.2.-'

$3;

‘4
E

.5,- z
r 3'4 '_ .

I
v
K

a
J .
i-

1‘.

3255566
'5‘ - .9

:w’
.0 P.

’7')

n

L,
*.»3 ha
.k"9"‘*'
SJIEP‘E'
‘ , , .

5‘ g

”1.15% ‘

v?

o

.

Mr:
a...

r

L

,.
O.

.l

.J‘

M:
.P .

;- W?“ E
. :‘l-fiq". J
-.wz~' ,
i .

f
5i

V
‘-

V‘ .1
"gm
:fi‘

$3?“
uyfi§§”%
. ‘AIW-

5%?
z ' j ‘V' ‘
* :$%fikm “ T"

n _2

~ - L
“W “$
‘r 3'6:
”‘6:

m
11

r {L‘w

.( 5: Q!" E
y 4w%gfi%rg%’
@557) e¥§fi~fl

, €141}: lg}

”.3

i F
.« 3’1“?"
f b- 3‘1 .73" l ‘
'4 szEm-A‘ *

~ 'vs‘éfl
v4.44“

flfinuv
‘rzrégéww ,

.
v ~ "-

 

mm IHIIUUINHIIIlllllllllllllIllllllllllllllllllllHill

5, 01409 5644
(HQ?)

This is to certify that the

. . thesis entitled .
[(53) o a/fl; 7/90/3777] /7)¢7 7/ [5759,27 éflllé’ffi
fl7 (771 ab 7% , Lino/a? fl/flz M1

presented by

/L9/~ 197/ #674

has been accepted towards fulfillment
of the requirements for

M 'A" degree in Té/ecommanich‘Oh

WCW’

Major professor

Date 3/78/45”

0-7539 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

UBRARY
Michigan State
University

 

 

 

PLACE N RETURN BOX to roman this checked horn your rocord.
TO AVOID FINES Mum on or More data duo.

DATE DUE DATE DUE DATE DUE

L_|

   
 

  

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

USES OF CABLE TELEVISION AMONG KOREAN STUDENT82IUJ
EXPECTANCY-VALUE APPROACH

BY

No-Kon Heo

A THESIS

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

MASTER OF ARTS

Department of Telecommunication

1994

ABSTRACT
USES OF CABLE TELEVISION AMONG KOREAN
STUDENTS: AN EXPECTANCY-VALUE APPROACH
BY

NO '- Kon H80

This thesis is aimed to examine the uses of cable
television by Korean students. Expectancy-value theory,
integrated with uses and gratifications research, provided a
theoretical framework for the present study. It was
hypothesized that students' expectancy-value judgements
would be strongly related to their attitudes toward cable
television and gratifications sought from exposure to cable
television. Analyses of survey data of one hundred Korean
Students in the United States supported the propOsed
hypothesis. An exploratory factor analysis further revealed
that Korean students watch cable television for instrumental
purposes in that Information was perceived by all three
viewer groups as the primary outcome while Arousal was

perceived as the least salient outcome.

ACKNOWLEDGEMENTS

Special thanks go to David McCarty for serving as
thesis adviser and for providing a lot of advice. Thanks ako

go to Charles Atkin and Thomas Baldwin, without whom this

research would not have been completed.

iii

Table of Contents

List of Tables ...................................... v

Chapter One - Introduction
Overview ....................................... 2

Chapter Two - Literature Review and Hypotheses

Prior Research on Cable Television ............. 8
Uses and Gratifications Research .............. 16
Expectancy - Value Theory ..................... 20

Chapter Three - Methodology

The Sample and Procedure ...................... 40
Measurement ................................... 43
Data Analysis ................................. 49

Chapter Four - Results
Measurement of Variables ...................... 52
Testing Hypotheses ............................ 65

Chapter Five - Summary and Discussion

Summary of Findings ........................... 96
Discussion ................................... 104
References ........................................ 113

iv

List of Tables

Table 1 Characteristics of the sample ..................... 41
Table 2 Television Viewing Motives: Primary Factor Loading
...................................................... 54
Table 3 Mean scores for viewing motives by different viewer
groups ............................................. 56
Table 4 Factor matrix for Expectancy-Value Judgements
(zbiei) of Cable Viewing ........................... 58
Table 5 Mean scores for Composites of Expectancy-Value
judgementswbiei) variables ....................... 60
Table 6 Correlation Coefficients between Attitudes and
Expectancy-Value Judgements (Ebiei) ................ 65

Table 7 Correlation Coefficients between Attitudes and
Composites of Expectancy-Value Judgement Variables

(Ebiei) ............................................ 67
Table 8 Correlations Coefficients of attitude with Zbd,
Zei, and Zbiei ..................................... 68

Table 9 Regression of Attitude on Expectancy-Value
Judgements (2biei), Belief (Ebi), and Evaluation

(Eei) ............................................... 69
Table 10 Regression of Attitude on Expectancy-Value

Judgement Variables ................................. 71
Table 11 Correlation Coefficients of ZGSi with Ebi, Zei,

and Zbiei: cable only .............................. 73

Table 12 Correlation Coefficients of Gratifications
Sought<GSi) with bi, ei, and Expectancy-Value

Judgements (biei) television. (Cable Only).... ...... 74
Table 13 Correlation Coefficients between both Composites of

2biei and ZGSi ....................................... 76
Table 14 Regression of ZGSi on Expectancy-Value Judgements

(Zbiei), Belief (Zbi), and Evaluation (Eei) ......... 78
Table 15 Regression of 2631 on the Composites of

Expectancy- Value Judgements (Zbiei) ................. 79

Table 16 Correlation Coefficients between Students'

Attitudes and Gratification Sought (EGSi) from Cable
Television ........................................... 83

Table 17 Correlation Coefficients between Attitudes and

Composites of Gratifications. (Cable only). ...... 83
Table 18 Correlation Coefficients of Attitude with Intention
to subscribe to cable television ..................... 84

Table 19 Regression of Students' Intention to subscribe to
Cable Television on Students' Attitude toward Cable
Television ........................................... 85

Table 20 Correlation Coefficients of Students' Intention to
Subscribe Cable Television to Gratifications Sought

(2651') . Cable only. . ............................... 87
Table 21 Regression of Students' Behavioral Intention on
Gratifications Sought(2GSi) ........................ 88

Table 22 Partial Correlation Coefficients of Students'
Intention to Subscribe to Cable Television to

Expectancy-Value Judgements (Ebiei) with Control of
Attitude and Gratifications Sought (2681) ........... 90

vi

Chapter I

Introduction

Overview

In recent years American television environment has
undergone dramatic changes with the advent of new
communication technologies, which include cable television.
Since the middle of the 19603 when cable television began to
be viewed as an alternative to the three existing commercial
television networks (Henke, 1974), cable television now has
become a major competitor to networks for claiming viewer's
"attention," "money," and "time." (Becker & Schoenbach, 1983)
and it changed the patterns of television viewing behavior. By
1993, cable television was available to 97 percent of the
total US households and cable penetration was reaching 59
percent of the total television households (TV dimensions,
1994). During this period, on the other hand, the prime-time
audience share for major networks combined was 60 percent,
compared to 87 percent in 1980 (Broadcasting and cable, 1994).
This networks' shrink can partly be explained from the fact
that cable television was perceived by adults to be very
Similar to network television in satisfying traditional
communication needs (Perse and Courtright, 1993). Networks'

2

3

revenue share per audience was already surpassed by cable
television (Electronic Media, 1989). Of more importance is
that declines in both networks' viewing share and revenue
share are expected to continue in the future (Krugman and
Rust, 1993). Moreover, basic subscribers and pay-cable
subscribers watch more television than nonsubscribers (Heeter
and Greenberg, 1987 or Perse and Ferguson,1993). This finding
has important implications for networks in that the actual
television viewing time did not increase during the 19803
(Bower, 1985). One major contributor to the success of cable
television is the greater channel capacity which can
accommodate a variety of viewer interests and gratifications
sought. Scholars from the tradition of economic models have
gone as far as to suggest that the future success of
television hinges on the production of programs well focused
to viewer interests (Waterman and Grant, 1991). So,
understanding of the viewers' needs is very important.

As of 1994, a typical cable television household could
receive 37 channels (TV Dimensions, 1994) depending on the
regions and systems. Some may argue that as long as the
tradition of American political economy of the mass media

industry (Neuman, 1985) coupled with viewers' excessive tastes

4

for entertainment fares (Cerbner, Gross, Signorielli, &
Morgan, 1980) continues reinforcing homogeneity of television
programming to attract as many viewers as possible, the
increased channel capacity does not necessarily guarantee
diversity of television content. Nevertheless, to the extent
that cable television has the potential of offering various
program options to viewers, there is a practical increase in
diversity (Webster, 1986). Viewers, in fact, have a variety of
program options in news and information, children's
programming, religious programming, sports, music, and
minority programming.

As television contents diversify, viewer's involvement in
television consumption becomes more active and selective. For
instance, viewers actively' use remote control device to
"graze" across multiple channels until they settle down on a
specific content or channel that they perceive would result in
maximum satisfaction of television viewing (Morgan, Shanahan,
and Harris). Viewers' active participation also can take the
form of "zapping" which is intended to skip unwanted contents
(Heeter, 1985; Greenberg and Heeter, 1985). The sheer result
of these activities by viewers is greater viewer control

during communication processes. Viewers can control what to

5

watch or not to watch through their own evaluation of a
particular medium or content based on their perceived
utilities of that medium or content in satisfying their needs.
Blumler (1979) suggested that to better understand how
actively viewers participate in the communication, we must go
beyond studying "what" medium or content viewers choose to or
not to watch and must look at "why" they select what they
watch and what rewards they get from the exposure. In fact,
Lin (1993) suggested. that active jparticipation. in. media
consumption enhances gratifications obtained and the effects
of exposure. In a similar vein, Palmgreen and Rayburn (1979)
and Galloway and Meek (1981) maintain that seeking of
gratification may enhance viewers' future media exposure. This
is consistent with the uses and gratifications perspective
which assumes that television viewing is a self—motivated and
goal-directed activity to seek various gratifications. So
viewers are seen as active rather than passive participants in
the communication process (Katz, Blumler, and Gurevitch,
1974) . These active viewers become more active in the new
television environment where content choice options are
diverse (Walker, 1988).

The present study is to investigate how students' seeking

6
of gratifications from cable television viewing can explain
their cable television viewing behaviors. More specifically,
this study tries to examine what gratifications Korean
students in the United States are seeking from cable
television viewing and how this seeking of gratifications is
related to their exposure to cable television. Korean students
are sampled for two purposes. So far, most of television
gratifications research have been done for US college
students. So the first purpose is to see how Korean students
are different from US students in attending to cable
television. The other goal of the study is to demonstrate the

basic tenets of uses and gratification approach widely

replicated in the US to new sample : Korean college students.

Chapter II

Literature Review and Hypotheses

Prior research on cable television

From the early days of development, cable television
became a subject matter because of its envisioned implications
for the television industry and academic settings. For both
groups, assessment of audience exposure to cable television
was an essential part of their research. For instance, the
media and related industries are spending millions of dollar
each year to document, analyze, and project patterns of
exposure. The net result of these efforts is indexed by
"ratings", which sometimes determine the performance of a
particular medium or program. A.C. Nielsen produces standard
audience» measurement data. for' both cable television. and
broadcast networks and stations nationally which includes
ratings, audience size, and simple demographics of television
audiences. Arbitron's Scan America integrates viewing habits
with product purchasing behavior.

These ratings can have only limited value in the complex
and competitive television environment. This concern has often

been addressed by the cable television industry. Virginia

9

Westphal of Viacom Cable suggested that "perhaps operators
should ask whether audience ratings adequately describe the
way cable viewers use their cable networks" (Making Cable,
1986). Gunter (1993) further suggested that "broadcasters and
their program.makers become better informed, not simply about
how neny viewers are watching television, but how involved
they are in what they are watching. Although important, the
traditional television ratings do not necessarily indicate the
degree to which viewers are involved in, and receptive to,
what they are watching."

