

Variable B Beta F Sig F
AGE .035813 .226268 5.921 .0180
GENDER 1.022736 (212895 6.912 .0109
EDUCATION -.087226 -.055542 .474 .4940
NO. OF FAMILY -.307541 -.129803 2.574 .1140
AVAILABILITY 1 .542642 .534348 32.337 .0000
AVAILABILITY 2 .257216 5246668 9.330 .0034
AMOUNT OF TV .147887 .103404 1.197 .2783
CGE-TV .011201 .090251 1.000 .3214
RCGE-TV -1.339151 -.073926 .739 .3935
INTERPERSONAL .094178 .096019 .857 .3584
RADIO NEWS -.036081 -.043721 .236 .6287
NEWSPAPER -.014207 -.015787 .033 .8561
INCOME .003353 .005212 .003 .9535
(Constant) -1.234921 .624 .4327
Multiple R .82306

R Square .67743

 

 

 

89
independent variables together and the dependent variable.
The R square (R”) is .68, which means 68 percent of the
whole variance in theamount of broadcast news viewing is
explained by the independent variables included in the
multiple regression equation.

In the multiple regression analysis, it is a common
practice to interpret the Beta coefficients as showing the
relative importance of independent variables (the true
relationships between independent variables and a dependent
variable). However, what a Beta coefficient means is the
unique contribution of an independent variable to the
variability of a dependent variable, or in other words, the
correlation between an independent variable and a dependent
variable after the influences of all the other independent
variables are partialed out.

When independent variables are strongly correlated with
one another as is the case in this study (multi-
collinearity), neither the correlation coefficients nor Beta
coefficients reflect in any absolute sense the true
relationships between independent variables and dependent

variables.1 Until the causal relationships among all the

 

1It is misleading to interpret that a Beta coefficient
(multiple partial correlation coefficient) shows the true
relationship between two variables. The following example
illustrates this.

education level .7

.6 ‘\“‘\H\‘ amount of TV news
’fifl”ffl’r,HHa—~' viewing
income .4

90

independent variables included in this study are revealed
theoretically, no statistical techniques known so far
provide a satisfactory solution. Therefore, in this study,
rather than depending solely on one of these two, both
correlation coefficients, as the indicators of the observed
relationships and Beta coefficients, as the indicators of
the unique relationships after the influences of all the
other variables are controlled, will be used in
complementary fashion for model constructions and hypothesis
testing.

The following model of news viewing in a broadcast—

channel-only situation (Figure 5) is based on the combined

 

In this case the Beta coefficient of education level will be
.46 (.7 — .6 x .4) and the Beta coefficient of income will
be —.02 (.4 — .6 x .7). The problems of these Beta
coefficients become obvious if the true relationships among
these variables are assumed as follows:

.6 .7
income education level-——————-amount of TV
news viewing

 

In this case, the true relationship between income and
amount of TV news viewing is 0. Therefore, the Beta above
reflects the true relationship between income and amount of
TV news viewing much better than the correlation coefficient
(.4). The true relationship between education level and
amount of TV news viewing, however, is .7 instead of .46.
Therefore, the Beta coefficient of education level is far
off from reflecting the true relationship, while the
correlation coefficient reflects the true relationship.
Thus, until we can decide the relationships among
independent variables (theoretically), neither correlation
coefficients nor Beta coefficients tell us the true
relationships between multiple independent variables and a
dependent variable. For a more detailed discussion about
this issue, see the variance partitioning chapter of
Pedhazur (1982, pp. 174-220).

91
use of these two coefficients. The numbers outside
parentheses are zero—order Pearson's correlation
coefficients. The numbers in parentheses are Beta
coefficients. When either one of these coefficients was

significant, the variable was included in the model.

Broedcast News Viewing in a Mulri-Channei Situation

In a multi-channel situation, besides the major
theoretical factors of the amount of broadcast news viewing
in a broadcast-channel-only situation, news channel
awareness, channel repertoire, and value perception of news
channels are included.

Qerreieripn_nnrrir. Table 10 shows the correlation
matrix among all the variables related to broadcast news
viewing in a multi-channel situation. Broadcast news
viewing is significantly correlated with amount of 24-hour
news viewing (I = .3345, p < .001), education level (r = —
.2883, p_ < .001), CGE-TV (r = .3691, p < .001), RCGE-TV (r
= .2592 p < .01), viewer availability 1 (r = .4962 p <
.001), amount of TV viewing (I = .2457 p < .01), amount of
newspaper reading (I = .1859 p < .05), and age (r = .4407
p < .05). Also, many independent variables are
significantly correlated one another as was the case in the
amount of broadcast news viewing in a broadcast—channel-only

situation. For example, education level and 24 hour news

92

 

Viewer
Availability 1

 

 

 

 

 

 

 

 

.68***(.53***)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.18
Viewer (.25*)
Availability 2
.32**(.09)
CGE—TV TV news
Viewing
.48***(.10)
Amount of
TV Viewing
Age .47**
(.23*)
Gender .17(.21*)
* p < .05
** P < .01
*** P < .001
Figure 5. TV News Viewing in a Broadcast—Channel-Only

Situation

 

913

asuHa Ho

noRoHHnHoHH HERE on «REES H.H-81H 8 So 88... on 2 no: 3855 5 H SHHHHAHHSRH HHS-«HS

9.28.3 HHHHoE. mum 87.2 82.2 93E ~35 H58
sacs scan

HH.uocH amen .HHHH...
mum -.HHHH...-.HoHH
namuac .HHHH... .HHHH. u.oHoH
xnoH-H< .HHHH.. .HHHH.. .OHHH .HHoH...
HHHHHHUHHHHH H .HHHH... .HHHH .oomH .HHHH. .HHHH
HaoHHHuHHHHH H .HHHH .OHHH .OOOH .OHHH .HoHH -.HOHH
HcHHoSAHH -.OHHH .HHHH .HHHH...-.OHHH .OHHH .oHHH .omHH
ononHoHuo .HHHH .HHHH... .OHOH .HOHH. .HHHH... .ooHo -.HHHH -.ooHH
euHao HouooenHos .HHHH .HHHH... .OHHH .HHHH .oHHo -.oooH .HHoo .HHHH
Ha cHacHsm .HHHH.. -.oooH -.OHHH .HHHH .OHHH .HHHH... .HHHH. .oHHH
masseuvou .HHHH. .HHHH -.OHHH .HHHH .HOHH. .HHHH -.HHHH .OOHH
xHaHo scam -.OHHH .OOHH .OHHH -.HHHH .HOHH .OHHH -.OHHH .oHHH
HawouuonmoauH .HHHH .HHHH -.OHHH .oHHH .HHHH. .HHHH -.oaoo .oHHH
HeoaHpa ospeeoH -.oooH .oooo -.OHOH .oHHH .HHHH -.oHHH .HHHH. -.oooH
Hmo .HHOH... .HHHH. -.HHHH.. .HHHH .OHOH .HHHH...-.HHHH...-.HHHH
cannon .HHHH .HOHH -.HHHH.. -.oHHH .HHHH -.oHHH .oHHH -.HHHH
HHEHHH uHuo -.ooHH .HOHH .HHHH .OHHH .oHHo -.HHHo. .HHHH. -.HHHH
Hones» -.HHHH .oHHH .HHHH...-.oHHH .HOHH. -.HHHH. -.HOHH .HHHH

enosHaa you cosmos HpaHHa

HHHo

Hmo -.HHHH..
cosmos -.oHHH .HHOH
HHEHHH «Hue .oHHH -.HHHH.. -.OHHH
Heooso .OHoo .HHHH -.HHHH .HHHH..

so on canon" HHH

H.H-HHoa HHHSHHHosuao . HA .oH. .. HA .oH. ... HA .ooH

Haves- aspen
noHno
.HHHH.

-.OHHo -.oHHH
.HHHH .oHHH
.HHHH -.HHHH
.oHoH .oHHH

-.oHHH .oHHH
.HHHH. -.HHHH
.oHHo -.HHHH
.ooHH .oHHH
.HHHH. a.ooHH

ac

xano

canHaH Haven sacs

.oHoH
u.HHHo
u.oHoH

.oHHH

.HHHu

.ouHH

.oHHH
-.HHHH

.HHHHHH

.HHHH... .HHHH...
a.oHoH -.oHoo

.HHuu... .ooHH

.ooHo .onH
u.ooHH .ooHH

.omuH .ouuH

Hanan-
wonmoaaH

n.0HuH
.HHHH
.HuHH..
.HHHH
.ouHH

94
channel awareness, CGE-TV and RCGE—TV, amount of newspaper
reading and amount of interpersonal communication, amount of
radio news listening and amount of interpersonal
communication all show correlations higher than .40.
ult' re e sion ana sis. As was the case in a

broadcast-channel-only situation, the examination of the
scatterplot and residual plot of each independent variable
by dependent variables did not show any strong evidence of
non-linearity, heteroscadacity and the existence of
significant outliers. Table 11 shows the results of the
multiple regression analysis. The relationship between the
independent variables together and the dependent variable
(R) is .75. The variance of broadcast news viewing
explained by the independent variables (R2) is .56.

From the previous correlation and multiple regression
analyses, the model of broadcast news viewing in a multi-

channel situation can be presented as Figure 6.

-H ur V' i a ulti-Cha S't at'on
To analyze the amount of 24-hour news viewing, the same
set of independent variables contained in the analysis of
the amount of broadcast news viewing in a multi-channel
situation is included, except that the dependent variable

here is 24—hour news viewing.

95

Table 11.
Multiple Regression Analysis Results of the Amount of
Broadcast News Viewing in a Multi—Channel Situation

 

 

 

 

Variable B Beta F Sig F
AGE .040978 .309748 13.231 .0004
GENDER -.020562 -.004696 .004 .9478
EDUCATION -.272237 -.201873 6.523 .0120
INCOME -.016925 -.028722 .153 .6961
N0.0F PEOPLE .197485 .085321 1.457 .2298
AVAILABILITYl .348899 .383919 25.241 .0000
AVAILABILITYZ .255706 .295639 16.592 .0001
RCGE-TV 2.273609 .117639 2.010 .1590
24-HOUR NEWS .128228 .132460 2.830 .0952
CGE-TV .017367 .124735 2.687 .1039
RADIO NEWS -.108381 -.142025 3.830 .0528
INTERPERSONAL .101524 .119878 2.232 .1379
VALUE -.480538 -.104781 1.956 .1646
TV VIEWING -.064745 -.057112 .609 .4368
AWARENESS .113549 .030136 .168 .6823
PREMIUM -.096888 -.021334 .095 .7579
REPERTOIRE -.038753 -.007889 .011 .9159
NEWSPAPER .005328 .006479 .007 .9312
(Constant) -.506720 .117 .7331
Multiple R .74988

R Square .56232

 

 

 

96

 

Education —.29***(—.20*)
Level

 

 

 

 

CGE-TV .37***

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(.12)
RCGE-TV (.12)
. 50***
Viewer (.38***)
Availability 1 --Hrr-~§-‘-‘~§“--~
Broadcast

Viewer .16(.30***) ._____HH———— News
Availability 2 Viewing

 

 

 

 

 

 

 

Amount of . 5**(.06)
TV Viewing

 

 

 

 

24—News .33***
Viewing (.13)

 

 

.19*
Newspaper (.01)
Reading

 

 

 

 

 

Age .44***(.31***)

 

 

 

* P<.05
** P<.01
*** P<.001

Eignre é- Broadcast News Viewing in a Multi—Channel
Situation

97
The correlation matrix among variables in a multi—
channel situation was presented previously (Table 10).
Here, for the convenience of readers, just the correlation
between the amount of 24—hour news viewing and other

variables will be presented in Table 12.

Regreseion Analysis. After examining the scatterplots
and residual plots, regression analysis between independent
variables and the amount of 24-hour news viewing was carried
out. The result is presented in Table 13. Based on these
analyses, a 24—hour news viewing model is illustrated as

Figure 7.

Hyperhesee Teeting

Based on the results presented so far regarding (a)
broadcast news viewing in a broadcast-channel-only
situation, (b) broadcast news viewing in a multi-channel
situation, and (c) 24-hour news viewing in a multi—channel
situation, we can test the hypotheses (H1 to H8) related to
the amount of TV news viewing.

H1: SES will be positively related to the amount of TV

news viewing in a multi-channel situation, while it

will not be related to the amount of TV news viewing
significantly in a broadcast—channel-only situation.

This hypothesis predicted that (a) a significant

positive relationship between SES and the amount TV news

Table 12.

98

The Correlations between Independent Variables and the
Amount of 24—Hour News Viewing

 

 

 

 

 

 

 

 

Variable Correlation Variable Correlation
Broadcast News .3345*** Newspaper .0063
SES -.1048 Radio News .1263
CGE-TV .1758* Interpersonal .0040
RCGE-TV .2591** Premium Channel .1299
Availability 1 .1659 Age .0609
Availability 2 .0411 Gender .2122*
Awareness .0946 No. of People .1033
Repertoire .3396*** Income .1041
Value perception .4127*** TV Viewing .0189

N of cases: 131 * p< .05 ** p< .01 *** p< .001

Table 13.

Multiple Regression Analysis Results of the Amount of 24-
Hour News Viewing in a Multi-Channel Situation

 

 

 

 

 

Variable B Beta F Sig F
AGE .023892 .174828 2.830 .0953
GENDER -.356979 -.078916 .896 .3457
SES -.139894 -.100422 1.127 .2907
INCOME -.057746 -.094866 1.233 .2691
NO. OF PEOPLE .311975 .130479 2.513 .1157
VALUE PERCEPTION 1.648823 .348038 17.917 .0000
RCGE-TV 4.399280 .220352 5.297 .0232
BROADCAST NEWS .187290 .181307 2.830 .0952
REPERTOIRE .912452 .179807 4.414 .0378
AWARENESS .518728 .133272 2.454 .1200
PREMIUM CHANNEL .554342 .118163 2.179 .1426
CGE-TV -.016385 -.113924 1.623 .2053
TV VIEWING AMOUNT -.083878 -.071626 .700 .4044
AVAILABILITY 1 .035647 .037972 .148 .7011
AVAILABILITY 2 .020362 .022790 .063 .8023
INTERPERSONAL .017655 .020181 .045 .8317
NEWSPAPER -.014240 -.016764 .037 .8486
RADIO NEWS .002894 .003671 .002 .9661
(Constant) -2.815284 2.522 .1150
Multiple R .63318

R Square .40091

 

 

99

 

RCGE-TV .26**(.22*)

 

 

 

 

CHANNEL .34***(.13*)

 

 

 

 

REPERTOIRE -——H_______________________—
24-HOUR NEWS

 

 

 

 

"d’”””"”””fifflfr,HH——' Viewing
VALUE

PERCEPTION .41***(.35***)

 

 

 

 

AGE .21*(.17)

 

 

 

 

BROADCAST
NEWS VIEWING .33***(.18)

 

 

 

* P< .05
** P< .01
*** P< .001

Fignre 7. 24-Hour News Viewing in a Multi—Channel Situation

100
viewing will be found in a multi-channel situation, while
(b) no such relationship will be found in a broadcast—
channel—only situation, based on "interest maximization
theory." The relationships between SES and the amount TV
news viewing in three different situations are presented in
Table 14.
Table 14

The Relationships between SES and the Amount of TV News
Viewing

 

Broadcast News Broadcast News 24—Hour News
Broadcast-Only Multi-Channel Multi—Channel

 

Correlation —.1397 —.2883*** -.1048
Beta -.0555 -.2019* -.1004

 

 

 

 

* p < .05, *** p < .001

Throughout three different situations, both correlation
and Beta coefficients are negative. Contrary to the
prediction in H1, SES shows a significant negative
correlation with the amount of broadcast news viewing in a
multi-channel situation. Therefore, H1 is not supported.
The actual results (the overall negative relationships
between SES and the amount of TV news viewing and the
stronger negative relationship in broadcasting news viewing
in a multi-channel situation) strongly suggest that exactly
the opposite direction of changes proposed in H1 might be
thergeee: that is, SES tends to have a negative relationship
with the amount of TV news viewing, and this relationship

becomes stronger (in a negative direction) when interest

101
maximization is possible due to multi—channel development
(this will be discussed in detail in Chapter V.)