A related concern arises from the difficulties of
measuring exposure created by the unique characteristics of
new television environment. Increased content options provided
by cable television and audience fragmentation make it almost
impossible to keep track of each individual viewer's
television viewing behavior by traditional diaries and meters
(Webster and Wakshlag, 1985), let alone measure "why" and
"how much" they are involved in the exposure. Ratings data, in
fact, do not report these matters. Moreover, it is believed
that traditional quantitative measurements of cable television
exposure are underestimating the performance of cable

television. And standard measures of audience size make no

10

distinction between viewers in terms of their involvement with
programs because they measure all "viewing" equally. Far more
important are qualitative measures that assess what values
viewers place on a particular medium or content and what
motivational schemes lead to the exposure (Gunter, 1992).
Measurement of audience exposure to cable television is
also important for academic researchers. Most of the research
done by academics centers on cable's impacts on the existing
networks and society, and audiences of cable television. For
example, Kaplan (1978) in telephone interviews found that
respondents, as a result of cable television viewing, reported
less viewing of local network television, reduced use of radio
for news, less often attendance of movies, and decreased
reading of newspapers. Similar result was held by Agostino
(1980) who found that in three or four-station markets, most
local stations experienced serious net loss of audience as
cable brought about more options to choose. Henke, Donohue,
Cook, and Cheung (1984) provided evidence that once
nontraditional national news sources become available, the
market share captured. by' network. news programs declined
substantially. This decline was most obvious among viewers who

had cable longer than among newer subscribers. Local news

1].

viewing was virtually not affected by cable subscription. More
recently, Krugman and Rust (1993,1987), based on Nielsen's
ratings data, found higher penetration of cable television to
be responsible for networks' decline in shares of audiences
and advertising revenues. All these researche implies that
cable television serves as a functional alternative to the
traditional 'media in satisfying the needs of television
viewing.

Other research cable television has examined audiences of
cable television and their reasons for subscribing. Collins,
Reagan, Abel (1983) tried to predict cable subScribership by
the discriminant analysis between subscribers and
nonsubscribers and found lower television use, having fewer
persons in the household, and lower income were related to
higher cable television subscribership. Becker, Dunwoody and
Rafaeli (1983) and Webster (1983), on the other hand, found
that cable subscribers spent more time with television than do
decliners. Agostino (1980) also identified greater
subscription rates were related with larger families. Ducy,
Krugman, and Echrich (1983) attempted to identify factors
distinguishing between basic subscribers and pay subscribers.

They found that higher income, more children in a household,

12

anxi a greater orientation toward movies are more likely to
describe pay subscribers. Krugman (1985), after review of the
previous studies, maintained. that cable subscribers were
younger, wealthier, better educated, and more avid television
consumers than their nonsubscribing counterparts. Yet, despite
their success in distinguishing subscribers from
nonsubscribers, not all the studies in this research tradition
were illuminative due to the inconsistencies in their
findings. umphrey (1988) and Collins et al. (1983) attributed
these contradictions to variances in cable services from
market to market. Sparkes (1983), on the other hand, pointed
to the lack of theoretical base as the more reasonable reason.
These studies of segmenting cable subscribers were based on
media variables and viewers' demographics for analyses and did
not take viewers' social and psychological needs as reasons
for subscribing cable television into account in their
investigations of cable television viewing.

So far, relatively little is known about the reasons
viewers have for cable television viewing. Nevertheless, some
research does provide information regarding reasons peOple
have for attending to cable television. Jeffres (1978) found

that the number one reason for cable subscribing was a greater

13

variety of programming. This was followed by improved
reception of television signals. Krugman and Eckrich (1982)
and Kaplan (1978) found receiving out-of-market stations and
sports programming that viewers would not ordinarily get to be
the primary reason for subscribing cable television. Rothe,
Harvey, and Michael (1983) found movies to be the top reason
for subscribing.

In a somewhat different approach, Jeffres (1978)
questioned how the patterns of psychological status such as
conscious intentions (motives) that people have for watching
television are affected when cable television greatly expands
available content. He outlined motivational schemes of
television viewing into five categories: "media-seeking,"
where viewer seeks the medium regardless of content; "generic
content-seeking," where viewer seeks a particular type of
program; 'progranr content-seeking," ‘where ‘viewer seeks a
particular program; "information-seeking," where the viewer
moves towards some content within a program,- and "mixed,"
where both television content and non-content bases are used
to direct one's viewing.

He anticipated that as viewers become familiar with cable

television, viewers' motivational scheme would change from

14

media-seeking to content-seeking, thus showing greater viewer
selectivity. He found, however, the opposite result that
showed increase in media-seeking motives and decrease in
content-seeking after the introduction of cable television.
Although the author acknowledged the difficulty of
generalizing the results because of measurement problems
involved, this study was the first to consider viewers'
psychological motives for television viewing in examining
cable television viewing. This study, however, still did not
look at different motivational schemes each segment of cable
television audience and non—cable audiences might have when
they attend to television fares. LaRose and Atkin(1988), based
(m1 a "confirmation/disconfirmatien paradigm," attempted to
predict cable subscribership by looking at how viewers'
reported satisfaction with cable television was associated
with their decision to watch or continue watching cable
television” They' found. that satisfaction. variables, when
operationalized as perceived value of cable, customer service,
complaint last year, and expectation met from cable viewing,
were better predictors of intention to disconnect than
demographic and media variables. The results of this study

imply that viewers' initial orientations, such as perceived

15

value and expected utility of cable television in satisfying
their needs, can be important variables of attitude formation
toward cable television which, in turn, serve as intervening
variables in determining viewing decision.

Since little research on cable television has addressed
the importance of individuals' attitudes and needs in
determining exposure, the present study tries to reconcile
this disappointment in a more theoretical framework. The uses
and gratifications perspective assumes that people expose
themselves to cable television to satisfy certain goals or
needs (Katz et al., 1974). Furthermore, they actively evaluate
various media and contents including cable television in an
expectation of gratifications obtained. Importance in this
perspective is its applicability to new television environment
where increased channels expand viewing options, thus
requiring more active participation from viewers. This notion
was addressed by Palmgreen, Wenner, and Rosengren (1985) who
maintained that "a major challenge that confronts uses and
gratifications researchers is that of adapting and molding the
current conceptual framework to deal with new communication
technologies" (p.34). So far, few uses and gratifications

researchers have focused on studying cable television.

Uses and Gratifications Research

The uses and gratifications approach to understanding
media related audience behavior has developed as a departure
from the "powerful media effects." Within this approach, more
attention was given to the power of audience rather than the
mass media institutions and their content. Audience members
are viewed as active agents whose exposure to media and media
contents is motivated by certain internal needs and specific
gratification-seeking motives. This seeking of gratification
is conceived to be a significant determinant of audience
members' exposure to a particular medium or content. In an
attempt to fulfill their gratification expectations, audience
members dictate their selection of media or content from among
various sources of need satisfaction. So media consumption is
viewed as voluntaristic and selective behavior rather than
passive attention to uniformly-patterned messages.

Researchers within this framework are commonly concerned
with "(1) the social and psychological origins of (2) needs,
which generate (3) expectations of (4) the mass media or other

sources, which lead to (5) differential patterns of media

16

17

sources, which lead to (S) differential patterns of media
exposure (or engagement in other activities), resulting in (6)
need gratifications and (7) other consequences, perhaps mostly
unintended ones" (Katz et al., 1974, p.20).

In its simplest form, uses and gratifications research
tries to explain the ways in which reasons for media
consumption or gratification seeking motives influence the
effects of consumption (Katz et el., 1974). Uses and
gratifications research is deterministic, in its approach in
that motivations of gratification seeking are seen as cause of
media consumption behavior in much the same way as basic human
needs motivate other behavior. So, for better understanding
of exposure to cable television, we must not look at what
content options are available, but look at what expectations
they have about those offerings (Becker & Schoenbach, 1983).

Research on uses and gratifications has centered on two
major areas. First, researchers have addressed the role of
social and psychological contexts in generating motivations
for media consumption (e.g., Greenberg, 1974,- Kippax and
Murray, 1980,- Palmgreen and Rayburn, 1979; Rosengren and
Windahl, 1977; Rubin, 1981). The general finding from these

researches is that the gratification seeking process had

18

empirical ties with such variables as age, education, sex,
income, and psychological variables including arousal motives.
More recently, Donohue, Palmgreen, & Rayburn (1987) examined
how social and psychological factors, including need for
activation, interact to produce different life styles and
patterns of media use.

A second major area of study includes typological studies
of media gratifications and their relationship to measures of
media or content exposure (e.g., Becker and Fruit, 1982;
Kippax and Murray, 1980; McLeod and Becker, 1974; Palmgreen
and Rayburn, 1982; Rubin, 1982, 1983; Wenner, 1983). These
studies commonly showed that gratification seeking is related
to medium and content choices. For example, Kippax and Murray
(1980) and Becker and Fruit (1982) consistently found that
audience members' comparisons of gratifications obtained from
different media are related to choices of medium or content
among various alternatives.

Some other studies have investigated the relationship of
gratification seeking to content choice within a single
medium. Rubin and Rubin (1981) found that viewing of daytime
serials and game shows was associated with seeking of

companionship, relaxation, arousal, habit, and passing the

19

time, while viewing television news, documentary magazines,
and talk shows was associated most strongly with an
information-learning motive.

Even though many of these studies have found significant
relationship between gratification seeking and exposure to
mass communication, the relationship still ranged from "low to
moderate correlations." (Palmgreen et al., 1985, p.30). More
recent studies even reported that gratification seeking is
weakly related to exposure (Rubin, 1985; Perse, 1986; Wenner,
1986; Babrow, 1986). Barbow (1986), for example, showed that
gratification seeking has been found to explain 8% to 10% of
the variance in levels of exposure to media. These findings
may have been related to lack of theoretical rigor in the
perspective claimed by Blumler (1979) and Elliott (1974).

This criticism resulted in the introduction of various
theories from other disciplines and theoretical integrations
and applications within the framework of uses and
gratifications. An approach that has special promise for
advancing understanding of the influence of gratification
seeking on exposure consists of research that integrates
expectancy-value theories with the uses and gratifications

framework.

Expectancy — Value theory

Expectancy-Value theories are based on the proposition
that a person's behavior is the result of her perceived
probability (expectancy or belief) that an object possesses a
particular attribute or that behavior will have a particular
consequence weighted by evaluation-that is, the degree of
affect, positive or negative, toward an attribute or
behavioral outcome (Fishbein and Ajzen, 1975, 1967; Galloway
and Meek, 1981; Palmgreen and Rayburn, 1982, 1984; Barbrow and
Swanson, 1988, 1989; Van Leuven, 1981). This theory assumes
that human behavior is 'volitional' and 'goal-directed' rather
than 'capricious' and also assumes that people consider the
implications of their actions before they decide to engage or
not engage in given behavior.

Under' this theoretical. umbrella, Fishbein. and..Ajzen
(1980) have proposed a "theory of reasoned action" that
Specifies in a mathematical way the relationships among
beliefs, attitudes, and behaviors. According to this theory,
behaviors are directly caused by behavioral intentions to

perform (or to not perform) those behaviors. So, behavioral

20

21

perform (or to not perform) those behaviors. So, behavioral
intention is regarded as the single most important predictor
of behavior.

Measuring a person's behavioral intention is not always
easy, however. Fishbein and Ajzen(1980) suggested that a
person's behavioral intention is a function of her individual
and social factors. The individual factor, called attitudes,
refers to the positive and negative evaluation a person holds
toward performing the behavior. People usually differ in their
attitudes. We may assume that these differences will lead to
different behavioral consequences. The social factor in
determining behavioral intention refers to the subjective norm
that is the person's perception of social pressure put on him
to perform or not to perform the behavior in question.

In sum, a person is very likely to engage in a particular
behavior when she judges the behavior positively and when
people important to her think she should perform the behavior.
Attitudes and subjective norms are identified as independent
variables which indirectly affect final behavior through their
effects on behavioral intentions. In general, this model has
a recursive-chain causal structure where the preceding

variable becomes the antecedents of the following variable

22

(Liska, 1984). Fishbein and Ajzen's(1975) model deserves
further explanations for its influences in gratifications

study.

1 n Ev luati n r tifi ion

Because of its information-processing assumptions and its

goal-directed and volitional view of an audience in the
communication process, many uses and gratifications
researchers have attempted to adopt this theory in their
audience research.
For‘ instance, Galloway and Meek (1981) proposed a basic
expectancy-value model adapted for measurement of audience
uses and gratifications, and they suggested that a person is
likely to expose herself to a particular content (or medium)
if the gratifications sought from television viewing are both
highly expected and valued.

In the similar view, Palmgreen and Rayburn (1982) based
on the Fishbein and Ajzen's (1975) aforementioned attitudinal
and behavioral intention model, postulated that media exposure
is the result of attitudes toward the media object and the

seeking of various gratifications from the viewing experience.

23

positive and negative feelings about engaging in behavior such
as viewing a particular medium or content (Fishbein and Ajzen,
1975), are the function of one's beliefe about viewing a
medium or a content and one's eveleeeive reepenees.

In this way, students' attitudes toward cable television

viewing can be calculated from the following formula:

n

2 biei

i=1
where bi refers to the belief (or perceived probability) that
cable television 'viewing results in. a consequence of i
(obtaining of gratification i), and ei refers to the affective
evaluation of the consequence 1'. Thus a person's attitude
toward cable television viewing may be estimated by
multiplying each salient belief by evaluation and summing the
products over the domain of all relevant gratifications;

According to Fishbein and Azjen (1975), attitude is a

generalized predisposition to act in a consistently favorable
or unfavorable manner toward some object and does not
necessarily predispose a person to perform any Specific

behavior. Attitude, however, is related to behavioral

24

intentions, which is considered to be a direct cause of
behavior.

This model has drawn empirical support from Palmgreen and
Rayburn (1982) in a study of television news viewing. Strong

correlation. of .54 (p<.001) was found. between students'

attitudes toward television news and Stdei products. This

result was replicated by Barbrow and Swanson (1988), who found

a strong relationship between college students' attitudes and

their expectancy-value judgements (2biei) of television news

viewing.

By the same token, Palmgreen and Rayburn (1982, 1984)
further postulated that seeking of gratifications (GS) from a
particular medium or content is a function of both the beliefs
(expectancy) a person holds about exposure to a media or
content and the affective evaluations a person attaches to

that exposure. Accordingly, we can calculate students' general

tendency to seek various gratifications (2631') from cable

television viewing from the formula as follows:

I]
2 biei

i=1

25

where bi is the beliefs (expectancy) that a particular media
source (e.g. cable television) provides a gratification i or
beliefs that exposure to cable television results in
fulfillment of gratification i, and ei is the affective
evaluation of the gratification i. From this reasoning, it is
logical to expect a person to participate in cable television
viewing through the active seeking of gratifications as long
as she thinks that seeking is good and believes that the cable
television viewing will result in obtaining of what she
sought.

.Although the same formula is ‘used. to jpredict both
attitudes toward cable television viewing and gratification
seeking behavior, it does not necessarily mean that attitudes
and gratification seeking are tautological concepts. According
to Palmgreen and Rayburn (1982), attitudes and gratification
seeking are two distinct generalized orientations. However,
they proposed that the antecedents (biei) of attitudes toward
cable television are also the antecedents of gratifications
sought from the cable television viewing. 80, biei is supposed
to predict both attitudes and gratification sought (GSi)
independently (p. 567).

This ‘model further' posits that strong' beliefs

26

(expectancy) or strong evaluation alone do not necessarily
lead to a strong seeking of gratifications in a particular
direction. For example, a person may perceive that being
entertained by viewing television or a particular program
would be very attractive but may not necessarily lead herself
to active seeking of that gratification if her expectation
about the television's or the program's ability to entertain
her is low. Moreover, the same Ibehavior (viewing cable
television) may lead to 'many different consequences to
different people. So, the two components, beliefs (expectancy)
and evaluations, need to be combined and then summed over all
gratifications.

Based on this reasoning and empirical support, the

following hypothesis is formulated:

Hla: There will be a positive relationship between student's

attitudes toward cable television viewing and the

strength.of their expectancyevalue judgements (2biei) of

exposure to the cable television.

27

Hlb: When Zbi and Zei are allowed to be related to the

attitudes independently, the strength of the
relationships will become weaker (or nonsignificant) than

the strength of relationship between attitudes and

combined measure of Zbiei.

Even if positive attitudes do not always guarantee media
exposure, we may, however, expect that students who value a
particular gratification sought from viewing cable television
highly and believe cable television has something to offer
that gratification are more likely to subscribe to cable
television. In reverse, students whose beliefs (expectancy)
and evaluation of cable television are low are less likely to
subscribe to cable television services. This tendency, for
example, will be most prominent among nonsubscribers who may
have been dissatisfied with the existing cable services or
among students who are actively exposing themselves to other
media alternatives. For the latter students, the costs of
obtaining the information they seek from cable television may
be perceived to be more than the costs of obtaining the same

information (i.e. comparable gratifications) from other media

28

sources (Palmgreen and Rayburn, 1982, 1984).

Previously, it was suggested that a person's high
expectancy (beliefs) and evaluations of media source (2biei)
result in active seeking of multiple gratifications(ZGSi) from

the media source. This formulation, in turn, suggests:

H2a: There will be a positive relationship between students'

seeking of various gratifications(2GSi) from cable

television viewing and the strength of students'

expectancy-value judgements (Zbiei) of exposure to the

cable television.

H2b: When Ebi and 2ei are allowed to be related to the ZGSi

independently, the strength of the relationships will

become weaker (or nonsignificant) than the strength of

the relationship between ZGSi and combined expectancy-

value judgements (Zbiei) .

If Zbiei is correlated to both attitudes and

29

gratifications seeking measures (2651), then it is reasonable

to expect:

H3: There will be a positive relationship between attitudes

and 2631' of cable television viewing.

In sum, students who believe strongly that viewing basic
cable television or a particular channel (e.g., CNN) is very

informative or students who believe strongly that viewing a

movie channel (e.g., HBO) is entertaining may feel very
positive toward. these attributes (e.g., entertaining' and
informative). Students' expectancy-value judgements (Zbiei)

then become antecedents of both generalized orientation to
seek various gratifications from.cable television viewing (or
pay-channels) and attitudes toward those media sources.
Palmgreen and Rayburn's (1982) model has been subject to
empirical tests by subsequent studies (Palmgreen and Rayburn,

1982, 1984; Barbrow and Swanson, 1988) to examine the

relationships among variables of Zbiei, EGSi, attitudes, and

television news viewing. One common finding in these studies

30

was that expectancy—value judgements (Zbiei) did not directly

predict television viewing. This may have been so because of

the model's recursive-chain causal structure, where attitudes

or 2631 were mediating the effects of 2biei on the television

viewing intention which, according to Fishbein and Ajzen
(1975), is the direct cause of television viewing behavior.

It also explains why the predicting power of attitudes and

EGSi for media exposure was higher than Zbiei (Palmgreen and

Palmgreen, 1982). The stronger the relationships of Zbdei to

attitudes and 2031 are, the weaker the relationship between

2biei and cable television viewership will be.

 

Although measures of attitudes and ZGSi predicted

television news viewing behaviors, the correlations were low
to moderate (Palmgreen and Rayburn,1982). These relatively
nonsignificant relationships assume that there may be other
intervening variables (Barbrow and Swanson, 1988). Indeed,

Palmgreen and Rayburn (1982) argue that "the various media

31

consumption studies show low to moderate correlations (.15 -
.40) between. the: gratifications ‘measures and consumption
indices." (p.30).

One possible explanation may be that measurement issues

can effect the ability of attitudes (or ZGSi) to predict

behavior (e.g., television exposure) (Davison and Jaccard,
1979). For example, Palmgreen and Rayburn's (1982) definition
of attitudes toward television news viewing as satisfaction
with the news might have caused poor prediction of attitudes
for exposure. Earlier, Fishbein and Ajzen's model (1980)
suggested that the immediate determinant of a person's overt
behavior was that person's intention to perform (or not to
perform) that behavior, and that a person's behavioral
intentions could be predicted indirectly by measuring a
person's attitude and subjective norm. Subjective norms refer
to beliefs about the Ibehavioral expectations of others,
weighed by the motivation to conform to them. In other words,
subjective norms refers to a person's perceptions of the
social pressures to perform or not perform a particular
behavior in question.

Barbrow and Swanson (1988) cautioned to distinguish

32

subjective norms from various social-structural forces such as
availability of or accessibility to cable television that
might also influence a person's exposure to a media source.
Palmgreen and Rayburn (1982) also suggest that the behavioral
influence of significant others may effect media exposure.
For instance, they found that among those respondents who
usually deferred viewing decisions to other household members,
viewing behavior of the decision maker predicted television
viewing much better than gratifications sought and obtained
(1979). Similarly, Blumler (1979) also emphasized the
normative influences on what individuals aim to get out of
media exposure to the media source.

Barbrow and Swanson (1988) also reasoned that normative
perceptions might play a key role in determining audience
exposure levels and tested it empirically in students' soap
opera viewing in which self-concept, a normative predictor,
was the single strongest predictor of viewing intentions.
This notion of normative influence on communication behavior
would be very important in understanding the patterns of
family communications or group communications, especially in
the new video media environment where cable television and

video cassette recorders provide audience members with

33

multiple program choices, including adult programming. In
such an environment, normative influences can take various
forms such as "mediation", "rules", "prohibitions", and
"discussions" (Atkinq Greenberg, and. Baldwin, 1991).
Nevertheless, gratification researchers have not examined this
notion extensively.

According to Fishbein and .Ajzen's (1973, 1980)
contention, subjective norms can be measured by a person's
normative belief --her expectations that significant others
endorse performing the behavior-- and her motivation to comply
with significant others.

This model also specifies that a person's subjective norms,
combined with her attitudes toward an act determine behavioral
intentions that are the direct cause of overt behavior. Both
Fishbein and Ajzen (1975) and Liska (1984) agree that
behavioral intentions predict overt behavior to the extent
that behavior are volitional rather than habitual, skillful,
or cooperative. Accordingly, behavioral intention of
television viewing may be a poor determinant of television
exposure for students whose television viewing behavior is
influenced by habitual and time-consuming activities.

Moreover, Fishbein (1973) suggests that behavioral prediction

34

can be improved if we can measure intentions that a person
holds for other alternative courses of action. For example,
even though a student's intention to subscribe to cable
television is very high, she does not necessarily subscribe to
cable television if her intention regarding other sources of
media (e.g., newspapers, VCR, or magazines) is much higher.
In other words, students frequently do not do what they intend
to do, but are only constrained by a lack of resources and
opportunities (Liska, 1984). In general, however, volitional
or goal-directed behavior is best predicted by intention.

In sum, Fishbein and Ajzen's (1975) "theory of reasoned
action" specifies that the single best predictor of cable
television viewing (or subscribing) is the intentions students
hold regarding that exposure. These intentions are viewed as
a function of two other factors: (1) students' attitudes
toward cable television viewing, (2) students' subjective
norms with respect to the exposure. Students' attitudes about
the exposure is a function of their beliefs (expectancy) about

the exposure or perceived likelihood that exposure will result

in gratifications obtained Lbi) and the evaluations of

obtaining gratifications (ei). The subjective norms are a

35

function of the students' perceptions that Significant others

think the exposure should or should not be performed and the

student's motivations to comply with these others. Based on

the reasoning discussed above, we can expect that:

H4:

H5:

H6:

There will be a positive relationship between students'
attitudes toward cable television viewing and students'

intentions to subscribe cable television.

There will a positive relationship between students'

seeking of various gratifications (ZGSi) from cable

television viewing and students' intentions to subscribe

cable television.

The relationship between students' expectancy-value

judgements (Zbiei) and their intention to subscribe to

cable television will become weak (or nonsignificant)

when attitude and gratifications sought (ZGSi) are

controlled constant.

The reasoning behind this prediction relates to the mediating

36

effects of gratifications sought and attitude between
expectancy-value judgements and intention. Tests of this
hypothesis will provide interesting information for comparing

it to Triadis' (1980) contention that expectancy-value

judgements (Zbdei) may independently affect intentions and

behaviors.

H7: There will be a positive relationship between students'
perceptions of subjective norms and their intentions to

subscribe cable television.

H8: There will be a positive relationship between students'
intentions to watch cable television and their cable

television viewing.