H2: In a broadcast-channel—only situation, CGE-TV will

be positively related to the amount of TV news viewing.

H3: In a multi-channel TV situation, RCGE-TV will be

positively related to the amount of TV news

(particularly, 24-hour news).

These two hypotheses predicted that CGE-TV as the
indicator of TV news viewing intention in a broadcast-
channel-only situation (when program choice is limited and
TV viewing in news hours mean almost compulsory TV news
viewing) would be positively related to the amount of TV
news viewing in a broadcast—channel-only situation, while
RCGE—TV as the indicator of TV news viewing intention in a
multi-channel-situation (when program choice is unlimited
and TV viewing in traditional news hours does not
necessarily mean compulsory TV news viewing) would be
positively related to the amount of TV news viewing
(particularly with the amount of 24-hour news viewing).
Table 15 shows the relationships of these two variables with
TV news viewing in three different situations.

Table 15 shows that in a broadcast—channel-only
situation, CGE-TV has a significant positive correlation
with the amount of broadcast news viewing, while RCGE-TV has
little relationship with it. However, in a multi-channel
situation, both CGE—TV and RCGE-TV have positive

relationships with broadcast news viewing.

Table

15

102

The Relationships between CGE—TV and RCGE—TV and the Amount
of TV News Viewing

 

Broadcast News
Broadcast—Only

Broadcast News
Multi-Channel

24-Hour News
Multi-Channel

 

CGE-TV RCGE-TV CGE-TV RCGE-TV

CGE-TV RCGE-TV

 

 

 

 

 

 

 

 

 

 

Corr. .3217** .0237 .3691*** .2592** .1758* .2591**
Beta .0903 .0739 .1247 .1176 -.0716 .2204*
* p < .05, ** p < .01, *** p < .001

Furthermore, with regard to 24-hour news viewing in a

multi-channel situation, RCGE-TV shows a stronger

correlation than CGE-TV.

These results support H2 and H3.

H4: News availability will be positively related to the
amount of broadcast news viewing in both broadcast—
channel-only and multi-channel situation, while it is
not related with the amount of 24-hour news viewing.

This hypothesis predicted that when TV news is

available during limited hours (e.g., traditional news

hours) as is the case in broadcast news viewer availability

(whether a person can watch TV during news hours) will have

a strong relationship with the amount of TV news viewing.

Similarly, if TV news is available anytime, viewer

availability will have little relationship with the amount

of TV news viewing as is the case in 24-hour news. Table 16

shows the relationship between viewer availability and the

amount of TV news viewing in three different situations.

103

Table 16.
The relationship between Viewer Availability and Amount of
TV news Viewing.

 

Broadcast News Broadcast News 24-Hour News
Broadcast-Only Multi-Channel Multi—Channel

 

Avail 1 Availz Avail 1 Avail 2 Avail 1 Avail 2

 

Corr. .6844*** .1824 .4962*** .1642 .1659 .0411
Beta .5343*** .2467** .3839*** .2956*** -.0168 .0202

 

 

 

 

 

 

 

 

 

** p <.01, *** p < .001

Table 16 shows that in regard to broadcast news
viewing, both availability 1 (number of days in a week a
viewer watches TV during 6-7 PM) and availability 2 (number
of days in a week a viewer watches TV during 11-12 PM) show
either significant correlation or Beta coefficients.
Especially, viewer availability 1 (viewer availability in
the primary news hour) is very strongly related to the
amount of broadcast news viewing. However, with regard to
24-hour news, neither availability 1 nor availability 2 show
significant correlation or Beta coefficients. These results
confirm H4.

H5: News programming awareness will be positively

related to the amount of 24-hour news viewing.

This hypothesis proposed that in a multi-channel
situation, where there are so many different channels
available, 24—hour news programming will be watched more
when the offerings of these channels are well recognized by

viewers and vise-versa. The correlation and Beta

104
coefficients between programming awareness and 24-hour news
viewing are .0946 and .1333, respectively. Even though the
directions of these coefficients show that there are
positive relationships between these two variables, neither
of these two coefficients are significant in p < .05 level.
Therefore, the results do not support H5.

H6: The inclusion of 24-hour news channels in channel

repertoire will be positively related to the amount of

24-hour news viewing.

This hypothesis proposed that in a multi-channel
situation, since viewers tend to turn to only a limited set
of channels intentionally (channel repertoire), whether 24-
hour news channels belong to these intentionally watched
channels will affect the amount of 24—hour news viewing.
The correlation coefficient and Beta coefficient between
channel repertoire and the amount of 24-hour news viewing
are r =.3396, p < .001, and Here =.1798, p < .05,
respectively. Therefore, the results support H6.

H7: The value perception of news/information channels

(the willingness to pay extra money for these channels)

will be positively related to the amount of 24-hour

news viewing.

This hypothesis proposed that in a multi-channel
situation where viewers pay money (subscription fee) to
achieve a various set of extra values, those who perceive
the value of receiving additional news channels through

multi-channel subscription highly (those who willing to pay

105
extra money for these channels) will watch them more than
those who do not (those who put their values somewhere
else). The correlation coefficient and Beta coefficient
between value perception of 24-hour news channels and the
amount of 24-hour news viewing are r =.4127, p <.001, and
Here = .3480, p <.05, respectively. Therefore, the results

support H7.

TV Newe Viewing and News Learning

The second objective of this study is to investigate
how news learning from TV news viewing differs in broadcast-
channel-only and multi-channel situations. Besides TV news
viewing, SES and news learning motivationwere included as
other major theoretical factors of news learning. In
addition to these variables, news learning (knowledge) is
necessarily contingent upon other factors, such as the
amount of other media exposure (newspaper reading, radio
news listening, interpersonal communication) and other
demographic factors (age, gender, and household income), and
the date of telephone interview. These variables were also

included in the analysis.

n' L ur Ne s e
Among the four news events (the Flooding, War in
Bosnia, G-7 Summit Meeting, and Gays in Military), the event

Flooding showed the highest awareness (91 percent), while

106
the G-7 Summit Meeting marked the lowest awareness (49
percent). In average, the overal awareness of the four news
events (291 cases) was 71.2 percent (or three out of four
events)

Table 17
News Learning Level of 4 News Events

 

 

 

 

 

 

 

 

cases awareness (%)
Flooding 302 91.1
War in Bosnia 295 67.5
G-7 Summit 300 49.3
Gays in Military 300 75.0
Total 291 71.2
s e 'n o d st- - ‘ ituati n

In order to determine how the level of news learning is
related to the independent variables and how independent
variables are related one another, all possible sets of
correlations among all the variables were performed. Table

18 shows the resulting correlation matrix.

According to this correlation matrix, level of news
knowledge is significantly correlated with the three major
theoretical factors, the amount of broadcast news viewing, r
= .3268, p < .01, SE3, r = .3243, p < .01, and news

learning motivation r = .3611, p < .01. The other variables

1(37

Hugo 3.

0943.38 Ewan on 35.38 wanna 8 9. “Soy on an: 5% 5 . Roao-mnAwBEHbfiw $355.

2 a2»

mum
zofificg
mummy—won
5&0 Sam
Hawogmosb
so

$38.

583

ouwo

m on 088"

55.3 .2 so: mum

.unam..

.wwhu..
.onH..
.owso
.~.uu
.opog
.wwwu
u.~uow
.Huoo
u.o»uo

on

u.oomo
.HHHo .uuHN
.HHuq .ouuu
.oawp -.onmu
.ONQN u.on~
.mmma... -.ouuu
.omgu u.Hmmo
.oun¢ .H.-
.Hbmu .oHNQ

son»-
3.38

.ooum
.Huow
.Nwoa.
.quu
u.o~ug
.Hsoo
u.ooco

:2?
a?

.NuNQ.
..qu...
.NHom
n.0qq.
.moo»..
-.HUH.

396 Has? So
so: 3383.
.m~Hu...
.23 :23
.OHuo .oquu -.ouoo
.38: .33: .33
:88 :58 2:2.

~18an mwmswmwogoo . R .8. : R .8. .3 X .2:

cannon

IoNHWO
u.ouwo

H828

-.oqum

108
do not show significant correlations with the level of news
learning. The correlation matrix also shows a complex
pattern of relationships among independent variables (age
and the amount of TV news viewing, news learning motivation
and interpersonal communication, newspaper reading and radio
listening, newspaper reading and interpersonal
communication, newspaper reading and household income, and
interpersonal communication and household income).

In order to determine the relationship between all the
independent variables together and the level of news
learning, a multiple regression analysis was carried out.
The examination of the scatterplot and residual plot of each
independent variable by the level of news learning did not
show any strong evidence of non-linearity, heteroscadacity
and the existence of significant outliers. Table 19 shows

the results of the multiple regression analysis.

---—--—---_---¢------

The multiple correlation coefficient (R) is .58, which
is the correlation between the nine independent variables
together and the level of news learning. The R square (R?)
is .34, which means 34 percent of the whole variance in the
level of news knowledge is explained by the independent

variables included in the multiple regression equation.

109

 

 

 

Table 19.

Multiple Regression Analysis Results of the Level of News

Learning in a Broadcast-Channel—Only Situation
Variable B Beta F Sig F
MOTIVATION 2.589534 .248231 4.588 .0362
BROADCAST NEWS 3.587931 .329021 7.792 .0070
SES 4.390486 .278585 6.224 .0153
GENDER -6.376352 -.120936 1.173 .2831
RADIO NEWS 1.402719 .159234 1.404 .2407
INTERPERSONAL -l.162460 -.111566 .421 .5188
INTERVIEW DATE -.002984 -.075866 .477 .4923
NEWSPAPER -.577640 -.061058 .233 .6314
AGE .089208 .049180 .153 .6971
INCOME .204290 .029155 .049 .8250
(Constant) 248.106187 .481 .4904
Multiple R .58438
R Square .34150

 

 

 

 

110

As was discussed before, neither correlation
coefficients nor Beta coefficients are sufficiently valid
indicators of the relative strength of relationships between
independent variables and the dependent variable when
independent variables are substantially correlated.
Therefore, here again, both correlation and Beta
coefficients will be used as the indicators of the
relationships together.

Figure 8 shows the model of news learning in a
broadcast-channel—only situation. The numbers outside
parentheses are zero-order Pearson’s correlation
coefficients. The numbers in parentheses are Beta
coefficients. When either one of these coefficients was

significant, the variable was included in the model.

News Learning in a Multi-Chappel Situation

In order to determine how the level of news knowledge
is related to the independent variables and how independent
variables are related to one another in a multi-channel
situation, all possible sets of correlations among all the
variables were performed. Table 20 shows the resulting

correlation matrix.

111

 

.33**(.33**)

 

 

 

 

 

 

TV News

.32**(.28*) Level of
Education _fi News
Level Learning

 

 

 

 

 

 

 

News Learning

 

 

 

Motivation .36**(.25*)
* P< .05
** P< .01

*** P< .001

Eigppe . News Learning in a Broadcast—Channel-Only
Situation

112

33o no.

85.8.88 .51.. 08 43.58. nor—non 8 as 5.3 on an... 388. S . 5.8.8.9589 «5.88..

8.888.. 8.8 :85

“:68. 8am
mmm
203538
5828“.
agape amen
5838.855
ymo

$88.

588

8.8582 88 :5:

m cm 883

So

8583 9.0889,. 2.38.
8.8 8.8
.82: .38.:
.53.: 138:. .89..
.88.: .38 .58.
.88. .88: .28
.52. :88 38.8
.88: .28 .88
.38: .83... .88..
:88. .28 ...:S
.88... :88 :8:
.88 28.8

mum

.nuqu...
.o.oo
.Huop.
.ooao
u.~Ha~..
n.8uou.
.uu~o...
.o~.H

son»-
:38

.~.m~..
.nu.o..
.~ouu...
.oooo
.ooqp
.poum.
.ouao

:2?
E...

.Nqu...
..uaa...
.Nou....
.ouom
.ywuu
.oo.u

”......an manoeso . R .8. : R .8. ... v.” .28

8&0 H587 3o
.58 3288
.83.:
.89.. .88.
:88 .88: .88.
.8: .88 :82
.88 .83 .88

338.

l-HNHO
.23

Hanoao

£85

113

According to this correlation matrix, level of news
learning is significantly correlated with the amount of 24—
hour news viewing ; = .2254, p <.01, SES, ; = .4210, p
<.001, and news learning motivation, r = .2661, p <.001.
The level of news learning is also significantly correlated
with other media uses (newspaper reading, radio news
listening, and interpersonal communication), age, gender,
and household income. The correlation matrix also shows a
complex pattern of relationships among independent
variables. For example, age and broadcast news viewing,
education level and household income, newspaper reading and
interpersonal communication, and radio news listening and
interpersonal communication show strong correlations.

Table 21 shows the results of the multiple regression
analysis. The multiple correlation coefficient (R) is .65,
which means the correlation between the ten independent
variables together and the level of news learning. The R
square (R?) is .43, which means 43 percent of the whole
variance in the level of news learning is explained by the
independent variables included in the multiple regression

equation.

Based on these analyses, Figure 9 illustrates the
relationship between independent variables and the level of

news knowledge in a multi-channel situation,

114

Table 21
Multiple Regression Analysis Results of the Level of
News Learning in a Multi-Channel Situation

 

 

 

 

Variable B Beta F Sig F
SES 5.924624 .378489 27.590 .0000
AGE .554726 .370029 26.097 .0000
GENDER -11.124067 -.216670 10.766 .0013
INTERPERSONAL 2.124564 .218225 8.202 .0048
INTERVIEW DATE -.005603 —.153796 6.192 .0139
INCOME .776382 .112857 2.818 .0952
24-HOUR NEWS 1.112816 .109768 2.760 .0987
BROADCAST NEWS —1.185273 -.104665 2.046 .1546
MOTIVATION .999890 .099070 2.050 .1542
NEWSPAPER -.550534 -.059655 .683 .4099
RADIO NEWS —.285712 -.032184 .210 .6478
(Constant) 472.689562 6.555 .0114
Multiple R .65380

R Square .42745

 

 

 

115

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

24-hour news .23**
Viewing (.11)
Education
Level .42***
(,33***)
News Learning .27***
Motivation (.10)
News Paper .17*(-.06)
Reading Level of
If News
Learning
Radio
Listening .16*(—.03)
Interpersonal .22**
Communication (.22**) .23**
(.37***) -.20*
Age (-.22**) .29***
Gender (.11)
Income
* P< .05
** P< .01

*** P< .001

Eigppg_g. News Learning in a Multi—Channel Situation

116
Hypotheses Testing
Based on the results presented so far regarding (a)
news learning in a broadcast-channel-only situation, (b) and
news learning in a multi—channel situation, we can test the
hypotheses (H8 to H14) related to news learning and the role

of TV.
H8: SES will be positively related to news learning.