A summary diagram of students' attitude-behavior model of

cable television viewing will be like this:

37

Expectancy-Value Model of Gratification Sought from Cable

Television Viewing: adapted from Fishbein and Ajzen

 

 

(1975).
Zbiei ’//////a\ HI-—————aB
\zcsi‘ /
SN
bi = the beliefs (expectancy) that cable television viewing

will result in a gratification i.

ei

the affective evaluations of the gratification i.

A

attitudes toward cable television viewing.
EGSi = a general orientation to seek gratification i from

cable television viewing

SN subjective norms

BI = behavioral intentions to subscribe to cable television

II!
II

subscribing cable television

Besides the suggested hypotheses, the present study is

intended to examine other relationships between variables.

38

For instance, contrary to the Fishbein and Ajzen's (1975)
specifications of the relationships among variables, Liska
(1984) concluded that even if behavioral intentions were
better predictors of behavior than were attitudes, the direct
effects of attitudes were still substantial. In the same
corollary, Liska argues that beliefs alone may affect
behavioral intentions and behavior independently of the effect
of attitudes. In other words, attitudes do not completely
mediate the effect of beliefs on intentions and behavior,
because human information processing, linking beliefs to
attitudes, is inexact and inefficient. Liska further reasoned
that attitudes and subjective norms may not be independent.
They both appear to represent similar beliefs and to influence
each other. And most interestingly, contrary to the Fishbein
and Ajzen's (1975) assumption of recursive-causal structure of
the model, Liska suggests that behavior may affect both
intentions and attitudes. From this line of thoughts, the

present study raises the following questions:

RQ1 : Are attitudes related directly to behaviors?
RQZ : Are beliefs alone related directly to behavioral

intentions and behaviors?

Chapter III

Methodology

39

The Sample and Procedure

The sample in this study was drawn from a population of
all the Korean students enrolled at Michigan State University
in 1994. There were 381 Korean students listed in the
directory of Korean Student Organization. This number includes
students both in undergraduate and graduate levels and
students in English Learning Center. They are unevenly
populated both in Lansing and East Lansing, Michigan, areas
with the majority (90 percent) living in East Lansing area.
Twenty-five percent of the total students are living in on-
campus resident halls and 65 percent are living in three
different university apartments. Given the relatively small
size of sampling frame, the researcher tried to reach as many
students as possible. Thus the sample is purposive, not
randomly selected. Students were contacted by either telephone
calls or personal contacts and, as a result, a total of 100
students agreed to participate in the research.

The sample consisted of 64.2% of male and 35.8% of
female. The majority (69.1%) of students were ranged in age

between 25-34 and 12.4%, 9.3%, 9.3% were between 20-24, 18-19,

40

41

and 35-44, respectively. 91.7% of students were identified as
full-time student, and 8.3% as part-time student. 62.9% were

students.

married,

students, 40% had 1

and 37.1% were single

child, 34.3%,

and 3 children respectively.

 

 

 

 

 

 

 

Tdflel.
Characteristics of the sample

Sex Male 61 (64.2%)
Female 34 (35.8%)

Age .18-19 9 ( 9.3%)
20-24 12 (12.4%)
25-34 67 (69.1%)
35-44 9 ( 9.3%)

Status Full-time 88 (91.7%)
Part-time 8 ( 8.3%)

Marital status Single 61 (62.9%)
Married 36 (37.1%)

No. of children 0 12 (34.3%)
1 14 (40.0%)
2 8 (22.9%)
3 1 ( 2.9%)

Grade Freshman 13 (13.5%)
Sophomore 2 ( 2.1%)
Junior 5 (5.2%)
Senior 9 (9.4%)
Graduate 67 (69.8%)

Subscribership Basic only 54 (54.0%)
Pay 9 ( 9.0%)
Non-cable 37 (37.0%)

 

 

 

 

Among married

and 2.9% had 0,

 

42

Seventy percent of the sample were attending graduate
program, and 13.5% were freshman, 9.4% were senior, 5.2% were
junior, and 2.1% were sophomore. 63% of respondents were
subscribing cable television and 37% were not. Among those
respondents getting cable television, 86.2% were basic only
and 13.8% were pay subscribers.

The data were collected over a three-day period in
October 1994. A survey questionnaire was administered to 100
Korean students without rewards for participation. Students
were briefly instructed by the researcher about the purpose of
the study, anticipated length of time to finish, and the
proper" procedures to answer' the questionnaire. The

questionnaire was organized into eight major parts.

43

Measurement

Preliminegz geeeeiene (6 items). The first section of the
questionnaire (item 1-6), starting with a short introduction,
was a series of screening questions which included location of
residence (1: on-campus, 2= off-campus), ownership of a
working television set (1: Yes, 2: no), availability of cable
television (1: yes, 2: no), subscribership (1: yes, 2: no),
and number of pay services. If a respondent did not have any
working television set, he or she was instructed to quit
answering the questionnaire , on the assumption that they did
not watch television. Students who responded that they
received pay channels were asked to provide the number of pay

channels they received.

44

fireeifieegiene eegght free televieien viewing in generel (23
items). One of the most difficult problems facing scholars
conducting empirical researches on the gratifications audience
members seek from the media is measurement of the
gratifications themselves (Blumler, Gurevitch, and Katz,
1974). But how to measure gratifications is a core part of
uses and gratifications study (Palmgreen et al., 1985), as it
is in this study, as well. Becker (1979) suggests three
ways of gratifications measurement. First is to infer what
gratifications audience members are seeking based on
measurement of some separate, yet related 'variable. For
instance, inferences can be made about the informational needs
of audience members--and gratifications sought--based on such
variables as sex and age. The second method is to rely on
reports from the audience members. For this strategy,
respondents are usually provided a list of gratifications--
developed from earlier research--and asked to indicate which
of the gratifications applied to them. Finally, gratifications
can be 'measured. by ‘manipulating in field or laboratory
settings the gratification subjects have upon receipt of

various communication messages. In a simple experiment, for

45

example, subjects could be instructed to pay attention to a
given message for a specific reason somehow related to the
purpose of the experiment. These subjects then could be
compared to others having been given different instructions or
gratifications to learn of the gratifications on subsequent
behavior.

To measure gratifications students seek from television
viewing in this study, students were asked to rate on a five-
point scale their degrees of agreement on 23 reasons for
watching television in general (1: strongly agree, 5: strongly
disagree). For instance, respondents were asked to estimate on
a five—point scale how much they agreed with the gratification
statement, "I watch television. to learn something' new."
Twenty-three gratifications items were borrowed from the
earlier works done by Frank and Greenberg (1979) and Rubin
(1980).

These items were evaluated for all students
participating, regardless of whether they were receiving cable
television. The gratifications items were subject to factor
analysis to examine meaningful dimensions of television
viewing. Exploratory factor analytic methods have been the

prevailing tools in gratifications research (Dobos and

46

Dimmick, 1988). To examine interrelationships among factors,
oblique rotations and interfactor correlations methods were
employed. The earlier works done by Frank and Greenberg and
Rubin found five: dimensions: Diversion, Information,
Entertainment, Special Programs, and Arousal. The results of

the factor analysis will be compared with these dimensions.

Pr Pr r (20 items). The third section. asked
students what kind of program they preferred to watch. The
five-point scale ranged from very often (5) to very seldom
(1). Program types for the preference questions were based on
the A.C. Nielsen's categorization (1989).

Nielsen categorized programs into twenty types, including talk
shows (e.g., Larry King), the national evening news (e.g.,
CNN, ABC), the local news, morning news show (e.g., Today),
quiz and game shows, religious programs, feature programs,
general dramas (e.g., Matlock), comedies (e.g., Cosby), action
and adventure (e.g., MacGyver), do it yourself (e.g.,
Hometime), sports, travel shows or documentaries, hobbies,
music, outdoors, health (e.g., Getting Fit), home shopping,

and children's program.

47

Televieien (2 items) and cher Media Egposure (4 items). The

program preferences items were followed by items measuring
television exposure. Students are asked to rate on a five-
point scale the number of television viewing hours both on a
weekday and weekend, followed by the number of hours spending
reading newspapers and listening radio in a typical day.
Students are also asked how many movies they rent on video
cassette and how many movies they attend at a theater in a

typical week.

Eveleatiene (ei, 23 items). The fifth section asked students
to evaluate how important for cable television to have each of
the gratification items. For example, students were asked to
rate on a five-point scale (1: very good, 5: very bad) whether
it is good or bad for cable television to have such
characteristic as "makes me feel less lonely." These questions
were asked to all the students regardless of subscribing cable
television in the hope that different viewer groups might have

different perceived values for cable television viewing.

Beliefe (bi, 23 items). The sixth section addressed students'

perceived probability that cable television would provide them

48

with each of the same characteristics mentioned above. The
five-point scale ranged from "very definitely (1) to very
definitely not (5)." The expectancy - value scores were
obtained by multiplying measures for beliefs and evaluations
and then summing across items. So the possible scores ranged
from 24 to 600. Here again, all the students were asked to
rate their' perceived. probability' of cable television. in
satisfying each of the gratifications items. Students,
regardless of receiving cable television, were believed to

have different belief systems regarding cable television.

Aggieugee (3 items). To measure students' attitudes toward
viewing cable television, three bipolar attitudinal items
adapted from Fishbein and Ajzen (1980: appendix B) were used
(very good: 1, very bad: 5, very wise: 1, very foolish: 5,
very beneficial: 1, very harmful: 5). For a composite score of
attitudes, scores on each three items were summed with

possible range of 3 to 15.

Segjee5i1e_gegee (1 item). A single item adapted from Fishbein
and Ajzen (1980: Appendix B) was used to measure students'

subjective norms. Students were asked to rate on a five-point

49

scale ranging from very likely (1) to very unlikely (5) that
most people important to them. (e.g., family' members or

friends) think they should subscribe to cable television.

Beeevierel In;ee§ione (1 item). A single item behavioral item
was adapted from Fishbein and Ajzen (1980: Appendix B) and
asked students whether they intended to subscribe to cable
television or to continue subscribing. The five—point scale

ranged from very likely (1) to very unlikely (5).

Qegeggepgieee Items for demographics included students' age,
sex (1: male, 2: female), status in the university (1: full-
time, 2: part-time), number of children in the family, grades
(1: freshman, 2: sophomore, 3: junior, 4: senior, 5:

graduate), marital status (1: married, 2: single), and race.

Data Analysis

The major analytical tool used in this study was
exploratory factor analysis to identify underlying dimensions

of both television viewing motives in general and expectancy-

50

value items of cable television viewing. The purpose of this
was to examine possible discrepancies in students'
motivational schemes when viewing television in general and
cable television.

The second analytical technique was "Pearson-Product Moment"
correlations to examine relationships between variables in
hypothesized patterns. Finally, to further examine the
relationships between variables, regression analyses were

performed.

Chapter IV

Results

51

Measurement of variables

Televieien Viewing Me§ive§iene

To determine motivational schemes of television viewing in
general, a varimax-rotated principal components analysis of
responses with iterations on 23 gratifications items was
performed using Statistical Program for Social Science (SPSS).
.A varimax- rotation was used because of correlations among
various motives for television viewing (Rubin & Perse, 1987).
Exploratory analysis of students ratings on television viewing
in general produced six factors with minimum eigen values
greater than one. These factors explained 65.4% of the total
variance. Gratifications seeking statements with a factor
loading of .5 or greater were used in identifying and labeling
the factors.

Factor 1, labeled Escape, explained 28.0% of the total
variance after rotation. This factor included a total of six
statements (Crocbach Alpha:.85) such as "I watch television to
forget my problems for a while" and "to escape from the

reality of everyday life." Factor 2, labeled as Arousal,

52

53

statements (Cronbach Alpha:.70) including "I watch television
to get a feeling of adventure and excitement" and "to watch
uncensored programs." Factor 3, labeled as Relaxation,
explained 8.7% of the total variance. Five statements
(Cronbach: .80) were included in this factor such as "I watch
television because it relaxes me" and "to release tension."
Factor 4, labeled as Information, explained 5.6% of the total
variance. This factor consisted of three statements (Cronbach
Alpha:.71) such as "to learn something new," "to obtain useful
information for daily living," and "to learn more about what
is going on in the world." Factor 5, labeled as Pass—Time,
consisted of two items (Cronbach Alpha:.64) explaining 5.2% of
the total variance. Those are items such as "because it passes
the time away, especially when I am bored" and "to spend time
with others while they are watching." Factor 6, labeled as
Special Programs, explained 4.5% of the total variance. It
contained only one item "when there is something on I want to
see." So, it was difficult to provide meaningful
interpretation for this factor. Accordingly, this factor will
be exempted from subsequent analyses hypotheses. The summary

of factor matrix is in table 2.

Table 2

Television Viewing Motives: Primary Factor Loadings

54

Factor 1: Eacapo

Iakoa mo fool looo lonoly

"hon thoro io no ono oloo to

To Havo omotional oxporioncoo
Somothing to talk about with othora
Forgot my problomo for a whilo
Eocapo from tho roality of lifo

Factor 2: Arouaal

It is thrilling

Advonturo/ Exitomont

For companionohip

Witch unconoorod film

Bo intorooting and otimulating
It popo mo up

Factor 3: Rolaxation

It amuaoo mo

It rolaxoo mo

It is onjoyablo

It ontortaino mo

It roloaoo tonoion

Factor 4: Information

To loarn oomothing now

To obtain uooful information
To loarn moro about tho world

Factor 5: Paoa-timo

Paoa tho timo away whon horod
Spond timo with othoro

Factor 6: Spocial Programs

Somothing on I want to ooo

.73
.68
.72
.61
.80
.76

.16
-.09
.26
.07
.44
.38

.04
.13
.08
.21
.52

-olo
.13
.01

.22
.33

-o18

.07
.02
.35
.32
.09
.08

.65
.64
.49
.69
.63
.59

.13
.05
.20
.02
.11

.06
.30
.01

-.00
.22

.06

.38
.04
.05
.15
.07
.21

.18
.22
.28
.03
.05
.00

.63
.78
.77
.57
.56

.08
.05
.15

.25
.04

.11

-.18
-.15
.16
.24

.03

.10
.41
-.08
-.06
.26
.15

.40
-.01
.22
.51

.74
.71
.77

.00
.11

.35

.16
.38
-.09
.18
.16
.07

-.25
.16
.07
.19

-.03
.08

.06
.07
.08
.09
.09

"ozs
.02
.26

.76
.63

'o07

-.01

.13
-.00
-.02
-.08
-.10

-.23
-.11
.38
.12
-.04
.26

-.42
.26
.05

-.16
.03

.21
.06
.13

.17

.72

55

For the discriminant analysis among different viewer
groups, each motivational factor was compared in terms of
means and Standard deviations. For the entire sample of
respondents, Information was the most salient motive for cable
television viewing with a mean score of 2.40 (SD: .81)
follOwed by Relaxation (M: 2.64, SD: .72), Special Programs
(M: 2.73, SD: 1.08), Pass—Time (M: 2.93, SD: .98), Escape (M:
3.31, SD: .87), and Arousal (M: 3.35, SD: .68).

The same procedures were employed to different viewer
groups: basic-only, pay-subscribers, and non-subscribers. The
complete matrix of comparision among each viewer groups is
summraized in table 3.

Information was the most salient motive with mean score
of 2.44 (SD: .89) among basic-only subscribers, followed by
Special Program (M: 2.64, SD: 1.15), Relaxation (M: 2.74, SD:
0.72), Pass-Time (M: 3.10, SD: 1.02), Arousal (M: 3.39, SD:
.74), and Escape (M: 3.42, SD: .85).

For pay-subscribers, Information was most salient motive
with a mean score of 2.15 (SD: .65), followed by Relaxation
(M: 2.27, SD:.62), Pass-Time (M: 2.61, SD: .55), Special-

program (M: 2.78, SD 1.09), Escape (M: 2.94, SD: .98), and

Arousal (M: 3.00, SD: .61).

56

Table 3

Mean scores for viewing motives by different viewer groups

Motives Basic—only pay-subs non-subs
Escape 3.42 (.85) 2.94 (.98) 3.20 (.83)
Arousal 3.39 (.74) 3.00 (.61) 3.37 (.57)
Pass-time 3.10 (1.02) 2.61 (.55) 2.69 (.95)
Relaxation 2.74 (.72) 2.27 (.62) 2.57 (.69)
Special-Program 2.64 (1.15) 2.78 (1.09) 2.94 (1.01)
Information 2.44 (.89) 2.15 (.65) 2.44 (.72)
m... ""55; """"" 2 "6'2 """" 2' 1:3,"

Note: All measures could range from a low of 1 to 5 and lower
scores mean more salient motivation.
Inter—coding reliabilities (Crobach alpha) for each
motives were.85(Escape), .70(Arousal), .80(Relaxation).
64(Pass-Time), and .71(Information).

Television was viewed by those students who did not
subscribe to cable for the purpose of Information with a mean
score of 2.44 (SD: .72), followed by Relaxation (M: 2.57, SD:
.69), Pass-Time (M: 2.69, SD: .95), Special-Program (M: 2.94,
SD: 1.01), Escape (M: 3.20, SD: .57), and Arousal (M: 3.37,
SD: .57). The result shows no significant differences among

three viewer groups in terms of cable television viewing

motivations.

57
WHZLH ei
Before measuring students' expectancy-value judgement

(Zbiei) of cable television, expectancy—value scores were

calculated for each of the 23 gratifications items by
multiplying students' beliefs (perceived probability of cable
in satisfying each suggested gratifications) by evaluations
(perceived values of cable television to have each
gratifications features). The belief measures were ranged from
1 (very definitely) to 5 (very definitly not) and evaluation
measures rated between 1 (very good) to 5 (very bad). So, the
possible scores for expectancy-value (biei) could range 1 to
25. The products of expectancy (beliefs) and value
(evaluation), then, were subject to a varimax-rotated
principal components analysis. for' exploratory' factors. A
varimax-rotated factor analysis of 23 corresponding
expectancy-value (biei) items produced four exploratory
factors with minimum eigenvalue greater than 1 which explained
73.9% of the total variance. Items with a factor loading of .5
or greater were used to identify and label the factors (See

table 4 for complete matrix of primary factor loadings).

58

Table 4

Factor matrix for Expectancy-Value Judgements of Cable Viewing

F1 F2 F3 F4
Factor 1: Escape/Companionship
Pass the time away .69 .49 -.01 -.00
Feeling of companionship .63 .47 .17 .28
Feel less lonely .75 .38 .22 .09
When there is no one else to talk .84 .21 .22 .01
For emotional experiences .68 .31 .25 .31
Talk about with others .71 .19 .37 .16
Forget my problems for a while .80 .24 .18 .15
Escape from the reality 09 life .81 .26 .13 .14
Interesting and stimulating .66 .12 .43 .35
Peps me up .57 .25 .35 .43
Spend time with others .77 .10 -.00 .28
Factor 2: Entertainment/Relaxation
Amuse me .18 .77 .37 .19
Makes me relaxed .55 .68 .12 -.17
Makes me enjoyable .40 .73 .25 .15
Feeling of adventure and excitement .26 .63 .25 .35
Makes me entertained .23 .63 .50 .26
Releases me tension .51 .65 .13 -.15
Factor 3: Information
Information to learn something new .11 .18 .85 .02
Something I want to see .18 .28 .75 .20
Useful information for daily living .22 .16 .88 .02
Information about the world .18 .11 .90 .03
Factor 4: Arousal
Makes me thrilled .17 .48 —.11 .68
Uncensored programs .22 -.04 .17 .66

Inter-code reliabilities: .87(Escape/Companionship),
.80(entertainment/relaxation), .70(Information), .42(Arousal)

59

Factor 1, labeled Escape/Cbmpanionship, explained 51.5%
of the total variance. Items included in this factor were 11
items including escape items such as "forget my problems for
a while" and "escape from the reality of everyday life."
Companionship items such as "feeling of companionship" and
"when there is no one else to talk with" were loaded highly in
the same cluster. Factor 2, labeled Entertainment/Relaxation,
explained 10.8% of the total variance. It consisted of 6 items

including entertaining features of cable television such as

"it amuses 'me" and "makes me enjoyable" and relaxation

features like "it relaxes me" and "it releases tension."
Factor 3, labeled as Information , explained 6.3% of the total
variance. It contained 4 information-related features of cable
television such as "information to learn something new,"
"useful information for daily living," and "information about
the world." Factor 4, labeled as Arousal, explained 5.3% of
the total variance. Items included in this were "makes me
thrilled" and "uncensored program."

Based on this factor analysis, composite expectancy-value
judgement scales were created by calculating the mean scores

for each factor. So, four composite variable were constructed.

Table 5

Mean scores

60

for Composites

of Expectancy—Value

 

 

 

 

 

judgements (Ebiei) variables

Male Female Basic- Non-Cable Pay

only

Factor 1: 10.99 9.81 10.75 10.75 9.12
Escape/Companionship
Factor 2: 7.77 7.59 8.27 7.32 5.78
Entertain/Relaxation
Factor 3: 5.85 5.42 5.9 6.53 4.11
Information
Factor 4 10.96 13.02 12.43 11.2 9.17
Arousal

 

 

 

 

 

 

 

Note. All measures could range from a low of 1 to 25 with lower mean means

more salient variable.

The full matrix of expectancy-value judgement variables

is summarized in table 5.

judgement of EScape/COmpanionship was

For example ,

expectancy-value

constructed by summing

 

up all the scores of the products of beliefs and evaluations

across 11 items clustered in factor 1. In this manner, four

expectancy-value composites were created: Escape/COmpanionship

with a mean of 10.70 (SD: 5.12), Entertainment/Relaxation (M:

7.77, SD: 4.18), Information (M 5.90, SD: 4.41), and Arousal

(M: 11.67, SD: 5.38). Each viewer group was compared in terms
of salience for each expectancy-value judgements variables.

For respondents who received basic cable only, Arousal was

61

(M: 5.90, SD: 4.30) was perceived as a primary outcome from
cable television viewing , followed by Ehtertainment/
Relaxation (M: 8.27, SD: 4.59), Escape/Companionship (M:
10.75, SD: 5.14), and Arousal (M: 12.43, SD: 5.60).

The computed mean scores of each factors for non-cable
subscribers also indicated that Information Arousal was most
salienttly anticipated outcome of cable television viewing (M:
6.53, SD: 5.00), followed by Entertainment/Relaxation (M:
7.32, SD: 3.54), Escape/Chmpanionship (M: 10.75, SD: 5.15),
and Arousal (M: 11.2, SD: 5.24).

Among pay-subscribers, Information (M=4.11, SD: 1.91) was
the most strongly anticipated outcome of cable television
viewing, followed by Entertainment/Relaxation (M: 5.78, SD:
3.26), Escape/COmpanionship (M: 9.12, SD: 4.97), and Arousal
(M: 9.17, SD: 3.71). These findings show that "Instrumental"
rather than "ritualistic" motive is the primary expectation of
Korean students from cable television viewing (Rubin, 1984).
There were no big differences between male and female students
in terms of anticipated outcomes of cable television viewing.
For male students, Information (M: 5.85, SD: 4.26) was the
primary gratification factor expected from cable television

viewing, followed by Entertainment/Relaxation (M: 7.77, SD:

62

4.22), Arousal (M: 10,96, SD: 5.24), and Escape/COmpanionship
(M: 10,98, SD: 4.90). For female students, Arousal (13.02,
SD:5.40), followed by Escape/Cbmpanionship (M:9.81, SD:5.07),
Entertainment/Relaxation (M:7.59, SD:4.01), and Information
(M=5.42, SD=4.17).

In general, throught the entire sample of Korean
students, Information was most strongly expected while Arousal
was the least anticipated outcome of the cable television
viewing. In other words, Korean students tended to expose to
cable television, be it basic or pay—channel, with an
expectation of getting information for daily living and

information about the world (see table 5).

A i w r 1 vi win 3 1

General attitudes of students toward cable television
viewing were rated on the three adjective-pair statements. For
instance, students were asked to rate on a five—point scale
about the question "I think watching cable television is very
good (1)/ very bad (5) very wise (1)/ very unwise (5), and
very beneficial (1)/ very harmful (5)." The sum of the three

standardized scores created a composite attitude measure (M:

63

8.44, SD= 2.23).

mm

A single subjective norm item asked students about the
extent to which "most people who are important to me (e.g.,
family members or friends etc.) think I should subscribe cable

television" (M: 3.19, SD: 1.02).