This hypothesis predicted that a significant positive
relationship between SES and news knowledge exists both in a
broadcast-channel-only and multi-channel situations since
cognitive skill and prior knowledge tend to facilitate the
news learning of high SES people. The relationships between
SES and the level of news learning in two different
situations are presented in Table 22.

Table 22.
The Relationships between SES and the Level of News Learning

 

 

Broadcast-Only Multi-Channel
Correlation .3243** .4210***
Beta .2787* .3785***

 

 

 

 

 

* p<. 05 ** p< .01 *** p < .001

Table 22 shows that in broadcast-channel-only and
multi-channel situations, both correlation and Beta
coefficients are significant which is strong evidence of the
substantial relationship between SES and the level of news

learning. Therefore, the results support H8.

117

H9: News learning motivation will be positively related

to news learning.

This hypothesis proposed that news learning motivation
will have a positive relationship with the level of news
learning based on uses and gratification studies. The
relationships between news learning motivation and the
amount TV news viewing in the two situations are presented
in Table 23.

Table 23.

The Relationships between News Learning Motivation and the
Level of News Learning

 

 

 

 

 

Broadcast-Only Multi-Channel
Correlation .3611** .2661***
Beta .2482* .0991

 

 

* p< .05 ** p< .01 *** p< .001

Table 23 shows that news learning motivation and the
level of news learning is significantly correlated in both
situations, even though Beta coefficients (the unique
contribution of news learning motivation to news learning)
are insignificant in a multi—channel situation (that means
news learning motivation plays its role by interplaying with
other variables in this situation). Therefore, the results

confirm H9.

H10: Multi-channel TV subscribers will perceive TV news
as providing more variety in news stories than non-
subscribers.

118

H11: Multi-channel TV subscribers will perceive TV news

as providing more in-depth news stories than non-

subscribers.

H12: Multi-channel TV subscribers' evaluation of TV in

gratifying news learning motivations (content seeking)

will be higher than non-subscribers.

Hypotheses 10, 11, and 12 proposed that TV viewers in a
multi-channel situation will perceive TV as a better news
medium than TV viewers in a broadcast-channel-only situation
will. To test these sets of hypotheses, the evaluation of
TV in providing various news stories, the evaluation of TV
in providing in-depth news stories, and the evaluation of TV
in gratifying one’s news learning motivation between
broadcast-channel-only TV viewers and multi-channel TV
viewers were compared. Table 24 shows the t-test results of
the differences between them.

Table 24.
The Comparison of Broadcast-Channel-Only Viewers and

Multi-channel TV Viewers Regarding the Evaluations of TV as
a News Medium.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cases Mean SD t DF Sign
Broadcast 93 63.73 25.46
Variety 2.85 297 .005
Multi-channel 206 71.50 19.99
Broadcast 95 59.37 26.51
In-Depth 3.64 299 .000
Multi-channel 206 69.81 21.40
Motivation Broadcast 91 3.37 1.00
Gratifi- 5.31 283 .000
cation Multi-channel 194 3.93 .72

 

 

119

According to Table 24, in all three aspects of
evaluations broadcast-channel-only TV viewers and multi-
channel TV viewers show substantial differences. First,
with regard to variety evaluation (measured by 0 - 100
evaluation scale), multi-channel subscribers give about 8
points higher evaluation to TV. Secondly, with regard to
in—depth evaluation (measured by 0 - 100 evaluation scale),
multi-channel subscribers show more than 10 points higher
evaluation of TV. Finally, with regard to the evaluation of
the news learning motivation gratification (measured by 5
point scale), multi-channel subscribers show an average
response close to "good (4)" while broadcast-channel-only
viewers, close to "neutral (3).” Therefore, the results
support these three hypotheses.

H13: The amount of TV news viewing will be related to

news learning more strongly in a multi-channel

situation than in a broadcast-channel-only situation.

H14: In a multi-channel situation, the amount of 24-

hour news viewing will be related to news learning more

strongly than the amount of broadcast news viewing.

H13 proposed that since multi-channel TV tends to
provide more variety in news stories and indepth news
coverage as well as allowing a more flexible program choice
(which will reduce the ceiling effect), the amount of TV
news viewing will have a stronger relationship with the

level of news learning in a multi-channel situation than it

will in a broadcast-channel-only situation. In the same

120
vein, H14 proposes that since 24-hour news provides more
variety in news stories and indepth news coverage than
broadcast news, the amount of 24—news viewing will have a
stronger relationship with news learning than the amount of
broadcast news viewing.

In a multi-channel situation, the amount of TV news
viewing was computed by adding the amount of broadcast news
viewing and 24-news viewing together. Then, the correlation
and Beta coefficient between this total amount of TV news
viewing and news learning were produced. Table 25 shows the
correlation and Beta coefficients between (a) the news
knowledge and broadcast news viewing in a broadcast-channel-
only situation, (b) the news knowledge and broadcast news
viewing in a multi-channel situation, (c) the news knowledge
and 24—hour news viewing, and (d) the news knowledge and
total TV news viewing in a multi-channel situation.

Table 25.

The Relationship between TV News Viewing and the Level of
News Learning

 

Broadcastl Broadcastz 24-News Total News

 

Correlation .3268** .0115 .2254** .1420
Beta .3290** -.1047 .1098 .0255

 

 

 

 

 

 

 

Broadcastl: broadcast-channel-only situation
Broadcastz: multi-channel situation

Total news: broadcast news + 24-hour news

* p< .05 ** p< .01 *** p< .001

According to this table, TV news viewing in a

broadcast-channel-only situation is highly related to the

121

level of news learning. On the contrary, (total) TV news
viewing in a multi-channel situation is not significantly
related to the level of news learning. Therefore, H13 is
not supported from the results. In a multi-channel
situation, when we compare the relationships between
broadcast news viewing and the level of news learning and
and the relationship between 24-hour news viewing and the
level of news learning, the latter shows significant
correlation while the former does not. These results

support H14.

CHAPTER V

DISCUSSION AND CONCLUSION

W

In Chapter I, two research questions were presented
which guided the direction of this study. The answers to
and the results of the empirical evidence associated with
these questions have been presented in some detail in
Chapter IV. This chapter will summarize the main findings
and discuss their implications. A second purpose of this
chapter is to note some of the contributions and limitations
of this research. Finally, some suggestions for future

research are presented.

I ! ! Ii E K E' i'

The research questions posed in Chapter I asked how the
development of multi-channel TV changes TV news viewing and
news learning from it. In order to answer these questions,
the major differences in TV news viewing and news learning
from it between a broadcast-channel-only and a multi-channel
situation were analyzed through the cross-sectional
comparison of these two situations.

Based on past mass media and relevant psychology

literature, this study proposed fourteen hypotheses

122

123
regarding the relationships between various theoretical
factors and two major dependent variables (TV news viewing
and news learning). Hypotheses 1 to 7 were relevant to the
changes in TV news viewing, and hypotheses 8 to 14 were
relevant to the changes in news learning from TV news
viewing. As was presented in Chapter IV, some of these
hypotheses were supported and some were not. In this
section, the implications of these hypotheses test results

will be discussed.

E J l E H J!I-:I 1 IE 3 IX H M' .

SES_and_T!_news_yiewing. Hypothesis 1 predicted that
socio-economic status (SES) will become a significant
predictor of the amount of TV news viewing due to the
development of multi-channel TV. This hypothesis was
proposed based on past knowledge gap studies, based
particularly on interest maximization theory (Jeffres,
1978a). According to this theory, it is assumed that (a)
high SES people have stronger news interest/selectivity than
low SES people and (b) these differences in news interest
will result in different amounts of TV news viewing between
these two groups when the development of multi-channel TV
facilitates program choice based on individualized program
type interest (interest maximization).

From this theory, it was expected that SES would show a
substantial positive relationship with the amount of TV news

viewing in a multi-channel situation. This hypothesis,

124
however, was not supported. Actually, the results -- (a)
the negative correlation and Beta coefficients throughout
three different situations and (b) the stronger
(significant) negative relationship found in a multi-channel
situation -- strongly suggest that the relationship between
SES and TV news viewing (especially broadcast news viewing)
is opposite from what was proposed in this study. The
results also indicate that the relationship between SES and
TV news viewing tends to become stronger due to the
development of multi-channel TV.

Therefore, this finding leads us to the following re-
interpretations of the results regarding interest
maximization theory, the relationship between SES and TV
news viewing, and the changes in the relationship between
SES and the amount of TV news viewing due to the development
of multi-channel TV: (a) the development of multi-channel TV
facilitates interest maximization (or program choice based
on individualized program type interest), (b) SES tends to
have a latent negative relationship with TV news viewing,
and (c) the development of multi-channel TV realizes this
latent negative relationship between SES and TV news viewing
to an actual relationship.

Even though the prediction has been wrong with regard
to the direction of the relationship, this study did find
that the development of multi-channel TV strengthens (or

realizes) the existing (or latent) relationship between SES

125
and TV news viewing. This supports the interest
maximization theory. Another support for the interest
maximization theory comes from Youn’s study (1993a) which
investigated the relationship between program type
preferences and program choices in a broadcasting-channel-
only and a multi-channel situation. According to this
study, when program type preferences were elaborately
measured and other moderating variables were carefully
controlled, multi-channel TV viewers gratified their program
type preferences twice as well as broadcast-channel-only TV
viewers.

The support for the second interpretation is widely
available from past mass media audience studies (Becker &
Whitney, 1980; Brinton 8 McKown, 1961: Browne, 1978; MCClure
& Patterson, 1976; Robinson, 1974, 1975). These studies
consistently found that high SES people tend to depend on
printed media (newspaper) as their primary news source,
while low SES people, on TV news.

In this study, too, when the relationship between SES
and the primary news source was analyzed, the same result
was found. Table 27 shows that high SES people tend to
consider the newspaper as their primary news source, while
low SES people, TV, th;§gnere = 19.46, DE = 4, p < .001.
Therefore, high SES people may perceive TV not so much as a

news medium as they may perceive the newspaper as a news

Table 26.

126

Primary News Sources among Different SES Groups

 

 

 

 

 

 

 

 

 

 

SE81 SE82 SE83 SE84 SE85 total
Newspaper 15 44 53 18 10 140
(10.7) (31.4) (37.9) (12.9) (7.1) (100.0)
TV 1 14 38 16 16 85
(1.2) (16.5) (44.7) (18.8) (18.8)

 

Chi-Square 19.46041

DF = 4 Significance = .00064

SE81: grade school only or did not finish high school
SE82: high school graduate
SE83: vocational or technical school or some college

SE84:
SE85:

college graduate
post graduate work or graduate degree

 

127
medium. Put differently, for them TV may be considered
primarily as an entertainment medium.

If this is the case, then, when TV provides far
expanded options of program choices in both news and
entertainment, the direction of interest maximization for
high SES people is to turn to more non-news programs and
less news programs. 0n the contrary, low SES people, who
consider TV news as relatively more an important news
source, will consume the same (or even an increased) amount
of TV news programs compared to high SES groups when multi-
channel TV provides far expanded options of program choices
in both news and entertainment.

W. Hypothesis 2 and 3
predicted that in a broadcast-channel-only situation,
content gratification expectancy of TV (CGE-TV) will be a
significant predictor of the amount of TV news viewing,
while in a multi-channel situation, relative content
gratification expectancy of TV (RCGE-TV) will be an
important predictor. These hypotheses were proposed based
on past uses and gratification studies, particularly based
on the distinction between content seeking (motivations) and
process seeking (motivations), and the expectancy-value
theory. The results supported these two hypotheses.

These results have two important implications. First,
the theoretical discussion on CGE-TV and RCGE-TV turned out

to be valid. CGE-TV and RCGE-TV are original theoretical

.128
concepts developed by this research. Even though a few
researchers (Babrow & Swanson, 1988; Galloway & Meek, 1981;
Palmgreen & Rayburn, 1982 & 1983) already applied the
expectancy value theory to TV news viewing, one of the most
important theoretical tasks for a meaningful application of
this theory was generally ignored by them -- the
investigation of values. Without knowing what values are
and what values are not relevant to an object (in this
study, TV news viewing), the expectancy value theory has
little meaning in practical sense. By clarifying these
values based on the theoretical distinction between content
seeking and process seeking in uses and gratification
studies, this study could conceptualize CGE-TV as an
indicator of TV news viewing intention when program choice
is limited and TV viewing in news hours is almost
compulsory, and RCGE-TV as an indicator of TV news viewing
intention when program choice is unlimited and TV viewing in
traditional news hours is not necessarily compulsory. The
fact that these two variables show the predicted
relationship patterns with the amount of TV news viewing is
good evidence to support their theoretical validity.

The second important implication of the results is
found from the different relationships which CGE-TV and
RCGE—TV have with broadcast news and 24-hour news viewing.
The fact that CGE-TV is substantially related to broadcast

news viewing means that in broadcast news viewing, even

129

though content seeking is the primary factor, process
seeking can also be involved with TV news viewing. On the
contrary, the fact that RCGE-TV is substantially related to
24-hour news viewing means that 24-hour news viewing happens
when a viewer pursues content seeking and only content
seeking (or little process seeking). If a viewer pursues
process seeking as well as content seeking, it can lead
him/her to different programs. Put differently, TV news
viewing for the purpose of entertaining or escaping is less
likely to happen in a multi-channel situation. In a more
general sense, this is interpreted as evidence of
motivational segmentation in TV viewing which is expected to
become more prevalent as the channel segmentation due to the
development of multi-channel TV goes on.

Relee_ef_epnez_fiee§er§. Hypothesis 4 predicted that
viewer availability will become a less significant predictor
of the amount of TV news viewing as the development of
multi-channel TV goes on. This hypothesis was supported
from the results. While viewer availability had a prevalent
relationship with the amount of TV news viewing in a
broadcast-channel-only situation, it showed a weaker
relationship with the amount of broadcast TV news viewing in
a multi-channel situation and no significant relationship
with the amount of 24-hour news viewing.

These results clearly indicate that the role of

structural constraining factors, such as the amount of

130
programming and programming schedule, etc. in program choice
and viewing, will become weaker as the development of multi-
channel TV goes on. With regard to TV news viewing, these
results have two implications: (a) TV news viewing will not
be confined to a certain time, and (b) incidental or
compulsory TV news viewing in the course of TV viewing
during traditional TV news hours will happen less often.
Therefore, as the development of multi-channel TV goes on,
the universal TV news viewing phenomenon of a limited number
of TV news programs at a certain time is expected to become
weaker, if not completely disappear.