Beeevierel Integtiene
Students' intention to subscribe cable television was
measured on a single behavioral intention item of "I intend to
subscribe to cable television (or continue subscribing). A
five-point scale ranged from very likely (1) to very unlikely

(5). The mean score was 2.72 (SD: 1.15).

Behevier (Level ef Expeegge te geble Televieien)

Students were asked to rate number of hours they spend
watching cable television on a typical weekday. A five-point
scale ranged from 1 (0 to 1 hour) to 5 (more than 4 hours).
A third (32%) of respondents have reported viewing cable
television 1-2 hours on a typical weekday and a quarter(25%)

of respondents said they watch 0-1 hours a weekday. Viewing

64

levels for weekend days were also rated on the same scale with
24% of respondents were watching 2-3 hours, followed by 21% on
0-1, 20% on 1-2, 18% on more than 4, and 16% on 3-4 hours.
Exposure behavior also was obtained by asking subscribership
to cable television (Yes: 1, No: 2). Sixty three percent of
the respondents were regular cable subscribers and 37% were

not subscribing to cable television.

Testing Hypotheses

The first hypothesis (Hla) predicted. a jpositive
relationship) between. students' attitude toward. cable

television viewing and the strength of their expectancy—value

judgements (Zbiei) of cable television viewing. Zero-order

correlation between attitude and Zbiei measures indicated

strong' relationship 'with. correlation. coefficient of .465

(p<.001). The results are summarized in table 6.

Table 6

Correlation Coefficients between Attitudes and Expectancy-
Value Judgements (Ebiei)

Attitude .4652

Note. Attitude score are obtained by summing up three

attitude items. Expectancy-Value Judgements (Zflaiei) summed
scores of the products of beliefs (bi) and evaluations (ei)
on 23 items.

p<.001.

For more detailed explanations, correlation

65

66

coefficients were computed between attitude measures and
each of the four expectancy-value judgements composites
(i.e., Escape/COmpanionship, Entertainment/Relaxation,
Information, and Arousal). The results in table 7, again,
support the first hypothesis of strong relationship between
attitude and expectancy-value judgements. Students'
expectancy-value judgements of Information were most
strongly related to their attitude toward cable television
(.43, p<.001), followed by Entertainment/Relaxation (.42,
p<.001), EScape/COmpanionship (.36, p<.001), and Arousal
(.22, p<.05). These findings generally supported the
prediction that students' beliefs (perceived probability)
about the consequences of cable television viewing (i e.,
gratifications obtained) and their evaluations of the
consequences would be antecedents of attitude formation
toward cable television viewing. This is consistent with
findings of other research done by Babrow (1989), Babrow &

Swanson (1988), and Palmgreen & Rayburn (1984, 1982).

67

Table 7

Correlation Coefficients between Attitudes and Composites of
Expectancy-Value Judgement Variables (Ebiei)

Factor 1 Factor 2 Factor 3 Factor 4
Attitude .36 ** .42 ** .43 ** .22*
(N: 97) (N: 96) (N: 96) (N: 95)

Note. Factor 1 (Escape/Companionship) consists of 11 items.
The mean score of 11 items was entered in factor 1. Factor 2
(Entertainment/Relaxation) includes 6 items. Factor 3
(Information) includes 4 items. Factor 4 includes 2 items.
*p<.05; **p<.001.

The second hypothesis predicted that when Eku.and Zei

are independently related to attitude, the strength of the

relationships between attitude and each Zbi or Zei will
become weaker (or nonsignificant) than the strength of
relationship between attitude and 2biei combined. This

prediction was made due to the combined effects of Zbi and

Eei on attitude. The results in table 8 show that when Ebi

was controlled constant, the correlation coefficient between

attitude and Zei was .41 (p<.001) demonstrating a reduction

of the strengh.

68

Table 8

Correlations Coefficients of attitude with Zbi, Eei, and
Ebiei.

2b.]: a 28.7: b 2b.]: 8.1..
Attitude .41* -.13 .46*
(N=96) (N=97) (N=96)

*p<.001.
a When Zei was controlled. b.When Ebi was controlled.

When the attitude was related to Zei with control for Zbi,

the relationship was nonsignificant supporting the suggested
hypothesis (1b). This finding indicates that students'
beliefs rather than evaluations toward cable television are
more consistent antecedent of their attitude formation,
since "evaluations are viewed as relatively stable elements
that are the products of an individual's needs and value
system (Palmgreen & Rayburn, 1984, p.540)."

Hypothesis 1a was further tested by regression
analyses. The proposed expectancy-value model poses that

expectancy-value judgements best predict attitude. So,

69

students' attitude toward cable television was regressed on
students' expectancy-value judgements regarding cable
television viewing. The result in table 9 shows that

attitude was strongly predicted by expectancy—value
judgements (2biei)(Beta: .47, F(1,82): 22.65, p<.001)

supporting hypothesis 1a.

Table 9

Regression of Attitude on Expectancy-Value Judgements
(Ebiei), Belief (Zbi), and Evaluation (Eei)

Variables E SE B Beta T
2.....-
Ebi 08 .15 51** 5 7
Eel 05 .14 32* 3 1

Note. For 2biei, R: .47, R square:.21, F(1,92)= 22.65.
For Ebi, R:.52, R square:.27, F(1,93)=32.64.

For Eei, R:.33, R:.11, F(1,93)= 10.08.
*p<.01, **p<.001.

Interestingly, when regression analysis was employed,

attitude was predicted most strongly by Zbi (Beta:.51,

70

p<.001). At the same time, attitude was also regressed on

Eei with Beta of .32 (p<.01) but weaker than when regressed

on 2biei. These findings from regression tests partly

supported the hypothesis 1b.

Prediction of attitude was subject to further
regression analyses of attitude on four expectancy-value
judgement variables: Escape/Companionship,
Entertainment/Cbmpanionship, InfOrmation, and Arousal. Here
again, attitude toward cable television was predicted when
attitude was regressed on four composites of expectancy-
value judgements variables (R square: .27, F(4.79): 7.45,
p<.001). The summary of regression test is shown in table

10.

In this analysis, factor 3, Information was the only
significant factor in predicting attitude (Beta=.27, p<.05).
Viewed from these findings, it is safe to conclude that
Korean students' attitudes toward cable television viewing
are related to their expectancy-value judgements (especially

Information expectation) of the medium.

71

Table 10

Regression of Attitude on Expectancy-Value Judgement
Variables

Variable B SE B Beta T

Factor 4 .04 .05 .10 .89
Factor 3 .14 .06 .27 2.37*
Factor 1 -.01 .07 -.03 -.21
Factor 2 .16 .08 .30 1.90

Note. R:.52, R square:.27, F(4,92): 7.46, p<.001.
Factor 1 (Escape/Companionship) consists of 11 items. The
mean score of 11 items was entered in factor 1. Factor 2
(Entertainment/Relaxation) includes 6 items. Factor 3
(Information) includes 4 items. Factor 4 includes 2 items.
*p<.05.

The third hypothesis (2a) predicted a strong

relationship between students' expectancy-value judgements

(Ebiei) and their gratification seeking (ZGSi). To test

the hypothesis, a caution was made. ZGSi was measured by

asking students reasons for viewing television in general
rather than cable television and Zbiei was a measure of
students' expectancy-value judgements toward cable

television. We may, then, expect that students who are not

receiving cable television (non-subscribers) will rate their

72

reasons for television viewing based on what they are
getting (usually over-the-air), while students who are
subscribing to cable television will rate their reasons for
television viewing based on cable television. But, when
asked to rate their expectancy-value judgements regarding
cable television, students in each groups will not rate in
the same way. For instance, a non-cable student whose
expectation and perceived value of cable television in
satisfying their need of "entertainment" is very low may
seek other media (usually over-the-air) sources to meet that
need. For this group of students, the relationship is very
unpredictable because of too many variables involved. One
possible relationship is negative relationship between
expectancy-value judgements and gratifications seeking in
which students actively seek entertainment need from
television, but, mostly from over-the-air television because
of low anticipation of meeting that need by subscribing to
cable television. On the other hand, a student whose
expectancy-value judgement of cable television in meeting
their psychological needs is very high may actively seek
various gratifications by subscribing to cable television.

For this kind of student group, the relationship between

73

expectancy-value judgement and gratification seeking will be
very strong. Accordingly the hypothesis (2a) was tested for

cable subscriber only.

Table 11.

Correlation Coefficients of ZGSi with 2191', Eei, and
Zbiei: cable only

*p<.01., **p<.001.
.. When Eei was controlled. to. When Zbi was controlled.

The result in table 11 shows a strong relationship(.74,

p<.001) between expectancy-value judgements (Zbdei) and

gratifications seeking (EGSi) behaviors from cable
television viewing among cable subscribers. For further
analyses, correlation coefficients of gratifications sought
(GSi) with bi, ei, and biei were computed for each
corresponding gratifications items. Table 12 shows that

gratifications seeking was strongly related expectancy-value

judgements in 22 out of 23 items with correlation

74

Table 12

Correlation Coefficients of Gratifications Sought(GSi)

with bi, ei, and Expectancy-Value Judgements (biei)
television. (Cable Only).

biei bia eib
It is thrilling .33 .31 .05(n.s)
Learn something new .44 .32 .21*
It amuses me .47 .19(n.s) .30
Pass the time away .46 .17(n.s) .26*
It relaxes me .48 .38 .15(n.s)
It is enjoyable .47 .31 .08(n.s)
Feel adventure and excitement. .42 .25* .21*
Companionship .40 .40 .03(n.s)
Something I want to see .04(n.s) .03(n.s) .05(n.s)
Uncensored program .56 .50 .20*
It entertains me .53 .20(n.s) .28
Makes me feel less lonely .64 .34 .29
No one else to talk with .72 .37 .38
Have emotional experiences .59 .40 .22*
To talk with others .47 .27 .27
Forget my problems .67 .34 .29
To release tension .58 .30 .24*
Escape from reality of life .66 .42 .21*
Useful information for living .36 .26 .11(n.s)
Be interesting and stimulating .47 .40 .08(n.s)
Learn about the world .46 .06(n.s) .43
peps me up .40 .04(n.s) .32
spend time with others .51 .31 .11(n.s)

*p<.05. All others are p<.001.

..When ei was controlled. b.When bi was controlled.
coefficients ranging from .33 to .72 (p<.001). Most strong
relationship was found on the item, "no one else to talk

with (.72, p<.001) and "forget my problem (.67, p<.001). The

75

least correlation was found on the item, "it is thrilling"
(.33, p<.001) and similarly on the item, "useful information
for living" (.36, p<.001). Correlation analyses were further
employed to each dimensions of gratification seeking and
expectancy-value judgements. Results of 4 by 5 correlation
analyses between two composites of expectancy-value
judgements and gratifications sought show that two variables
are strongly related. In 18 out of 20 cases, the correlation
coefficients were found to be strong at significant levels
with a range of .27 to .78 (See table 13).

Interestingly, the relationship was most strong between

the corresponding composites. For instance, students'
information seeking motive (ZGSi factor 4) from cable
viewing was most strongly related to their expectancy-value
judgement of Information (EbUei factor 3)(.50, p<.001).

When the same gratification seeking motive was related
to other expectancy-value judgement factors, the

relationship tended to be weaker (.23 to .40). For example,

the strongest relationship of Escape factor of EGSi was

with Escape/COmpanionship of Zbiei (.78, p<.001). And

Relaxation was most strongly related with

76

Table 13

Correlation Coefficients between both Composites of Ebiei
and EGSi.

Ebiei F1 Ebiei F2 Ebiei F3 Zbiei F4
EGS F1 .78*** .59** .27* .27*
268 F2 .38*** .37** .09(n.s) .54***
265 F3 .54*** .72*** .29* .39**
203 F4 .33* .40*** .50*** .23*
EGS F5 .56*** .52*** .30* .13(n.s)

Note. EGSi factor1:Escape, 2:Arousal, 3:Relaxation,
4:Information, 5:Pass-Time. Factor 6 Special Program was
excluded in the analysis because of low variance. Zbiei
factorl:Escape/Companionship, 2:Entertainment/Relaxation,
3:Information, 4:Arousal.