Hypothesis 5 predicted that in a multi-channel
situation, news channel awareness will be substantially
related to 24—hour news viewing. This hypothesis,
however, was not supported. These results indicate that
news channel awareness is not substantially related to the
amount of TV news viewing. Several explanations can be
given regarding these results. The first explanation is the
problem in measurement. In this study, news channel
awareness was measured by asking people whether they
recognized the programming content of two specialized 24-
hour news channels, C-SPAN and C-NBC. We cannot exclude the
possibility that some people could not answer the questions
well because they simply did not recognize the exact channel
names, even though they had good awareness of the actual

content of these channels. Also, the awareness of C-NBC

131
channel content may not be a valid measure of news channel
awareness. C-NBC may be too specialized a business channel
to mean news to many people. The second explanation is that
news channel awareness as a type of knowledge is affected by
many other factors as well as by the amount of TV news
viewing. For example, a significant correlation was found
between SES and news channel awareness, r =.4515, p <001.
Finally, awareness might not be the sufficient condition for
TV news viewing, even if it is a necessary condition. In
other words, even though news channel awareness is required
for the viewing of these specialized news channels, this
awareness may not necessarily lead to the viewing of these
channels.

Hypothesis 6 predicted that in a multi-channel
situation, the inclusion of news channels in the channel
repertoire will be substantially related to 24-hour news
viewing. This hypothesis was supported by the results.
These results confirm the notion that as the number of
available channels increases, channel repertoire as a form
of predisposition toward channels will play a more important
role (Heeter & Greenberg, 1988). Hypothesis 7 predicted
that in a multi-channel situation, the (monetary) value

perception of news channels will be substantially related to

132
TV news viewing. This hypothesis also achieved significant

support from the results.1

D J ! E H 1!’-:l 1 TM 3 H I .
Ty_neu§_yiewing. This study proposed that, since news
learning from TV news viewing is not likely to be affected
(reduced) by ceiling effect so much as is the case in a
broadcast-channel-only situation, this will bring changes in
viewers' perceptions of TV as a news medium and the actual
news learning of TV news viewing. In hypotheses 10, 11, and
12, it was predicted that multi-channel TV viewers would
perceive TV as a better news medium than broadcast-channel—
only TV viewers in terms of news variety (H10), depth of
coverage (H11) and the expectancy of news learning
motivation gratification.(2mh) (H12). Hypothesis 13
predicted that TV news viewing will show a stronger
correlation with the level of news learning in a multi-
channel situation. In the same vein, hypothesis 14
predicted that 24-hour news viewing will show a stronger
relationship with the level of news knowledge since 24-hour

news will be less affected by the ceiling effect.

 

1Here, one may question the theoretical meaningfulness
or uniqueness of value perception. However, the examination
of the correlations between value perception and other
independent variables reveals that value perception is
significantly correlated with only one variable, the
inclusion of 24-hour news channels in channel repertoire, ;
=2367, p < .05. This indicates the uniqueness of value
perception as a variable.

133

The hypothesis testing results turned out to be
somewhat inconsistent. The first three hypotheses (H10, 11,
and 12) regarding the changes in viewers' perception of TV
as a news medium, and the last hypothesis (H14) regarding
the relative roles of broadcast news and 24-hour news
viewing in news learning in a multi-channel situation were
supported from the results. However, the hypothesis
regarding the differences of the relationships between TV
news viewing and news learning between a broadcast-channel-
only and multi-channel situations (H13) was not supported.
Contrary to the expectation, the results turned out to be
just the opposite from what was proposed by H13: TV news
viewing was significantly correlated to the level of news
learning in a broadcast-channel-only situation, while it was
not in a multi-channel situation.

These results indicate that, even though TV is not as
good a news medium in a broadcast-channel only situation as
it is in a multi-channel situation and viewers perceive this
fact well, the relative importance of TV as a news medium is
still greater among broadcast-channel-only TV viewers than
multi-channel TV viewers. Then we need to ask why this is
the case.

One possible answer can be found from the measurement
of the amount of total news viewing in a 24-hour situation.
In this study, the amount of total news viewing was computed

by adding (a) the days in a week to watch broadcast news and

134
(b) the days in a week to watch 24-hour news. This
measurement of total news viewing, however, is prone to
error since we do not know whether (a) and (b) are identical
in terms of measurement scale.

If we assume that this measurement was valid, one
plausible answer for this question can be suggested from the
differences in the media environment between broadcast-
channel-only TV viewers and multi-channel TV viewers. Put
differently, broadcast-channel-only TV viewers may live in a
relatively limited media environment compared to multi-
channel subscribers. Therefore, for broadcast-channel-only
TV viewers, the easily and widely available TV news may be
the dominant (if not the only) news source of important
social events. In this case, the amount of TV news viewing
will be directly related to the level of news learning. 0n
the contrary, multi-channel subscribers who live in
relatively unlimited media environment may have a variety of
news sources about important social events compared to
broadcast-channel-only TV viewers. Therefore those who do
not watch TV news still can get informed from other news
sources.2 In this case, the amount of TV news viewing will
not show a very strong relationship with the level of news

learning.

 

2Actually, positive correlations were found between the
amount of other media exposure (newspaper reading, radio
news listening, and interpersonal communication) and the
level of news learning in a multi-channel situation.

135

Past studies which investigated the overall media uses
of multi-channel TV subscribers and non-subscribers (Becker
et.al., 1983; Collins et.al., 1983: LaRose & Atkin, 1988:
Rothe et.al., 1983), however, do not provide consistent
findings regarding the differences in the media environments
between these two groups, even though some of these studies
(LaRose & Atkin, 1988: Rothe et.al., 1983) suggest that
multi-channel TV subscribers tend to use other media more.
Thus, until we can accumulate more evidence, the status of
the explanation presented above will remain tentative.

Boles_gf_gthez_fiagtgzs. Besides TV news viewing, this
study examined the relationships between such factors as
SES, news viewing motivations, other media uses, and the
level of news learning. Hypothesis 8 predicted that SES
will have a positive relationship with news learning due to
the differences in cognitive skill and prior knowledge
(internal representation) of different SES groups. The
results supported this hypothesis. The results also
supported hypothesis 9 which predicted a positive
relationship between news learning motivation (content
seeking) and news learning.

Even though this study did not hypothesize the
relationship specifically, it was expected that the amount
of other media exposure would be positively related to news
learning. Especially, as was consistently found from past

studies (Atkin et. al., 1976; Becker & Whitney, 1980:

136

Becker, et. al, 1978; Israel & Robinson, 1972: Katz et. al,
1977; McClure & Patterson, 1976; Miller & MacKuen, 1979;
Nordenstreng, 1972: Robinson, 1972 and 1974: Robinson &
Levy, 1986), newspaper reading was expected to have a
stronger relationship with the level of news learning than
TV news viewing. However, the relationships between other
media exposure (particularly newspaper reading) and news
learning turned out to be not as strong as were expected.
In a broadcast-channel-only situation, none of the other
media factors showed significant correlations with news
learning (see Table 20). In a multi-channel situation, even
though significant correlations were found, the sizes of
correlations were still relatively small (see Table 22).

One explanation for this small effect of other media
factors can be provided from the measurement of news
learning used in this study: the awareness of six major news
events. The actual questions used to tap in the awareness
of these events focused on the very basic aspects (partly
due to the limitation of the telephone interview) rather
than the relatively complicated aspects of these news
events. In this case, this measurement might not have
discerned the differences in the news learning level of the
more sophisticated aspects of these events to which other
media exposure, especially newspaper reading, is likely to
contribute. Put differently, if the news awareness

questions asked more detailed aspects of these events, than

137
newspaper reading would have shown a stronger relationship
with the level of news learning. Another explanation is
that newspaper reading may not be for news but for other
information such as shopping information or features, while
TV news viewing is mostly for news. In other words, more
selectivity exists in newspaper reading through which
readers can avoid news. This may be why newspaper reading
showed a weaker relationship with the level of news learning

than TV news viewing.

MW):

In answering the questions of how the development of
multi—channel TV affects TV news viewing and news learning
from it, this study makes several contributions.

First, the factors identified by this study turned out
to be highly relevant and explanatory in investigating TV
news viewing and news learning. For example, with regard to
TV news viewing, the factors included in the analyses could
explain about 40 to 70 percent of the whole variance. In
case of news learning, 34 to 42 percent of the whole
variance was explained.

Secondly, the investigation of the relationships
between these factors and TV news viewing and news learning
throughout this study provides important implications on how
TV news viewing and news learning from it will change as the

development of multi-channel TV goes on. For example, this

 

138

study clearly shows that in a multi-channel situation the
impact of programming or availability factors, which
override individual difference factors and lead to universal
TV viewing phenomenon, will become less important. On the
contrary, this study found some evidence that such trends as
the maximization of personalized program type interest
(interest maximization) and the high degree of motivational
segmentation (RCGE-TV) will become more prevalent in TV
viewing including TV news viewing. These findings generally
indicate that TV news viewing will become more an individual
phenomenon rather than a social phenomenon. They also
indicate that the role of TV news as the common base of news
learning will become seriously weakened (if not completely
disappear) as the development of multi-channel TV goes on.

Thirdly, the theoretical elaboration with regard to
expectancy value theory and knowledge gap hypothesis is an
important contribution of this study. More specifically
with regard to the expectancy value theory, this study could
clarify what constitutes the major value components in the
expectancy value equation when this theory is applied to TV
news viewing. With regard to the knowledge gap hypothesis,
the discussion about the cognitive skill and prior knowledge
(internal representation) based on cognitive psychology
literature provides an important theoretical bridge between
SES and news learning which has been generally missing or

ignored from past knowledge gap studies. This is an

139
important step toward the knowledge gap theory from the
knowledge gap hypothesis.

Finally, with regard to news learning from TV news
viewing, this study provides an important implication. By
re-interpreting the findings from past studies, this study
maintains that news learning from TV news viewing is not as
limited as past studies made us believe. Actually, this
study found substantial evidence that, in terms of the
baseline awareness of major news events, TV news viewing is
as important as, or more important than, other media

exposure.

5 !' E E ! S! I'
This study has several limitations. First, since the
data gathering for this study was carried out in
comparatively limited local areas covered by one multi-
channel TV system, it is unclear to what degree the results
can be generalized. Secondly, as was discussed previously
in this research, the replacement of repeated group analysis
(comparison of the same people before and after multi-
channel subscription) by cross-sectional analysis
(comparison of the multi-channel subscribers and non-
subscribers) was inevitable, but it still limited the
analyses in many aspects, especially in analyzing news
learning from TV news viewing. Thirdly, some of the factors

used in this study need more theoretical elaboration. For

140
example, with regard to channel repertoire and value
perception, one common question arises; what factors would
decide the inclusion of news channels in channel repertoire
or value perception of news channels initially? Without
answering this question, the meaningful interpretations of
the findings related to these factors will be very limited.
Finally, throughout this study, the independent variables
turned out to be highly correlated. Therefore, in order to
evaluate the true relationships between these factors and
the dependent variables as validly as possible, both
correlation and Beta coefficients were used complementarily.
Still, it should be noted that no methodological solution
provides a valid guide toward true relationships in this
multi-collinearity situation. The only true solution is
theoretical; that is, identify the theoretical relationships
among independent variables and reflect them in the
analyses.

In order to overcome the limitations in this study and
extend its scope, this study suggests the following as the
major directions for future studies. First, it is necessary
to replicate this study with different samples in order to
check the external validity of the findings. Secondly, if
possible (e.g., in an area where multi-channel TV has been
newly introduced), repeated group analysis is strongly
suggested even with a relatively smaller sample size in

order to test the hypotheses proposed in this study more

141
validly, particularly the changes in news learning from TV
news viewing. Thirdly, further theoretical elaboration of
the individual factors included in this study and the
relationships between them are necessary for more valid
methodological analyses and interpretation. Finally, future
studies need to expand the scope of this study by
investigating how the development of multi-channel TV
affects other aspects of TV news viewing. For example, it
is speculated that one of the most fruitful topics for
future research is to identify how TV news viewing patterns
(e.g., duration of viewing, channel changes among various
news programs/stories, attentiveness, depth and breadth of
TV news story exposure, etc.) are affected by the

development of multi-channel-TV.

APPENDIX I

THE DESCRIPTIONS OF 6 NEWS EVENTS

Appendix I.

I] H . E .
Three main groups exist in the former Yugoslavia; the
Serbs, Bosnians, and the Croats. The Yugoslavian crisis
commenced during the month of March 1991 due to the
independence movements of the Balkans. In May 1991,
Croatia called for independence which was protested by the
Serbs in that province. With this and other problems
emerging, a civil war broke out when the Serbian president,
Slobodan Milosevic, tried to bring Croatia under Serbian
control. Yugoslavia was being pulled apart by three major
provinces of Serbia, Bosnia Hercegovina and Croatia. Serbia
is the most powerful in weaponry and other aspects. On
March 1, 1992 the Bosnians voted for independence which was
approved a month later by the European community and the
United States. This caused a break out of tension amongst
the different ethnic groups in Bosnia. In April 1992,
Serbian forces supported by air warfare, tanks and heavy
artillery invaded Bosnian towns and seized 80 percent of the
land. Then, the violence against Bosnians and Croats,
similar to Nazi genocide, started and is still going on.

G-7 Summit Meeting is an annual meeting of seven
industrialized demographic country leaders (Britain, Canada,
France, Germany, Italy, Japan, and the U.S.). In this
year's meeting, those issues of expanding trade and
extending more generous aid to Russia, etc. were discussed.
An important U.S-Japanese agreement emerged unexpectedly
after the formal summit ended. The U.S. had sought a
"framework" agreement with Japan to guide future
negotiations aimed at reducing Japan's enormous surpluses in
trade with the U.S. (nearly $50 billion a year currently).
The Japanese pledged "to achieve a highly significant
decrease" in the trade surplus and to negotiate "sets of
objective criteria" for gauging progress. This agreement is
considered as an important first step which should lead to
more Japanese buying of American goods and services -- autos
and parts, computers, telecommunications equipment,
insurance and financial services.

142

143

I] E] 3 . !l “.3 !

This summer (during July and August), one of the
biggest floods in the U.S. history struck the upper
Mississippi areas. The flooding of the Mississippi and
Missouri and tributaries has contributed to the deaths of 48
people and caused at least $ 10 billion in damage, flooding
17 million acres in nine states in the mid-west (by the 2nd
of Aug). Congress passed a $4.7 billion relief package for
the flood victims.

President Clinton announced a policy (July 19th)
allowing gays and lesbians to serve in the military, but
only if they are discreet and don't engage in homosexual
conduct. The policy falls far short of what Clinton
promised during last year's presidential campaign: an
unqualified lifting of the armed services' 50 year old ban
on homosexuals. Clinton's policy, released after a six-
month study by Pentagon officials, has been dubbed "don't
ask, don't tell, don't pursue." It permanently would forbid
recruits from being asked if they are homosexual and would
end military "witch hunts" of suspected gays and lesbians.
It would permit homosexuals a "zone of privacy" as long they
don't engage in homosexual acts.

APPENDIX II

TELEPHONE SURVEY QUESTIONNAIRE

Appendix 2.

Wire

>U1< Before we begin, let me tell you that any information
you give me will be kept strictly confidential. Let
me also tell you that this interview is completely
voluntary.
Should we come to any question that you don't want to
answer, just let me know and we’ll go on to the next
question.

>A1< First, how long have you lived in Michigan?
<1> LESS THAN 1 YEAR
<2> 1 TO 3 YEARS
<3) 4 TO 6 YEARS
<4) 7 TO 10 YEARS
<5> MORE THAN 11 YEARS

<8> DO NOT KNOW
<9> REFUSED - NO ANSWER

>A2< Do you subscribe to cable televison?

<1> YES
<5> NO

<8> DO NOT KNOW
<9> REFUSED - NO ANSWER

>A3< Do you read the newspaper?

<1> YES
<5> NO

<8) DO NOT KNOW
<9> REFUSED - NO ANSWER

144

145

>B1< Next, I would like to read you several statements and
have you tell me to what extent you agree or disagree
with each statement.

I try to make myself a more knowledgeable person from
what I read or view.

Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

>B2< I am the type of person who feels left out if I'm not
familiar with what’s going on in society.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

>B3< If I can be entertained from what I read or view, I
don’t care much about the informational content.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOHEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

146

>B4< It is important to me to have well-informed opinions
about controversial issues.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

>BS< I tend to turn to what I read or view just to pass the
time.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

>B6< I tend to avoid complicated material that I have to
try hard to understand.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOHEWHAT AGREE
<3> NEUTRAL

<4> SOHEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

147

>B7< I like to hear the ideas of others, in order to have
more objective ideas of my own.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

>BB< I tend to read or view material to relax rather than
to gather new information.

(Would you say you strongly agree, somewhat agree,
somewhat disagree, or strongly disagree?)

<1> STRONGLY AGREE
<2> SOMEWHAT AGREE
<3> NEUTRAL

<4> SOMEWHAT DISAGREE
<5> STRONGLY DISAGREE

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

(if the answer for A2 is 1, go to Cla; If the answer for A2
is 5, go to Clb)

>CIa< The next set of questions are about your television
viewing habits. Through out the survey, television
refers to both broadcast and cable television.

On average, how many days per week do you watch
television from 6 to 7 pm?

<1> 1 DAY PER WEEK

<2> 2 DAYS PER WEEK
<3> 3 DAYS PER WEEK
<4> 4 DAYS PER WEEK
<5> 5 DAYS PER WEEK
<6> 6 DAYS PER WEEK

<7> EVERYDAY
<8> NEVER

<97> OTHER:SPECIFY
<98> DON'T KNOW
<99> REFUSED - NO ANSWER

 

148

>C1b< The next set of questions are about your television
viewing habits.

On average, how many days per week do you watch
television from 6 to 7 pm?

<1> 1 DAY PER WEEK

<2> 2 DAYS PER WEEK
<3> 3 DAYS PER WEEK
<4> 4 DAYS PER WEEK
<5> 5 DAYS PER WEEK
<6> 6 DAYS PER WEEK

<7> EVERYDAY
<8> NEVER

<97> OTHER:SPECIFY
<98> DON'T KNOW
<99> REFUSED - NO ANSWER

>C2< On average, how many days per week do you watch TV
from 11 PM to 12 midnight?

<1> 1 DAY PER WEEK

<2> 2 DAYS PER WEEK
<3> 3 DAYS PER WEEK
<4> 4 DAYS PER WEEK
<5> 5 DAYS PER WEEK
<6> 6 DAYS PER WEEK

<7> EVERYDAY
<8> NEVER

<97> OTHER:SPECIFY
<98> DON’T KNOW
<99> REFUSED - NO ANSWER

>C3< On average, how many hours per day do you spend
watching television?

<O-24>

<98> DON’T KNOW
<99> REFUSED - NO ANSWER

149

>Dl< Do you watch the local evening news or the network

news on

<1>
<2>
<3>
<5>

<8>
<9>

television?

YES, LOCAL EVENING NEWS (SKIP le)
YES, NETWORK NEWS (SKIP Dla)

BOTH LOCAL AND EVENING NEWS

NO (SKIP Dla and 01b)

DON’T KNOW (SKIP Dla and le)
REFUSED - NO ANSWER (SKIP D18 and le)

>D1a< On average, how many days per week do you watch the
local evening news?

<1>
<2>
<3>
<4>
<5>
<6>
<7>
<8>

DAY PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
EVERYDAY

NEVER

O‘U‘lobUNH

<97> OTHER:SPECIFY
<98> DON’T KNOW
<99> REFUSED - NO ANSWER

>D1b< On average, how many days per week do you watch the

network

<1>
<2>
<3>
<4>
<5>
<6>
<7>
<8>

news (ABC,CBS, NBC or PBS)?

DAY PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
EVERYDAY

NEVER

aim-FUNH

<97> OTHER
<98> DON'T KNOW
<99> REFUSED - NO ANSWER

(IF A2 is 5, Skip E1 to E6)

150

>E1< Do you subscribe to premium cable channels? (such as
HBO, Showtime, Cinemax, or Disney)?

<1>
<5>

<8>

YES
NO

DON’T KNOW

<9> REFUSED - NO ANSWER

>E2< Do you watch the specialized cable news channels, such

as CNN,

<1>
<5>

<8>
<9>

CNN headline news, C-SPAN, or C-NBC?

YES
NO [go to E5]

DON’T KNOW[go to E5]
REFUSED - NO ANSWER

>E2a< If you had to, would you pay an additional fee to

recieve

these news channels?

<1> YES

<5> NO

<8> DO NOT KNOW
<9> REFUSED

>E3< On average, how many days per week do you watch these

channels?

<1>
<2>
<3>
<4>
<5>
<6>
<7>
<8>

DAY PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
EVERYDAY

NEVER

O‘U‘lwaH

<97> OTHER:SPECIFY
<98> DON'T KNOW
<99> REFUSED - NO

ANSWER[#Imd1=99][##md2=0][##b1ank=01

>E4<

151
On average, how many hours per day do you spend
watching these news channels (such as CNN, CNN
headline news, C-SPAN, or C-NBC)?
<0-24> HOURS

<98> DON’T KNOW
<99> REFUSED - NO

ANSWER[##md1=99][##md2=0][##blank=01

===>

>E4a< People tend to turn to a set of cable channels

>E5<

intentionally, and other channels, unintentionally by
channel hopping.

Would you say you turn to CNN or other cable news
channels intentionally or unintentionally?

<1> CHANNELS TURNED TO INTENTIONALLY
<5> CHANNELS TURNED TO UNINTENTIONALLY

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

<1>
<5>

<2>
<3>
<4>
<6>
<7>
<8>
<9>
<10>
<1l>
<12>
<l3>
<14>
<15>
<16>
<17>
<18>

<97>
<99>

Can you tell me what type of programming is shown on
C-SPAN?

YES: SPECIFY PROGRAMMING
NO [go to E6]

US HOUSE AND SENATE (CONGRESS)
MICHIGAN HOUSE AND SENATE (LEGISLATURE)
NEWS BROADCASTS

GOVERNMENT HEARINGS (CONGRESSIONAL)
COURT CASES-TRIALS

TALK SHOWS: LARRY KING, CROSSFIRE, ETC
POLITICAL BROADCASTS

SPANISH CHANNEL

SPORTS CHANNEL

DOCUMENTARIES

WORLD NEWS

LOCAL GOVERNMENT

ENVIRONMENTAL PROGRAMMING
BUSINESS-STOCKS

EDUCATIONAL

MOVIES

OTHER:SPECIFY
REFUSED - NO ANSWER

>E6< Can you
C-NBC?

<1>
<5>

<2>
<3>
<4>
<6>
<7>
<8>
<9>
<10>
<11>
<12>
<l3>
<l4>

<97>
<99>

152

tell me what type of programming is shown on

YES:SPECIFY TYPE OF PROGRAMMING
NO

TALK SHOWS-INTERVIEWS

NEWS BROADCASTS-INFORMATIONAL PROGRAMMING
BUSINESS SHOWS-BROADCASTS

FINANCIAL PROGRAMMING (STOCKS-WALL STREET)
DOCUMENTARIES

CHRISTIAN BROADCASTS

STITUATION COMEDIES

COURT CASES-TRIALS

GOVERNMENTAL PROGRAMS

SPORTS

EDUCATIONAL

SOAP OPERAS

OTHER:SPECIFY
REFUSED - NO ANSWER

>F1< The next set of questions deal with how you rate

televisi

How woul
you. Wo
somewhat

<1>
<2>
<3>
<4>
<5>

<8>
<9>

>F2< How woul

on.

d you rate television as a way of entertaining
uld you say it is very good, somewhat good,
poor or very poor.

VERY GOOD
SOMEWHAT GOOD
NEUTRAL
SOMEWHAT POOR
VERY POOR

DON’T KNOW
REFUSED - NO ANSWER

d you rate television as a way to relax?

(Would you say it is very good, somewhat good,

somewha

<1>
<2>
<3>
<4>
<5>

<8>
<9>

t poor or very poor. )

VERY GOOD
SOMEWHAT GOOD
NEUTRAL
SOMEWHAT POOR
VERY POOR

DON’T KNOW
REFUSED - NO ANSWER

153

>F3< (How would you rate television) as a way to make you
more knowledgeable.

(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1> VERY GOOD

<2> SOMEWHAT GOOD
<3> NEUTRAL

<4> SOMEWHAT POOR
<5> VERY POOR

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

>F4< (How would you rate television) as a tool for helping
you develop well-informed opinions about
controversial issues.
(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1> VERY GOOD

<2> SOMEWHAT GOOD
<3> NEUTRAL

<4> SOMEWHAT POOR
<5> VERY POOR

<8> DON'T KNOW
<9> REFUSED - NO ANSWER

>F5< (How would you rate television) at presenting other
people's ideas that can be compared to yours?

(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1> VERY GOOD

<2> SOMEWHAT GOOD
<3> NEUTRAL

<4> SOMEWHAT POOR
<5> VERY POOR

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

>F6< (How would you rate television) as a way to pass the

time.

(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1>
<2>
<3>
<4>
<5>

<8>
<9>

>F7< (How would you rate televsion) as a way to escape from

154

VERY GOOD
SOMEWHAT GOOD
NEUTRAL
SOMEWHAT POOR
VERY POOR

DON'T KNOW
REFUSED - NO ANSWER

complicated daily material.

(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1>
<2>
<3>
<4>
<5>

<8>
<9>

>F8< (How would you rate television) as a way of making you
familiar with what is going on in society.

(Would you say it is very good, somewhat good,
somewhat poor or very poor.)

<1>
<2>
<3>
<4>
<5>

<8>
<9>

VERY GOOD
SOMEWHAT GOOD
NEUTRAL
SOMEWHAT POOR
VERY POOR

DON’T KNOW
REFUSED - NO ANSWER

VERY GOOD
SOMEWHAT GOOD
NEUTRAL
SOMEWHAT POOR
VERY POOR

DON’T KNOW
REFUSED - NO ANSWER

 

 

 

155

>Gl< On a scale of 0 to 100, (where 100 is the highest
rating and 0 is the lowest rating), how would you rate
television as a method of providing a variety of news
stories?

<0 - 100> POINTS
<998> DON’T KNOW
<999> REFUSED - NO ANSWER

>G3< On a scale of 0 to 100, (where 100 is the highest
rating and 0 is the lowest rating), how would you rate
television as a method of providing in-depth news
coverage?

<0 - 100> POINTS
<998> DON'T KNOW
<999> REFUSED - NO ANSWER
>H1< Next, I would like to ask you a couple of questions

about reading the newspaper and listening to the radio.

On average, how many days per week do you read the

newspaper?
<1> 1 DAY PER WEEK
<2> 2 DAYS PER WEEK
<3> 3 DAYS PER WEEK
<4> 4 DAYS PER WEEK
<5> 5 DAYS PER WEEK
<6> 6 DAYS PER WEEK

<7> EVERYDAY
<8> NEVER

<97> OTHER:SPECIFY
<98> DON'T KNOW
<99> REFUSED - NO ANSWER

156

>H1a< On average, how many days per week do you listen to
the news on the radio?

<1> 1 DAY PER WEEK
<2> 2 DAYS PER WEEK
<3> 3 DAYS PER WEEK
<4> 4 DAYS PER WEEK
<5> 5 DAYS PER WEEK
<6> 6 DAYS PER WEEK
<7> EVERYDAY

<8> NEVER

<97> OTHER

<98> DON'T KNOW

<99> REFUSED - NO ANSWER

>H2< On average, how many days per week do you talk about
news stories with your family or friends?

<1>
<2>
<3>
<4>
<5>
<6>
<7>
<8>

DAY PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
DAYS PER WEEK
EVERYDAY

NEVER

001$wa

<98> DON'T KNOW
<99> REFUSED - NO ANSWER

>H3< Compared to the newspaper, would you say that
television is better, about the same, or worse in
providing a variety of news stories?

<1>
<2>
<3>

<8>
<9>

BETTER
ABOUT THE SAME
WORSE

DON'T KNOW
REFUSED - NO ANSWER

157

>H4< Compared to the newspaper, would you say that
television is better, about the same, or worse in
providing in—depth news coverage?

<1> BETTER
<2> ABOUT THE SAME
<3> WORSE

<8> DON’T KNOW
<9> REFUSED - NO ANSWER

>H5< What is your primary source for news? Would you say
TV, newspaper, family members, friends, radio, or
something else?

<1> TV

<2> NEWSPAPER

<3> FAMILY MEMBERS
<4> FRIENDS

<5> RADIO

<6> COMBINATION

<97> OTHER:SPECIFY

<98> DON’T KNOW

<99> REFUSED - NO ANSWER

>Il< Now, I would like to ask you a few questions about
some national and international news stories.

Are you aware of the flooding that has occured this
summer?

<1> YES
<5> NO[go to 12]

<8> DO NOT KNOW-DO NOT REMEMBER
<9> REFUSED [go to 12]

158

>Ila< From what you have heard, which area, the East Coast
including Florida, the Upper Mississippi River area, or the
West Coast, has been most seriously damaged by the
flooding?

(IF THE RESPONDENT SAYS ’DO NOT KNOW’ DO NOT PROBE,
BUT RECORD ’DON'T KNOW.’)

<1> EAST COAST AREA INCLUDING FLORIDA
<2> UPPER MISSISSIPPI RIVER AREA

<3> WEST COAST AREA

<7> OTHER: SPECIFY

<8> DON’T KNOW
<9> REFUSED -NO ANSWER

>12< Are you aware of the war in Bosnia (formerly
Yugoslavia)?

<1> YES
<5> NO [goto 14]

<8> DO NOT KNOW—DO NOT REMEMBER
<9> REFUSED [goto I4]

>12a< From what you heard, is the war a result of a
conflict between students and the government: a
conflict among countries in Eastern Europe: or a
conflict among different ethnic groups?

(IF THE RESPONDENT SAYS ’DO NOT KNOW’ DO NOT PROBE,
BUT RECORD 'DON’T KNOW.’)