Entertainment/Relaxation factor of Ebiei (.72, p<.001).
Arousal was also most strongly related with the
corresponding Arousal factor of Ebiei (.50, p<.001). Pass-
Time factor of EGSi was related with both Escape/
COmpanionship and Entertainment/Relaxation factor of Zbiei
(.56 and .52, p<.001).

In sum, the complete matrix of correlation coefficients

between composites of Ebiei and EGSi supported the

77

suggested hypothesis of a strong relationship between ZGSi

and Zbiei. In other words, we may expect that students who

strongly believe that viewing cable television will lead to
certain outcomes (i.e., gratifications sought) and also
evaluate that outcome highly would actively seek various
gratifications from cable television viewing.

Since two variables were found to be Strongly related,
an attempt was made to predict gratification seeking

behaviors from cable television.

To predict gratifications seeking (ZGSi) behavior of

students from cable television, regression analysis was

employed on the sum of students' expectancy-value judgements

(2biei) and Ebi and Eei. The results in table 14 shows

that students' expectancy-value judgements strongly
predicted their gratification seeking from cable television
( R: Beta: .74, R square: .54, F(1,60)= 63.69, p<.001). More
importantly, students' gratification seeking was more
predictable when students' beliefs and evaluations were
combined than when beliefs and evaluations were
independently tested (R : Beta: .69, R square: -48, F(1,61):

54.94, p<.001; R : Beta:.71, R square:.51, F(1,61), p<.001

78

respectively).

Table 14

Regression of ZGSi on Expectancy-Value Judgements (2biei),
Belief (Zbi), and Evaluation (Zei).

Variables B SE B Beta T
2... 7. 9;
2b; 68 .99 69 7 40
Eel 59 .08 71 7 60

Note. For Ebiei, R: .74, R square:.54, F(1,60)= 63.69.
For Zbi, R:.69, R square:.48, F(1,61):54.94.

For Zei, R:.71, R square.51, F(1,61): 58.08.
p<.001.

Students' gratifications seeking was subject to further
regression analysis on the composites of expectancy-value

judgements. (see table 15).

Regression of gratification seeking (ZGSi) on the four

composites of expectancy-value judgements (Ebiei) clearly

supported the hypothesis 2a ( R: Beta: .77, R square:.52,

F(4,59): 17.90, p<.01). In more detail, ZGSi was strongly

regressed on Escape/Chmpanionship (Beta: .54, p<.001). These

79

Table 15

Regression of ZGSi on the Composites of Expectancy-Value
Judgements (Ebiei).

Variable B SE B Beta T

Factor 4 .31 .26 .12 1.18
Factor 3 -.41 .39 -.12 -1.00
Factor 1 1.48 .50 .54 2.98*
Factor 2 .74 .62 .24 1.12

Note. R:.77, R square:.59, F(4,59)= 17.90, p<.001.

Factor 1 (Escape/Companionship) consists of 11 items. The
mean score of 11 items was entered in factor 1. Factor 2
(Entertainment/Relaxation) includes 6 items. Factor 3
(Information) includes 4 items. Factor 4 includes 2 items.
*p<.001.

results strongly supported the suggested hypothesis that
students' gratification seeking from cable television will
be positively related to their expectancy-value judgments of
cable television. And their expected escape/companionship
outcome strongly predicted their various gratifications

seeking behaviors (Beta=.54, p<.001). Other expected motives

were statistically insignificant.

Hypothesis 2b predicted that students' expectations

80

(Zbi) and their evaluations (Eei) toward cable television

viewing combined (Zbiei) will predict their gratifications

seeking behavior more strongly than either Zbi and Eei

alone was allowed to predict gratifications seeking. The
results in table 11 clearly supported the prediction when

the correlation coefficients of gratification seeking

(EGSi) with Zbi only was reduced to .30 (p<.01) from .74

(p<.001), with the control of Eei. The same result was

found between EGSi and Zei with control of 2bi. The

correlation coefficient was again reduced to .30 (p<.001)
from .74 (p<.001). This hypothesis was further tested by
examining the entire 23 by 23 GSi verses bi and ei
correlation matrix (see table 12). In 22 out of 23 cases,
there were reductions in the strength of relationships when
each bi and ei was separately related to the corresponding
gratification seeking items (GSi). The means of
correlations were reduced from .48 to .28 (bi) and to .20

(ei). The regression tests in table 13 also supported the

hypothesis when ZGSi was more strongly regressed on 2biei

(.74, p<.001) than it was regressed on either Ebi (.69,

81

p<.001) or Zei (.71, p<.001). These findings, again,

demonstrate the combined effects of beliefs and evaluations
on students' gratification seeking behaviors (Palmgreen &
Rayburn, 1984).

The fifth hypothesis (H3) predicted a positive
relationship between students' two generalized orientations
toward cable television: attitudes and their gratification
seeking behavior. This prediction was made because if

students' expectancy-value judgements are related to both

their attitudes and gratification seeking (EGSi), it is

reasonable to expect a positive relationship between

attitudes and gratification seeking (EGSi). In other words,

a student who has positive attitude toward cable television
may actively seek various gratifications from the exposure
to cable television. By the same token, we may expect that

a student who feels his or her gratifications sought have
been met by exposure to cable television before may also
have positive attitude toward cable television. Here again,
a caution was made before testing the hypothesis.
Gratifications sought were measured by asking students to

rate their underlying reasons for attending television in

82

general. So, students not subscribing cable television may
have rated their reasons based on their experiences with
over-the-air or other media sources rather than cable
television. For this group of students, it is very difficult
to examine the relationship between their seeking of
gratifications and their affective responses (attitudes)
toward cable television. For example, a student who has
positive attitude toward cable television may not
necessarily seek "Information" from cable television.
Instead, they may satisfy the same need by exposing
him/herself to other media sources. In this case, this
student's affective responses have nothing to do with their
seeking of "Information" from television. For this reason,
the hypothesis was tested for cable-only group.

The results in table 16 supported the hypothesis when
students' attitudes toward cable television and their
gratification sought were moderately related (.35, p<.01).
The hypothesis was further analyzed for each composite of
gratifications sought (table 17). For 3 out of 5 factors,
attitudes and gratifications sought were moderately related
at the significant levels. Information was most strongly

related with attitudes (.36, p<.01) followed by Relaxation

83

Table 16

Correlation Coefficients between Students' Attitudes and
Gratification Sought (ZGSi) from Cable Television.

Cable Non-Cable

Attitude .35* .26 (n.s)

Note. Cable (N:65), Non—Cable (N:35).
*p<.01

(.28,p<.05), and Arousal (.24,p<.05). These findings showed
that students' attitudes toward cable television were
related to their seeking of gratifications from cable

television, but only moderately.

Table 17

Correlation Coefficients between Attitudes and Composites of
Gratifications. (Cable only)

Attitudes .18(n.s) .24* .28* .36** .14(n.s)

Note. F1(Escape), F2(Arousal), F3(Relaxation),
F4(Information), F5(Pass-Time).
*p<.05., **p<.01.

84

The sixth hypothesis (H4) predicted a strong
relationship between students' attitudes toward cable
television and their intention to subscribe cable
television. The correlation coefficient of attitudes with
intention to subscribe to cable television was .53 at p<.001
level for all respondents. Table 18 summarizes the results
among different viewer groups. The relationship was held for
both cable (.54, p<.001) and non-cable (.44, p<.001) groups.

So, the hypothesis was supported.
Table 18

Correlation Coefficients of Attitude with Intention to
subscribe to cable television.

Intention .54 .44

Cable (N=65), Non-Cable (N=35). p<.001.

Since students' attitudes and behavioral intention to
subscribe to cable television were strongly related,
regression analysis was employed to predict students'

intention for subscribing to cable television. The results

85
Table 19

Regression of Students' Intention to subscribe to Cable
Television on Students' Attitude toward Cable Television.

B SE BETA T
All .27 .05 .53 5.95**
Cable .24 .05 .54 5.00**
Non-Cable .25 .09 .44 2.76*

Note. For all: R:.53, R square:.28, F(1,98)=35.45.; For
cable: R:.54, R square:.29, F(1,65):25.02.; For non-cable:
R:.44, R square:.20, F(1,35)=7.60.

*p<.0.01., **p<.001.

in table 19 supported the hypothesis again when students'
intention to subscribe to cable television was regressed on
students' attitudes (R: Beta: .53, F(1,98)= 35.45, p<.001).
Similar result was obtained when students' intention also
regressed on attitudes of both student viewer groups (basic:
R: Beta: .54, F(1,65): 25.02, p<.001; non-cable: R: Beta:
.44, F(1,35)= 7.60, p<.001). So the proposed hypothesis (H4)
was supported. And this finding is also consistent with that
of Babrow (1989) who found students' intention to watch soap

opera was strongly predicted by their attitudes toward soap

opera viewing.

86

The seventh hypothesis (H5) predicted a positive

relationship between students' gratification sought (EGSi)

and their intention to subscribe to cable television. This
hypothesis was tested for students subscribing cable
television because of the possible inconsistency between two
viewer groups in measuring gratifications sought.
Correlation analysis showed that students' intention was
positively related to their gratifications sought, but the
relationship was moderate (.34, p<.001). The relationship
was further examined between intention and each dimension of
the gratifications sought. The results summarized in table
20, once again, showed that intention was low to moderately
correlated to gratifications sought. Relaxation motive was
most strongly related to intention (.55, p<.001), followed
by Arousal (.27, p<.05), and Information motive (.25,
p<.05), respectively. Escape and Pass-Time motives were
nonsignificantly related to intention (.14 and .07 at
nonsignificant levels). To further examine the relationship
between intention and gratifications sought, students'
intention to subscribe to cable television was subject to

regression analysis. The complete regression analysis is

87

Table 20

Correlation Coefficients of Students' Intention to Subscribe

Cable Television to Gratifications Sought (ZGSi). Cable
only.

ZGSI
F1 F2 F3 F4 F5
Intention .14(n.s) .27* .55** .25* .07(n.s)

Note. F1(Escape), F2(Arousal), F3(Relaxation),
F4(Information), F5(Pass-Time).
*p<.05., **p<.001.

shown in table 21.

Overall, students' intention was moderately related to
the sum of 23 gratifications sought (R: Beta: .34, F(1,63):
7.58, p<.01). The hypothesis was further tested by
regressing intention on each composite of the gratifications
sought (R square: .34, F(1,63): 5.79, p<.001. Among five
gratification motives Relaxation motive was most strongly
related to students"intention to subscribe to cable
television (.65, p<.001). Other motives were statistically

nonsignificant in predicting intention.

88

Table 21

Regression of Students' Behavioral Intention on
Gratifications Sought(2GSi).

B SE Beta T
""""""" i;;;,'.;""""'3fi6m’i;""3’i6""3’;;""’
Arousal .09 .15 .08 .60
Relaxation .93 .20 .65 4.56*
Information .14 .19 .10 .69
Pass-Time -.28 .18 -.25 —.16

Note. R:.59, R square:.34, F(1,63)=5.79, p<.001.
*p<.001.

The eighth hypothesis (H6) predicted that the
relationship between students' expectancy-value judgements
and their intention to subscribe to cable television will
become very weak (or nonsignificant) when attitudes and

gratifications sought are controlled constant. This

prediction was made because gratifications sought (ZGSi)

and attitudes mediate the effects of expectancy—value

judgements (2biei) to intention.

When attitudes and gratifications sought (ZGSi) were

89

controlled constant, the correlation coefficient between

Zbiei and intention was nonsignificant (—.03 at

nonsignificant level), compared to .33 (p<.01) when

attitudes and Zbiei were not controlled for. This was

further analyzed between each composite of Zbiei and

intention with a control of proper variables.