<1> THE CONFLICT BETWEEN STUDENTS AND THE
GOVERNMENT

<2> THE CONFLICT AMONG COUNTRIES IN EASTERN EUROPE

<3> THE CONFLICT AMONG DIFFERENT ETHNIC GROUPS

<7> OTHER:SPECIFY

<8> DON’T KNOW
<9> REFUSED -NO ANSWER

>I3<

>13a<

>I4<

>I4a<

159

Are you aware of the Summit Meeting with the leaders

of seve

<1>
<5>

<8>
<9>

n democratic countries in Tokyo last month?

YES
NO [goto IS]

DO NOT KNOW-DO NOT REMEMBER
REFUSED [go to 15]

From what you have heard about the Summit meeting,

was th
peace
trade

(IF T
BUT

<1>
<2>
<3>

<7>

<8>
<9>

Are you

<1>
<5>

<8>
<9>

e key issue between the US and Japan keeping
in East Asia: helping Somalia; or reducing the
surplus?

HE RESPONDENT SAYS ’DO NOT KNOW’ DO NOT PROBE,
RECORD ’DON’T KNOW.’)

KEEPING PEACE IN EAST ASIA
HELPING SOMALIA
REDUCING THE TRADE SURPLUS

OTHER:SPECIFY

DON’T KNOW

REFUSED -NO ANSWER

aware of the issue of Gays in the Military?

YES
NO [go to J1]

DO NOT KNOW-DO NOT REMEMBER
REFUSED [go to J1]

From what you have heard about this issue, will
President Clinton's policy allow all Gays and

Lesbia
allowe
serve

ns to serve in the military: they will not be
d to serve: or that they will be allowed to
only if they are discreet?

(IF THE RESPONDENT SAYS ’DO NOT KNOW’ DO NOT PROBE,
BUT RECORD ’DON’T KNOW.’)

<1>
<2>
<3>
<7>

<8>
<9>

ALL GAYS AND LESBIANS CAN SERVE IN THE
MILITARY

NO GAYS AND LESBIANS CANNOT SERVE IN THE
MILITARY

SERVE AS LONG AS THEY ARE DISCREET

OTHER
DON’T KNOW
REFUSED -NO ANSWER

160

>J1< Finally, I have a few background questions that will
be used for statistical purposes only.

What is your age as of your last birthday?
<18-110> YEARS
<998> DO NOT KNOW
<999> REFUSED
>gen< RECORD GENDER OF RESPONDENT HERE: ASK ONLY IF
IN DOUBT
<1> MALE
<5> FEMALE
>J2< How many people currently live in your household

including adults and children?

<1> PERSONS [go to J4]
<2-20> PEOPLE

<9a> DO NOT KNOW [go to J4]
<99> REFUSED [go to J4]

>J3< Are there any children in the household (that is,
individuals less than 18 years of age)?

<1> YES
<5> NO [go to J4]

<8> DO NOT KNOW [go to J4]
<9> REFUSED -NO ANSWER [go to J4]

>J3a< How many children are in the household?
<1-25>

<98> DON’T KNOW
<99> REFUSED -NO ANSWER

161

>J4< What is the highest grade or level of education that
you have completed?

<1>
<2>
<3>
<4>
<5>
<6>
<7>
<8>

GRADE SCHOOL ONLY

DID NOT FINISH HIGH SCHOOL
HIGH SCHOOL GRADUATE
VOCATIONAL OR TECHNICAL SCHOOL
SOME COLLEGE

COLLEGE GRADUATE (BA, BS)

POST GRADUATE WORK

GRADUATE DEGREE (MA, MS,Ph.D)

<97> OTHER:SPECIFY
<98> DO NOT KNOW
<99> REFUSED -NO ANSWER

>J5< What is your main ethinic or racial background? Would
you say you are Black (or African American), White or
(Caucasian), Asian (or Pacific Islander), Native
American, or Hispanic?

>J6< What
<1>
<2>
<3>
<4>
<5>

<6>
<7>
<8>
<9>
<10>
<97>
<98>
<99>

<1>
<2>

<3>
<4>
<5>

BLACK OR AFRICAN AMERICAN

WHITE OR CAUCASION (GERMAN, NORWEGIAN,
ITALIAN, ETC)

ASIAN OR PACIFIC ISLANDER

NATIVE AMERICAN

HISPANIC

<97> OTHER:SPECIFY
<98> DO NOT KNOW
<99> REFUSED -NO ANSWER

is your occupation?

UPPER WHITE COLLAR: Businessperson/
Professional, Self Employed

LOWER WHITE COLLAR: Sales, Managers, Clericals.
Legal Aids, Clerks

UPPER BLUE COLLAR: Skilled Trades, Police,
Fireman, Animal Goomer

LOWER BLUE COLLAR: Operatives Laborers,
Construction

GOVERNMENTAL EMPLOYEE: State Federal Local Gov't
employee

FARMER

HOUSEWIFE-HOUSEHUSBAND

STUDENT

UNEMPLOYED

RETIRED

OTHER:SPECIFY

DO

NOT KNOW

REFUSED

>J7<

>J7a<

>J7b<

>J7c<

>J7d<

>J7e<

>J7f<

162

To get a picture of peOple's financial situation, we
need to know the general range of incomes of all
households we interview. Now, thinking about your
household's total annual income from all sources
(including your job) in 1992, did your household
receive $30,000 or more in 1992?

<1-12>
<1> NO [go to J7h]
<6> YES[go to J7d]

<93> DON'T KNOW/NO OPINION
<99> REFUSED/NO ANSWER

Was it $35,000 or more?
<1-12>
<7> YES
<6> NO OR DON’T KNOW/REFUSED

Was it $40,000 or more?

<1-12>

<8> YES [go to J7c]

<7> NO DON’T KNOW/REFUSED [go to J7a]

Was it $45,000 or more?

<1-12>

<9> YES

<3> NO OR DON’T KNOW/REFUSED
Was it $50,000 or more?

<1-12>

<10> YES[goto J7f]

<9> NO OR DON’T KNOW/REFUSED [go to J7h]
Was it $55,000 or more?

<1-12>

<11> YES

<10> NO OR DON’T KNOW/REFUSED
Was it $60,000 or more?

<1-12>

<12> YES
<11> NO OR DON’T KNOW/REFUSED

163
>J7g< Was it $10,000 or more?
<1-12>
<2> YES
<1> NO OR DON'T KNOW/REFUSED
>J7h< [equiv J7][allow 2]Was it $15,000 or more?
<1-12>
<3> YES[goto J7i]
<2> N0 OR DON'T KNOW/REFUSED [go to J7g]
>J7i< Was it $20,000 or more?
<1-12>
<4> YES[goto J7j]
<3> NO OR DON’T KNOW/REFUSED
>J7j< Was it $25,000 or more?
<1-12>

<5> YES
<4> NO OR DON'T KNOW/REFUSED

 

 

E'lJ' l

Abott, E. A. (1978). "Effects of year-long newspaper energy
series on reader knowledge and action." Presented to
the Association for Education in Journalism, University
of Washington, Seattle.

Allen, I. L., & Colfax, J. D. (1968). "The diffusion of news

of LBJ's March 31 decision." Journali§m_gnaz§gzly
45:321-324.

Anderson, D. et. al. (1981). ”The effects of TV program
comprehensibility on preschool children's visual

attention to television." Qn11d_neyglgpment. 52:151-
157.

Anderson, R. C. et. al. (1977)."Frameworks for comprehending
discourse." American_Educational_8esearsh_lonrnal.

14:367-383.
Arbitron RatinQS- (1983). IEQ:uax_sable_diar¥_testi_a
summa:¥_nf_finnings. Laural. MD: Author-

Atkin, C. K. (1990). "Exposure Motivations." Unpublished
manuscript. Department of Communication, Michigan State
University.

Atkin, C. K., Galloway, J., 8 Nayman, O. B. (1976). "News
media exposure, political knowledge and campaign

interest." _Jgnznalism_gnartezly. 53: 231-237.

Atkinson, J. W. (1957). ”Motivational determinants of risk-
taking behavior."Psychologiga1_neyiew.64:359-372.

Babrow, A. S., & Swanson, D. L. (1988). "Disentangling
antecedents of audience exposure levels: extending
expectancy-value analyses of gratifications sought from

television news." Communicatign_ugnggraphs. 55:1-21.
Bailey, G. A. (1971). "The public, the media, and the

knowledge gap."

4:3-8

Baldwin. T.. & McVoy. 0- (1988). Cable_99mmuni§atign.
Englewood Cliffs, N.J.: Prentice-Hall, Inc.

164

165

Baldwin, T. F., Barrett, M., & Bates, B. (1992a). "Influence
of Cable on Television News Audiences." Jenraaliem

Quarterly. 69(3):651-658.

Baldwin, T. F., Barrett, M., & Bates, B. (1992b). "Uses and
Values for News on Cable Television." Jearnal_er

Br9ad2asting_§_Electreni§_Media 36(2): 225-234-

Baldwin, T. F. et. al. (1988). "News viewing elaborated." in
C. Heeter & B. S. Greenberg (eds. ) (1988).

Qableyieging. Norwood, NJ: Ablex.

Bantz, C. R. (1982). "Exploring use and gratifications: a
comparison of reported uses of television and reported

uses of favorite program type" Communication_nesearch
9. 352- 379.

Barber, J. (1979). "Not the New York Times: what network
news should be." Waehingren_nenthly (September):14-21.

Bates, B. J. (1989). "Evolving into an information society:
Problems and issues." in Jerry L. Salvagio (ed.). The

Issues. Hillsdale, NJ: Lawrence Erlbaum Associates,
Inc., Pub.

Batra, R., 8 Ray, M. (1983). "Operationalizing involvement
as depth and quality of cognitive response." in R.
Bagozzi 8 A- Tybout (edso). Adxanses_in_gonsnmer
reeeareh (vol. 10). Ann Arbor, MI: Association for
Consumer Research. pp.309-313.

Becker. L. (1979). "Measurement of Gratifications.”

Communisation_neseareh 6(1): 54-73.

Becker, L., Dunwooody, S., & Rafaeli, S. (1983). "Cable's
impact on use of other news media." Jenrna1_er

Broadcasting- 27(2)=127-140-

Becker, L., & Whitney, D. C. (1980). "Effect of media
dependencies audience assessment of government. "

Communieatign_nesearch 7: 95-120-

Becker, L. et. al. (1978). "Debates' effects on voters'
understanding of candidates and issues," in G. F.

Biship et- al- (eds.) The_£residential_nebates. New

York: Praeger.

Bentz, C. R. (1982). "Exploring uses and gratifications: a
comparison of reported uses of television and reported
uses of favorite Program type-" Communisation_fiasearch.
9(3)L 352-379.

166

Berry, C. (1983). "Learning from television news: a critique

of the research." Journal_of_flroadsastingo 27(4)

Bever, T. G. (1970). "The cognitive basis for linguistic
structures," in J. R. Hayes (ed. ).

Dexelonment_of_Language New Your: WileY-
Blumler, J. G., & Katz, E. (eds. )(1974). The_fleee_e£_flasfi

O
- Q - _
ouu . . 0;- 0‘ - e. . . 0.-

Beeeareht Beverly Hills, CA: Sage.

Booth, A. (1970). "The recall of news items.” £ublie_Qpinlen
Quarterly. 34:604-610.

Bowman, G., & Farley, J. (1972). "TV Viewing: Application

Of a formal choice model.“ Applied_£eenemlee, 4:245-
259.

Bransford, J. D., & Johnson, M. K. (1972). ”Contextual
prerequisites for understanding: some investigations of
comprehension and recall." I9urnal_9f_Yerbal_Learninn
and_Yerbal_Behaxior 11: 717-726.

Brinton, J. E., & McKown, L. N. (1961). "Effect of newspaper
reading on knowledge and attitude." Jeurnallem

Quarterly 38:187-195.

Britton, B. K. & Tesser, A. (1982). ”Effects of prior
knowledge on use of cognitive capacity in three complex
cognitive tasks."

Benayler. 21: 421-436.

Britton, B. K. et. al. (1982). "Effects of text structure on
use of cognitive capacity during reading." 1eurnal_er
Ednsational_£§¥gholngx- 74(1):51-61-

Britton, B. K., a Price, K. (1981). ”Use of cognitive
capacity in reading: A processing operating
characteristic.“ 2ersentua1_and_notor_Skills. 52:291-
298.

Britton, B. K. et. al. (1979). nUse of capacity in reading
identical texts with different amounts Of discourse
level neanin9-" J9urna1_9f_Experimental_zs¥9hnlesxi

nunan_Learning_ann_Manorx- 5(3):262-270-

Britton, B. K. et. al. (1980). “Use of cognitive capacity in
reading easy and difficult text: two tests of an
allocation of attention hypothesis." Jeurnal_er_3eaalng
Behayler. 12(1):23-30.

167

Britton, B. K. et. al. (1980). "Use of cognitive capacity in
reading: effects of processing information from text

for immediate recall and retention." Jeurnal_ef_£eading
Behaxigr 12(2)

Broadbent. D- E- (1958). 2ercention_and_§9mmunicatign- New

York: Pergamon Press.

Breadeaeting. (Sep. 24. 1990). "Network news: changing as it
remains the same." pp. 34-39.

Breageaeting. (Feb. 3, 1992). "CBS's Safer blasts network
news." p. 14.

Brown, J. W., Ettema, J. S., & Luepker, R. V. (1981)
"Knowledge gap effects in a cardiovascular information
campaign." Presented to the Association for Education
in Journalism, East Lansing, MI.

Brown, R. (Mar. 8, 1993). "Keeping the cable customer
satisfied." Broadcasting_§_cahle. 123(10):10-

Browne, K. (1978). "Comparison of factual recall from film

and print stimuli.” Jeurnaliem_Quarterly 55:350-53.

Bruno, A. (1973). "The network factor in TV viewing.
‘ ‘ and Marketing Research,
13(5):33-39.

Bryant, J. 5 Anderson, D. (eds) (1983). thldrenle

Qanpreheneien. New York: Academic Press.

Budd, R. W., MacLean, M.S. Jr., a Barnes, A. M.(1966).
"Regularities in the diffusion Of two news events."

Journalism_0uarterlxo 43:221-230-

Cantril, H. (1943). "Identification with social and economic
class." J9urnal_9f_Ahn9rmal_and_social_2§¥cneloax 38
(January):74-80.

Carter, B. (1985).
Middle Qlaeel London: Routledge and Kegan Paul.

Centers, R- (1949) Ihfl_2EYQhQl99!;9£_SQQifll_Ql§§§§fii_A
Claas_sgnaciousnessl Princeton. New Jersey:

Princeton University Press.

Cermak, L.S., & Craik, F. (eds.) (1979). Leyele_er

Eroceasing_in_nnman_nenorx- Hillsdale. N-J.= Lawrence
Erlbaum Associates, Pub.

168

Chaffee, S., & Roser, C. (1986). "Involvement and the
consistency of knowledge, attitudes and behavior.

Communicatian_3esearch 13: 373-400-

Chaffee, S., & Schleuder, J. (1986), "Measurement and
effects of attention to media news. Human_gemmunieatlen
Researeh. 13: 76: 107.

Chew, F. (1992). "Information needs during viewing of

serious and routine news." lsurnal_of_8r9adcasting_§
Electronic_fledia. 369(4)=453-466-

Clarke, P., & Kline, F. G. (1974) "Media effects
reconsidered: some new strategies for communication

researCh- " Communication_8esearch 1: 224’ 240-

Cliftan, C., & Slowiaczek, M. (1981), "Integrating new

information with old knowledge." Menery_and_gegnitian
9:142-148.

Collins, J., Reagan, J., & Abel, J. D. (1983). "Predicting
cable subscribership: local factors." Jeurnal_ef

Breaaeaetlng. 27(2):177-183.

Cutler, N. E., 8 Danowski, J. A. (1980). "Process

gratifications in aging cohorts." Jeurnaliem_Quarterly.
57:269-277.

Darmon, R. (1976). "Determinants of TV viewing." Jeurnal_er
Adyertlelng. Research 16:17-20.

DeFleur, M. L. et. al. (1992). "Audience recall of news
stories presented by newspaper, computer, television,
and radio." Qeurnallem_Quarterly. 69(4): 1010-1022.

Donahue, G. A., Olien, C. N., & Tichenor, P. J. (1989).
“Knowledge Gaps and Smoking Behavior." Paper presented
at the annual conference of the Midwest Association for
Public Opinion Research, Chicago, Illinois, November
17, 1989.

Donahue, G. A., Olien, C. N., & Tichenor, P. J. (1975).
"Mass media and the knowledge gap.” cannunleatlen
Research 2: 3-23-

Dooling, D. J. (1972). "some context effects in the speeded

comprehension of sentences." Jeurnal_er_rxperlmental
EEYQthQQY- 93:56-62-

Daaling, D. J., & Lachman, R. (1971). "Effects of
comprehension an retention of prose." Jeurnal_er

Exnerinental_zsxchalesx- 88:216-222-

169

Douglas, C. W., & Stacey, D. C. (1972L "Demagraphical
characteristics and social factors related to public
opinion on fluoridation." Jeurnal_er_2ublle_Healtn
Dentistry 32: 128- 234.

Eastman, M. (1982L
cher Writingel New York: The Modern Library.

Edelstein, A. S. (1973). "Decision-making and mass
communication: a conceptual and methodological approach
to public opinion." in P. Clarke (ed. ) Neu_flegel_rer
WW. Beverly Hillsn CA: Sage-

Edwards, W. (1954L "The theory of decision making."
W 51: 380-417-

Eichar. D- 1L. (1989) W
Amerieat New York: Greenwood Press.

Ettema, J. S., Brown, J. W., & Luepker, R. V. (1983).
"Knowledge gap effects in a health information

campaign." W 4'7: 516-527-

Ettma, J. S., & Kline, F. G. (1977L "Deficits, differences
and ceilings: Contingent conditions for understanding

the knowledge gap." Wren 4: 179-202-

Feather, N. T. (1959). "Subjective probability and decision
under uncertainty." Eeyenelegieal_3eyleu. 66: 150-164.

Findahl, 0., & Hoijer, B. (1985). "Some characteristics of
news memory and comprehension." 1eurnal_efi_fireaaeaetlng
WW- 29:279-296.

Findahl, O., & Hoijer, B. (1975). "Man as receiver of
information: on knowledge, social privilege and the

news . "W stockholm.

Fishbein, J. (1963). "An investigation of the relationships
between beliefs about an object and the attitude toward

that object.” BREEDLRQLBLIQDS- 16:233-240.

Fishbein, J., & Ajzen, I. (1975). 8911§£ITALLLLHQQI
Reeeareh. Reading, Mass.: Addison-Wesley.

Fry, D. L. (1979). "The knowledge gap hypothesis and media
dependence: An initial study." Paper presented to

Association for Education in Journalism, University of
Houston, TX.

170

Galloway, J. J. (1977L "The analysis and significance of

communication effects gaps." gemmunieatlen_3eeeareh 4:
363- 386.

Galloway, J. J., 8 Meek, F. L. (1981L "Audience Uses and
Gratifications: an expectancy model. " gemmunleatlen

REEQQIQD- 8(4): 435-449.

Gantz, W. (1978L "How uses and gratifications affects

recall of television news." Qeurnallem_Quarterly,. 55:
664- 672, 681.

Gaziano, C. (1988). "Community knowledge gaps." ertleal
Stud1es_in_Eass_Qommunicatign- 8: 351- 357.

Gaziano, C. (1983). "The knowledge gap: An analytical review
of media effects." gemmunleatlen_3eeearen. 10: 447-486.

Gaziano, C.(1982). "The influence of news media and citizen
groups on the Knowledge gap in an inner-city
neighborhood." Paper presented to the American
Association for Public Opinion Research, Hunt Valley,
MD (May).

Gardner, H. (1985). ' ' ' ’
Cosnitixe_3exalutian. New York: Basic Books.

Gardner, E. T., 8 Schumacher, G. M. (1977L "Effects of
Contextual organization on prose retention." Jeurnal_er

Educational_25¥chologx- 69:146-151-

Gelman, M. (Oct. 6, 1983). "Cable news fails to harm local
newscast viewership." Eleetrenie_ueala, p. 18.

Genova, B. K. L., 8 Greenberg, B. S. (1978). ”Interest in

News and the Knowledge 689-" Enb119_Qpinian_Quarterl¥-
43: 79-91.

Goodhardt, G. J., Ehrenberg, A. S. C., 8 Collins, M. A.
(1975) The_Ielexisian1And1encei_2atterns_9f_21ewins-
Westmead, England: D.C. Heath.

Graber, D. A. (1990L "Seeing is remembering: how visuals
contribute to learning from television news." Jeurnal

af.99mmunicatian 40(3)=134-155-

Graber, D. A. (1978L "Problems in measuring audience
effects of the 1976 debates." In G. F. Bishop et. al.

(eds ) Ine_Ereaidential_Debate511Med1a1_Electaral1_and
291191122rsnectixes

. New York: Praeger.

171

Greeno, J. G. (1977). "Language understanding and learning."
Paper presented at the Symposium an Individual
Differences, Cognition, and Learning. American
Association for the Advancement of Science, Denver,
Colo.

Greenwald, A. G., 8 Leavitt, C. (1985). "Cognitive Theory
and Audience Involvement." in Linda, F. Alwitt and

Andrew A. Mitchell (eds.) 2sYchologicel_zrocesses_and

Apnlieatiene. Hillsdale New Jersey: Lawrence Erlbaum
Associates, Pub. pp.221-240.)

Grotta, G., 8 Newsom, D. (Win., 1983L "How does cable
television in the home relate to other media use

patterns?" Q9urnalism.Qnarterl¥. 60: 558-591. 609.

Gunaratne, S. A. (1976). "MOdernization and knowledge: a
study of four Ceylanese villages.” Qannunieatien
Menegrapnet Singapore: Asian Mass Communication
Research and Information Center.

Heeter, C. (1988). ”The choice process model." in C. Heeter
8 B. S. Greenberg (eds.) (1988). Qableyieulng. Norwood,
NJ: Ablex, pp. 11-32.

Heeter, C., 8 Baldwin, T. F. (1988). ”Channel Types and
Viewing Styles." in C. Heeter 8 B. S. Greenberg (eds.)
(1988). gableyleylng. Norwood, NJ: Ablex, pp. 167-176.

Heeter, C., 8 Greenberg, B.S. (1988). ”A theoretical
overview of the program choice process." in C. Heeter 8
B. Greenberg (eds.) (1988). Qableyiening. Norwood, NJ:
Ablex, pp.33-50.

Henke, L. et. al. (1984). "The impact of cable on
traditional television news viewing habits." Jeurnallen

Quarterly 61:174-178.

Hofstetter, C. R., Zukin, C. 8 Buss, T. F. (1978) "political
imagery and information in an age of television."

Journalism19uarterlx 55: 562-569-
Hovland. C- et- al. (1953) cammunicetion.and.2ersuasian. New

Haven, CT: Yale University Press.

Huey. E. B. (1968). 2sxchalosY_and_2edanos!_9f_nead1ns-
Boston: MIT Press.

172

Hunter, J. E., 8 Gerbing, D. W. (1982). "Uni-dimensional
measurement, second order factor analysis and causal
models." in B. M. Stew 8 L. L. Cummings (eds. ). Reeeareh

in_Qrganizational_8ehaxior (Vol 4) Greenwich. CT: JAI

Press .

Jeffres, L. (1978a). "Cable TV and interest maximization."

Journalism_Quarterl¥ 55:149-154-

Jeffres, L. (1978b). "Cable TV and viewer selectivity."
Journal_9f_8roadcastins- 22:167-177-

Johnston, W. A., 8 Heinz, S. P. (1978). "The flexibility and
capacity demands of attention." Jeurnal_ef_Experimental
Es¥chologxi_§eneral. 107:420-435.

Journal of Communication (1989). "Special Issue: Information

689-" Journa119f199mmunicationo 39(3)

Israel, H., 8 Robinson, J. P. (1972). "Democratic
Characteristics of Viewers of Television Violence and
News Programs," George A. Comstock and John P. Murray
(eds ) .Telexision_in_Dax_ta_D£11L1f21_£atterns_9f_use-
Washington, D.C.: US Dept. of Health.

Kalsbeek, J. W., 8 Sykes, R. N. (1967). "Objective
measurement of mental load. " Aeta_2eyeneleglea. 27: 253-
261.

Kanerva, E. W. (1979). "How people acquire information: A
model of the public affairs information attainment
process." Paper presented to the Association for
Education in Journalism, University of Houston.

Katz, E., Adoni, H., 8 Parness, P. (1977L "Remembering the
news: what the picture adds to recall." Jeurnallem

Quarterly. 54: 231- 239.

Katz, E., Blumler, J. G., Gurevitch, M. (1974). "Utilization
Of mass communication by the individual. " in J. G.
Blumler and E. Katz (eds. ) The_fleee_efi4flaee

‘!'.'!_O . 0: ' ‘1 1 0! . é 0!.

Reeeareh. Beverly Hills, CA: Sage.

Katzman, N. (1974). "The impact of communicatiOn technology:
promises and prospects." J. of Communication 24. 4:47-
58.

Kent, K. E., 8 Rush, R. R. (1976). "how communication
behavior of older persons affects their public affairs

knowledge-" Journalism_9uarterlx1 53:40-46.

173

Kintsch, W. (1977). Memery_ang_gegnltien. New York: Wiley.

Kintsch, W. (1976). "Memory for prose." in Charles Cofer
(ed-) The_Structure_af_numan;uemarx- San FranciSCO:

Freeman .

LaRose, R., 8 Atkin, D. (1988). "Understanding cable
subscribers as telecommunications behavior.” Telematiee

and_1nfarmatics. 5(4): 377-333-

Larsan, J. F. (1980). "A review of the state of the art in

mass media disaster reporting. In E. M. Rogers, (ed. L
. Washington, DC.:National

Academy of Sciences. pp. 75-126.

LaBerge, D., and Samuels, S. J. (1974). "Toward a theory of
automatic information processing in Reading." gegnltiye

Peyehelegy. 6:293-323.

Lehmann, D. R. (1971). "Television show preference:

application of a choice model." leurnal_er_Marketlng
Reeeareh. 8:141-173.

Lemert, J. B. (Win, 1993). "Do televised presidential
debates help inform voters?" Journal_9f18r9adcasting
and_Electr9nic_Media. 37: 83-94. ’

Lopreato, J., 8 Lewis, L. 8. (ads) (1974). Seelal
Stratifleatien. New York: Harper and Row, pp. 182-192.

Lovlich, N. P., 8 Pierce, J. C. (1984)."knowledge gap
phenomena: effects of situation-specific and

transsituatianal factors." genmunleatien_neeearen. 11:
415-434.

Mechlnp. F- (1980). Knawledge_and_Knowledge_£rodnstian.

Princeton University Press.
Machlup, F. (1982). The_Branenee_ef_Learning. Princeton

University Press.

Masuda, Y. (1981). Internatlen_Seeietyl Tokyo: IITS.

McClure, R., 8 Patternsan, T. (1976). "Can democracy survive

television?" Journal_of_§9mmunication. 26:23-28.

McDonald, D. G. (1983). "Investigating assumptions of media
dependency research." Qanmunieatlen_£eeeareh. 10:509-
528.

174

McDonald, D. G., 8 Reese, S. D. (1987). "Television news and

audience selectivity, " Jeurnaliem_Quarterly, 64(4): 763-

768.

McGuire, W. (1968). "Personality and susceptibility to
social influence." in E. Borgatta 8 W. Lambert (eds. ),

Handbaak_af_2ersanalitY_ThearY_and_Besearch. Chicagoz

Rand McNally.

McGuire, W. (1989L "Theoretical foundation of campaign." in

R. Rice & c. Atkin (eds ) Bublic.§ommunication
Qampa1gne. Newbury Park, CA: Sage, pp.43- 65.

McLeod, J. M., 8 McDonald, D. G. (1985). “Beyond simple
exposure: Media orientations and their impact on

political processes." communication_nesearcn 12: 3- 34.

McNelly, J. T., 8 Molina, J. R. (1972). "Communication,
traffication and international affairs information in a

developing urban society." Journalism_ouarterlY1

49:316-326.

McNelly, J. T., Rush, R., 8 Bishop, M. E. (1968L
"Cosmopolitan media usage in the diffusion of

international affairs news." Jeurnaliem_Quarterly
45: 329- 332.

Hedin. D. L-. 8 Rose. B. (1991). cognitiye_ESYchalacY-
Orlando, Fl: Harcourt Brace Jovanovich, Inc.

Metzger, G. D. (1983). "Cable television audiences." Jeurnal
' . 23(4): 41-47.

Miller, A. H., 8 MacKuen, M. (1979). ”Learning about the
candidates: the 1976 presidential debates." Publ1e

Opinion_9uarterlY 43: 326- 346-

Mistler-Lachman, J. L. (1972). "Levels of sentence
comprehension in processing of normal and ambiguous
sentences."

Benayier. 11:614-623.

Moshavi, S. 0. (July 1, 1991). "CNN where have all the
viewers gone?" Breageaeting. 120(26):45—47.

Navon, D., 8 Gopher, D. (1979L "On the economy of the

human-processing system. Eeyenelegieal_3ey1eg. 86: 214-

255.

Newman, W. R. (1976L "Patterns of recall among television

news viewers." Public_991n19n_9uarterlY 40: 115-123-

175

Nielsen, A. C. (1983). ' ° e i e
Bep9rt_te_NII_Client_Eeeting§ Northbrook IL: A c.

Nielsen Co.

Nnaemeka, T. I. 0. (1976). "Issue legitimation, mass media
functions and public knowledge of social issues." Ph.
D. dissertation, University of Minnesota.

Nordenstreng, K. (1972). "Policy for news transmission." in

D. McQuail (ed. ) SQ_1Ql9gx_gf_Ma§s_CQmmunication§

Harmondsworth: Penguin.

Norman, D. A., 8 Borrow, D. G. (1975). "On data-limited and

resource-limited processes." gggn1t1ye_E§ychglggy 7: 44-
64.

Ortony, A. (1978). "Remembering, Understanding and
Representation," Cogn1t1ye_fic1enge 2: 53- 69.

Palmgreen, P. (1979). "Mass media use and political
knowledge-" Journalismfuonggzaphs 61.

Palmgreen, P., 8 Rayburn, J. D. (1979). "Uses and
gratifications and exposure to public television: a

discrepancy approach." 99mmnnicatign_ne§eazch. 6(2):
155-180.

Palmgreen, P., 8 Rayburn, J. D. (1985a). "An expectancy-
value approach to media gratifications, In K.
Rosengren, L. Wenner, 8 P. Palmgreen (eds). Mgd1a
Grat1f1cat19ns_Besearch1_9nrzent_2ezspectixes. Beverly
Hills, CA: Sage. pp. 61-72.

Palmgreen, P., 8 Rayburn, J. D. (1985b). "A comparison of
gratification models of media satisfaction."

Communicat19n_nonggraphs 52: 334-346.

Palmgreen, P., 8 Rayburn, J. D. (1983). "A response to
Stanford."99mmun195119n_8esearch 10: 253-257.

Palmgreen, P., 8 Rayburn, J. D. (1982). "Gratification
sought and media exposure: an expectancy value model."

Communican1on_nesearch. 9(4): 561-580-

Palmgreen, P., Wenner, L. A., 8 Rayburn, J. D. (1981).
"Gratification discrepancies and news program choice."

Communicaticn_flesearch 8(4): 451- 478.

Palmgreen, P., Wenner, L. A., 8 Rayburn, J. D. (1980).
"Relations between gratifications sought and obtained:

a study of television news." Cgmmun1ga319n_ne§eazgh.
7(2): 161-192.

176

Palmgreen, P., Wenner, L. A., 8 Rosengren, K. E. (1985).
1‘0 - a O|~ ;‘ ‘i ' at _ It °‘ 9‘_ 1.
Beverly Hills: Sage.

Pedhazur. E. J- (1982). Mult1ple_Regressign_1n_flehayigral
Research1_Erplana112n_and_£redicti_n. (2nd edition).

New York: Holt, Linehart and Winston.

Perfetti, C. A. (1979). "Levels of language and levels of
process." in Laird S. Cermak and Fergus I.M. Craik
(eds ) Lexels_gf_2rgceasing_1n_fluman_nemgr¥-
Hillsdale, New Jersey: Lawrence Erlbaum Associates,
Publishers. p. 161.

Perfetti, C. A., 8 Goldman, S. (1974). "Thematization and
sentence retrieval."

yerbal_flehaxigr. 13:70-79.

Petty, R. E. et. al., (1981)."Persona1 Involvement as a
determinant of argument-based persuasion, Journa1_g£

Eersonalitx_and_agcial_2sxcholns! 41: 847-855-

Posner. M- (1978). Chr9nQmatr1c_Explorafigns_of_M1nd.
Hillsdale, NJ: Lawrence Erlbaum Associates.

Rao, V. R. (1975). "Taxonomy of television programs based on
viewing behavior." J9nrnal_gf_uarkering_sesearch- 12:
355-358.

Reagan, J. (1993). "From "dependency" to the repertoire of
information sources." paper presented in BEA, Las
Vegas, April 1993.

Reagan, J. (1984). ”Effects of cable television on news

use." Journalism_Quarterl¥o 61:317-324-

Robinson, J. P. (1974). "Public opinion during the Watergate
crisis." Communicatign_nesearch 1: 391-405.

Robinson, J. P. (1972L "Mass communication and information

diffusion." in F. Gerald Kline et al. (eds.) Current
- ‘ a - - a o. Beverly

 

Hills, CA: Sage.

Robinson, J. P. 8 Levy, M. R. (1986). Ihg_flg1n_§gnzge1
Learn1ns_fr9m_Ielexis19n_Nsws- Beverly H1118: Sage-

Robinson, J. P. et. al. (1980). "Comprehension of television
news: how alert is the audience? paper presented to the
Association for Education in Journalism, Boston, Aug.
1980.

177

Robinson, M. J. (1975). "American political legitimacy in an
ear of electronic journalism: reflection on the evening

news." in Richard Adler (ed. ) Tolov vision as a Sooial
E9rce1_New_Approa_he§_to_TE_Qriticism. New York:

Praeger, 1975, pp. 101-114.

Robinson, M. J. (1974). "The impact of the televised
Watergate Hearing-" Journal_9f_gommunication 24:17-30.

Robinson, M. J., 8 Kohut, A. (1988). "Believability and the

press." Euplic_opin12n_ouarte2111 52: 174- 189
Roper Organization- (1985) Public_Attitudes_tgward

 

York: Television information Office.

Roper Organization. (1973).
ono_othor_uo§§_noo1o, 1959-1971, New York: Television

Information Office, pp.4-5.

Rosen, J. (May 13,1991). "The whole world is watching CNN."
. 252(18): 622-624.

Rosenberg, M. J. (1956). "Cognitive structure and

attitudinal affect." lQnInnl_Qi_AthIm§l_§nd_$Q§i§l
Eoyonology. 53:367-372.

Rosengren, K. E., 8 Windahl, S., (1972). "Mass media
consumption as a functional alternative." in D. McQuail

(edo) s9c19l9gx_9f_flass_99mmnnications. Harmondsworth.

England: Penguin.

Rossides, D- W- (1976). Ihe_American.§lass_sxsten:An
Introduction_To.Social.Stratificatign. Boston: Houghton

Mifflin Company.

Rothe, J. T., Harvey, M. G., 8 Michael, G. C. (1983). "The
impact of cable television on subscriber and

nonsubscriber behavior." Journal_9f_Adxertising
Rosoozoh. 23(4):15-23.

Rotter. J- B- (1954). Soc1al_Learn1ns_and_Qlinical
Eoyohology. Englewood Cliffs, N.J.: Prentice-Hall.

Rubin, A. M. (1984). "Ritualized and Instrumental Television

Viewing-" Journal_of_gommun1catign 34: 67- 77.

Rubin, A. M. (1983). "Television uses and gratifications:
the interactions of viewing patterns and motivations."

Journal_of_argadcasting- 27(1):37-51.

178

Rubin, A. M. (1981). "An examination of television viewing

motivations." Communicatign_8esear_h- 8(2): 141-165.

Rubin, A. M., Perse, E. M., 8 Powell, R. A. (1985).
"Loneliness, parasocial interaction, and local

television news viewing-” Human.99mmunication_se§sarch.
12(2):155-180.

Rubin, A. M., 8 Perse, E. M. (1987). "Audience activity and
television news gratifications." Qommon1oot1on
Booooroh. 14(1): 58—84.

Sahin, H., Davis, D. K., 8 Robinson, J. P. (1981).

"Improving the TV news." Irish_Brgadcastins_Reyiaw.
11:50-55.

Salmon, C. (1986). "Perspectives on involvement: Current
problems and future directions." In T. Kinnear (ed. ).

Adzances_1n_§9nsumer_8esearch Vol 11 Prove. UT:

Association for Consumer Research.

Salomon. G. (1979) Interact19n_gf_ued1a1_cegnition1_and
Loozn1ng. San Francisco: Jossey-Bass Publishers. p. 90.

Samuelson, M. E., Carter, R. F. 8 Ruggels, L. (1963).
"Education, available time, and mass media use."

Journalism_9narterlx. 40: 491-496.)

Schiller. H- (1984). Inf9rnat1Qn_and_1he_2r1sis_Econ9m¥-
Norwood, N.J.: Ablex.

Schiller. H- (1981) WhQ_Kn9ws1_Infgrma119n_1n_the_ase_nf
tho_£ortnno_§oo. Norwood, N. J.: Ablex.

Schneider, W., 8 Shiffrin, R. M. (1977). ”Controlled and
automatic human information processing: I. detection,

search, and attention.” Esyoholog1ool_noy1ow. 84:1-66.

Shiffrin, R. M., 8 Schneider, W. (1977). "Controlled and
automatic human information processing: 11. Perceptual
learning, automatic attending, and a general theory."

Esychglogigal_seyiew- 84:127-190.

Shingi, P. M., 8Mody, B. (April, 1976). "The Communication
Effects Gap: A field Experiment on Television and
Agricultural Ignorance in India." Commnn1oog1on
Rosooroh 3: 171-190.

Slack, J. D. (1984). "Surveying the Impacts of Communication
Technologies." in B. Dervin and M. J. Boigt (eds. )
° . Vol.5. Norwood.
N.J.: Ablex.

179

Smith. A. (Fall, 1982). "Information Technology and the Myth
of Abundance." Doooo1_§. 1-16

Son, J. et. al. (1987). "Effects of visual-verbal redundancy
and recaps on television news learning." Jogrna1_or

BrQadcasting_8_Electronic_Media. 31(2): 207-216-

Sparkes, V. M. (1983). "Public perception and reaction to
multi-channel cable television service." Journo1_or

Broadcasting- 27(2):163-175-

Sparkes, V. M., 8 Kang, N. (1986). "Public reactions to
cable television: Time in the diffusion process."

Journal_2f_Br9adcasting_8_Electrgnic_Media. 30: 213- -229.

Sperling, G., 8 Melchner, M. J. (1978). ”Visual search,
visual attention, and the attention operating
characteristic." in J. Requin (ed.), Arronr1on_§nd
Eorfiormonoo_111. Hillsdale, N.J.: Erlbaum.

Spitzer, S. P., 8 Denzin, K. (1965) "Levels of knowledge in
an emergent crisis." Soo1a1_£oroo§ 44:234-237.

Stauffer, J., Frost, R., 8 Rybolt, W. (1978). “Literacy,
illiteracy, and learning from television news."

Q9Eflflfli§fi§i§fl.3§§§fll§h 5' 221-232.

Stauffer, J., Frost, R., 8 Rybolt, W. (1981). "Recall and
Learning from Broadcast News: Is print better?" Journal

of_nrgadca§ting- 25(3):253-262-

Stauffer, J. et. al. (1983). ”The attention factor in
recalling network television news" 1onrna1_or
communication. 37:11-26).

Stern, A. (1973). "A study for the National Association for

Broadcasting." cited in M. Barrett (ed.) Tho_2o11r1o§
- . New York: Thomas Y.

Cromwell Co., p. 7.'

Stern, C. (Jun, 7,1993L "NCTA '93: New channels scramble
for Space-" Broadcasting_8_cable pp 34-38-50-

Television Audience Assessment, Inc. (1983). Ih§_flnlii:
Channel_EnxirQnment Cambridge. MA: Author.

Islexision_and_9able_fiagtbgok (1993). vol- 61- Washington
D.C.: Warren Publishing, Inc., p. D-801.

Thorndyke, P. W. (1977L "Cognitive Structures in
comprehension and memory of narrative discourse."

Cognitixe_2§xcnglogL 9: 77-110-

180

Thorson, E., 8 Lang, A. (1992). "The effects of television
videographics and lecture familiarity on adult gardiac
orienting responses and memory. " common1oor1on
Rooooron. 19(3). 346-369.

Thorson, E. et. al. (1985). "Message complexity and

attention to television." Qommon1oor1on_go§ooron.
12(4). 47-454.

Tichenor, P. J., Donohue, G. A., 8 Olien, C. N. (1970).
"mass media flow and differential growth in knowledge."

Public_991n19n_9narterlx- 34 159-170-
Tichenor, P. J., Donohue, G. A., 8 Olien, C. N. (1980).

0! El! 0: L'h. :10: :0' 0nM.: 0! ave

tho_£ro§§. Beverly Hills: sage.

Tichenor, P. J. et. al. (1973). "Community issues, conflict
and public affairs knowledge". In P. Clarke (ed.) Now
Made1s_f9r_Mass.§9mmun19aticn.3essarcho Beverly Hills=

Sage.

Tolman. E- C- (1932). 2nrp9s1xe_nehavior_1n_An1mals_and_uen.
New York: Appleton-Century-Crofts.

Torsvik, P. (1972). "Television and information."

Scand1naxian_Eglitical_Stud1es 7: 215- 234-

Treisman, A. (1979). "The psychological reality of levels of
processing." in L. S. Cermak, 8 F. Craik (eds. )
. Hillsdale, N. J.:

Lawrence Erlbaum Associates, Pub., pp. 301-330.

Tunstall, J. (1983). Tho_flod1a_1n_flr1ra1n. New York:
Columbia University Press.

Tyler, S. R. et. al. (1979). "Cognitive effort and memory."

 
 

ugngzg. 5. 607-617.

Vanneman, R., 8 Cannon, L. W. (1987). Iho_Anor1oan
Poroopr1on_or_olo§§1 Philadelphia: Temple Univ. Press.

Wade, 8., 8 Schramm, W. (1978). "The mass media as sources
of public affairs, science, and health knowledge."

Public_Qpin19n_QnarterlL 33: 197-209-

Waksberg, J. (1978). "Sampling Methods for Random Digit
Dialing."19nrnal_9f_the_Amer19an_Statistical
Won 73: 40-46.

181

Wattenberg, B. J. (1974). Ihe Bea; America. Garden City,
N.Y.:Doub1eday.

Webster, J. (1984). "Cable television' 5 impact on audience

for local news." Journa11sm_guarterly. 61: 419- -422.
Webster. J- (1983) Ihe_1mnnQt_9f_Cable_and_Pax_Ceb1e

I
1 o. 0; 0 ° ‘00 g:

Nat19nal_Ass991at19n_9f_Br9adga§ters Washington. D C-
Webster, J. 8 Agostino, D. (1982). Cable_and_fiay_gab1g

I _-
01 ‘13 z‘ '0 o 4..., ; ; : o. . og1°

Bep9rt_of_the_Broadcast_aesear2n_gen§er. Athens, Chic-

Webster, J. 8 Wakshlag, J. (1983). "A theory of program
choice." Communicatign_3eseareh. 11(4): 27-35-

Webster, J. 8 Wakshlag, J. (1972). "The impact of group
viewing on patterns of television choice." Journal_gfi

Broadcasting 26:445-455.
Weisberq. R- W- (1980). uem2r¥1_Ihgnght1_and_aehaxier. New

York: Oxford University Press.

Westerqaard. J., & Resler. H- (1975) Clasfi_1n_9anitalist
5921:12, London: Heinemann, p.425.

Wenner, L. A. (1985). "The nature of news gratifications.”

In K- E- Rosengren, et. al.. (eds) Ned18_§ratifigations
Reaear2h1_9urrent_2er§nsgtiyes. Newbury Park. CA: Sage-

Williams. P: (1987). IeshnQ19g1_and_commnniea:19n_aehaxior.

Belmont, CA: Wadsworth Pub. Co.

Wilson, C. E. (1974L "The effect of medium on loss of
information." Journalism_nnarterlx 51: 115-

Wiseman, F., 8 NCDonald, P. (1979). ”Noncontact and refusal
rates in consumer telephone surveys."quzng1_Q£

Marketin9_3eaeareh 16: 478-84-

Youn, S. (1993a). "Program type preference and program
choice in multi-channel situation." a paper presented
at 1992 BEA convention (Las Vegas).

Youn, S. (1993b). "The motivational differences in network
news viewing and 24 hour cable news viewing.’I
unpublished paper.

Youn, S. (1989). "Information society: the debates and
issues." Computgzg1sign. Seoul, Korea: Information Age,
Feb. pp.142-151 8 Mar. pp.176-124 (in Korean).

 

 

Il.|. . ...n
fip4fl‘hh4 4 h:
. l I

I
~