The results in table 22 clearly indicate that the
relationship between students' expectancy-value judgements
of cable television and their intention to subscribe (or
continue subscribing) are statistically nonsignificant. This
finding implies that attitudes and gratifications sought
mediate between two variables. The same result was found by
Babrow & Swanson' (1988) study of television news viewing in
which they found no evidence of direct effects of
expectancy-value judgements on intention. The results shown

here are sharp contrast to Triandis'(1980) contention that

Zbiei may contribute more directly to intention. The

present study further attempted to examine the direct

relationship of beliefs (Zbi) to behavioral intentions

(RQ2), as suggested by Liska (1984). To test this question,

90

Table 22

Partial Correlation Coefficients of Students' Intention to
Subscribe to Cable Television to Expectancy-Value Judgements

(Ebiei) with Control of Attitude and Gratifications Sought
(ZGSi)

No Control for Control for Control for

Control Attitude 2051' Both
.1; """"" ééi """"" 6.3.2.127"'"T6325;§"""I'6;2;';§
F2 .37** 12(n s) 20(n s) 03(n 8)
F3 27* 02(n s) .18(n s) - 02(n s)
F4 24* 14(n s) .15(n s) 09(n s)

Note. F1(Escape/Companionship),
F2(Entertainment/Relaxation), F3(Information), F4(Arousal).
*p<.05., **p<.001.

proper control of variables was necessary. In this case,

Zei, attitudes, and gratifications sought (EGSi) were
controlled for because of the possibility to mediate the
effects of Zbi on intention. When Zei, attitudes, and EGSi
were partialed out, the relationship between beliefs (Zbi)

and behavioral intention was statistically insignificant (—

.06 at nonsignificant level). Without control, the beliefs

(Zbi) were moderately correlated to intention (.33,p<.001).

91

Accordingly, in the present study, students' beliefs
had no direct effects on students' intention to subscribe to

cable television.

The ninth hypothesis (H7) predicted a positive
relationship between students' perceptions of subjective
norms and their intention to subscribe to cable television.
A single subjective norm item asked students to rate their
perceptions of social pressure from the people important to
them when they intend to subscribe to cable television. A
strong correlation coefficient of .60(p<.001) was found
between subjective norms and behavioral intention in
supportive of the prediction for a pOsitive relationship.
Regression analysis further supported the hypothesis when
intention was strongly regressed on subjective norms (R:
Beta: .60, F(1,96)= 53.58, p<.001). This finding provides an
additional evidence to support the importance of social
pressures when people make behavioral decisions (Babrow &

Swanson, 1988; Babrow, 1989; Fishbein and Ajzen, 1975).

The tenth hypothesis (H8) predicted a positive

relationship between students' behavioral intention to

92

subscribe to cable television and their actual viewing
behavior. Viewing behavior was measured by summing up the
amount of time students spend watching cable television
during both weekdays and weekends, and also measured by
asking their cable subscribership (1: Yes, 2: No).
Correlation analysis showed a positive relationship
between students' exposure levels and intention for
subscription. For example, a moderately strong correlation
coefficient (-.48, p<.001) was found between students'
amount of cable television viewing in a typical week and
their intention to subscrive to cable television. Because
lower numbers on the measurement scale of intention mean
higher degrees of intention and higher numbers on the
measurement scale exposure level mean higher viewing
behaviour, the negative relationship between two variables
indicates a positive relationship between students'
intention to sbscribe to cable television and their viewing
behavior. This result was confirmed when intention was
positively related to the students' cable television
viewership relationships (.38, p<.001: subscribing: 1,
nonsubscribing: 2). Regression analysis also showed a

relatively strong relationship between two variables when

93

students' exposure levels of cable television were
negatively regressed on intention (R: Beta: -.48, F(1,95):
27.53, p<.001). So, these findings supported the proposed
hypothesis of a strong relationship between viewing behavior
and viewing intention.

To answer the second research question (RQ2) that the
possible direct effects of attitudes on cable viewing
behavior, correlation analysis was administered. When
variables such as gratifications sought, viewing intention,
and subjective norms were properly controlled constant,
there were no relationship between students' attitudes and
their viewing behavior (-.08 at nonsignificant level). So,
this finding did not support the Liska' (1984) conclusion
that attitudes can either directly or indirectly influence
the behavior. In this study among Korean students' cable
television viewing behavior, only indirect effect of

attitudes was found.

Chapter V

Summary and Discussion

94

95

For both television audiences and researchers who
examine the effects of the medium, the newly developed
television environment is a challenge. It is challenging for
audiences because they have to reallocate their limited
resources when they expose themselves to the diversified
media messages, so as to obtain maximum satisfactions from
the viewing. So, audience members' exposure to television
can be seen as an individual rather than collective decision
making process. Under this circumstance, each individual is
an "active" rather than "passive" agent in the communication
process. As a result, any kinds of new media consumption
must be understood in close relation to the process of
decision making by an individual member.

New media environment is also challenging to media and
audience researchers. So often did they rely on quantitative
data in their attempts to untangle the complexities of the
media consumption and its effects on society. For them,
"quality" of exposure is sometimes preferred in favor of
"quantity." So, measuring how many watch a particular
medium or content is the primary concern to determine the

performance of that medium or content. This practice can be

96

1993), in which those "numbers (e.g., number of total
audiences or number of hours contributed to television
viewing)" must be shared by many competing media sources.
Quantitative date, then, becomes less useful than
qualitative data. The present study was an attempt to
investigate this concern in the context of cable television
viewing by college students. More specifically, this study
concerned with students' use of cable television as it is
related to their motivational schemes for media uses. So, it
was individual in the level of analysis, and concerned
qualitative aspects of television viewing. Uses and
gratifications perspective was employed as the theoretical
framework in association with expectancy-value theory. The
validity of this attempts has been demonstrated by
researchers in various media settings (Babrow, 1989; Babrow
& Swanson, 1988; Galloway & Meek, 1981; Palmgreen & Rayburn,

1984,1982).

Summary of Findings

T 01 f T 1 v i n Vi win

Students' reasons for television viewing generated six

97

factor solutions that combined explained 65.4% of the total
variance. The six factors were labeled Arousal(3.35), Escape
(3.31), Pass-Time (2.93), Special-Programs (2.73),
Relaxation (2.64), and Information (2.40). Special-Program,
anchored by one item, was excluded from the subsequent
analyses because of difficulty in deriving meaningful
interpretation. Both basic-only students and non-cable
students sought Information motive most strongly while
Arousal was least strongly sought. Overall, students
receiving basic only were seeking various gratifications
most strongly (mean: 2.96) among three viewer groups,
followed by non-cable (mean: 2.87), and pay subscribers
(2.62). The factor analyses clearly indicated that Korean
students used television for two distinctive motives
"instrumental" and "ritualistic" (Rubin, 1982). According to
Rubin, ritualistic use of television is associated with
diffuse motives such as pass-time, relaxation, habit, and
escape. Viewers in this type of viewing show higher affinity
with the medium and they concern less with the content.
Instrumental use, on the other hand, is associated with
purposive and selective use of media content, usually for

information seeking reason. Viewed from this distinction,

98

Korean students seem to show instrumental orientation in

their uses of cable television.

WISE).

Measures of students' expectancy-value judgements
produced four underlying dimensions, which explained 73.9%
of the total variance. The four composites were labeled
Escape/Cbmpanionship, Entertainment/Relaxation, Information,
and Arousal. Information was the most strongly expected
gratification sought among all viewer groups followed by
Entertainment/Relaxation. Here again, Arousal was perceived
as the least possible outcome from cable television viewing.
As in the gratifications sought, students' expectancy-value
judgements of cable television clearly showed two
distinctive orientations toward cable television viewing:

"ritualistic" and "instrumental."

Findin Rela H h
The first hypothesis (Hla) predicted a positive

relationship between students' expectancy-value judgements

(Zflaiei) and their attitudes toward cable televiSion

99

viewing. This expectation was supported by a correlation of
.47(p<.001). When each composite of the expectancy-value
judgements was related to attitudes, the relationships were
ranged from .22 to .43 at significant levels. The hypothesis
was further tested by regressing attitudes on expectancy-
value judgements and the result again supported the
hypothesis (Beta: .47, p<.00‘1).

The second hypothesis predicted weak (or

nonsignificant) relationship between students' attitudes and

either Ebi (with control of Zei) and Eei (with control of

Ebi). This expectation was partially supported. When Ebi

was related to attitudes independently of Eei, the
relationship was reduced, but the relationship was still
substantial (.41, p<.001). Furthermore, attitudes were much

highly regressed on Ebi (.51, p<.001) than on Zbiei (.47,

p<.001). On the other hand, when Zei was related to

attitude, the relationship was nonsignificant (-.13). As a
result, the hypothesis (Hlb) was partially supported.
The third hypothesis (H2a) predicted a positive

relationship between students' expectancy—value judgements

100
of cable television and their gratifications sought(EGSi).
The correlation coefficient score of .74 (p<.001) between

EGSi and Ebiei was found to support the hypothesis. When

correlation analyses were employed to test relationships

between each composites of ZGSi and Zbiei constructs, the

relationships ranged from .23 to .78 at the significant

levels. Regression analyses also supported the prediction

when ZGSi was regressed on Ebiei (Beta=.74, p<.001).

The fourth hypothesis predicted a reduction in the

strength of relationship when either Ebi and Zei was

independently related to ZGSi. This expectation was

supported when the relationship was reduced to .30 (for both
Etu.and Eei) from .74.

The fifth hypothesis (H3) predicted a positive
relationship between students' attitudes toward cable

television and gratifications sought from cable viewing.

This expectation was supported when a moderate relationship

was found between attitude and ZGSi (.35, p<.05) among

cable subscribers. Furthermore, three out of five ZGSi

101

factors were moderately related (.24 to .38) to attitudes at
the significant level (p<.05).

The sixth hypothesis (H4) predicted a positive
relationship between attitudes and students' intention to
subscribe to cable television. The correlation analysis
supported the prediction for both viewer groups (cable: .54,
p<.001; non-cable: .44, p<.001). Regression test further
upheld the prediction when intention was regressed on
attitudes toward cable (Beta: .54, p<.001) and non-cable
(Beta: .44, p<.01) subscribers.

The hypothesis H5 predicted a positive relationship

between students' intention to subscribe to cable television

and their gratifications sought (EGSi) . The correlation

was moderate (.34, p<.05), still supporting the hypothesis.
Relaxation motive(.55, p<.001) was most strongly related to
intention, followed by Arousal (.27, p<.05), and Information
(.25, p<.05). And intention was most strongly regressed on
Relaxation motive (.65, p<.001).

The hypothesis H6 predicted a weak (or nonsignificant)

relationship between students' intention and their

expectancy-value judgements when attitudes and 26751 were

102

controlled. This prediction was supported when the

relationship was found to be nonsignificant (-.03)

statistically when both attitudes and EGSi were controlled

constant. This result was also true when each composite of

2biei was nonsignificantly related to intention when

attitudes and 265i were controlled constant again.

The hypothesis H7 proposed a positive relationship
between students' perceptions of subjective norms and
behavioral intention. A strong relationship (.60, p<.001)
between two variables supported the prediction. And
intention was strongly regressed on subjective norms (Beta:
.60, P<.001).

The hypothesis H8 predicted a positive relationship
between students television viewing behaviors (exposure
levels and subscribership) and their intention to subscribe
to cable. Both correlation and regression analyses provided
evidences to support the prediction.

Based on these findings, a complete diagram of
expectancy-value model of students' cable television viewing

can be drawn as follows:

103

Expectancy-Value Model of Gratifications Sought from Cable

Television Viewing : Adapted from Fishbein and Ajzen's Model

Attitude\
/ “.48
Ebiei

\BI————7‘B

\ /V4\
2631 /3"

 

 

Note. BI (behavioral intention to subscribe to cable
television), SN (subjective norms), B (exposure levels and
subscribership)

 

Discussion

Expectancy-Value Judgements of Gratifications

Results of the present study generally support the
proposition of a well—known attitude-behavioral model that
posits that attitude is determined by the expected outcomes
of the behavior (beliefs) and the perceived values of those
behavioral outcomes (evaluations). Attitude, then, becomes a
determinant of behavioral intention, which is known as a
direct cause of the behavior. When integrated in uses and
gratifications study, this model has gained considerable
supports to predict media exposure behavior. More
specifically, a generalized media orientation of
gratifications seeking (like attitude) was thought to be
determined by the expectations of obtaining those
gratifications, weighted by the subjective evaluations of
those gratifications. When applied to cable television
viewing by Korean college students in the United States, the
present study found a strong positive relationship (.74,
p<.001) between students' expectancy-value judgements and
their gratifications sought from cable television viewing.

An exploratory factor analysis indicated that Korean

104

105

students expect four dimensions of gratifications from cable
television viewing: Escape/Cbmpanionship,
Entertainment/Relaxation, Information, and Arousal.
Information was perceived by all three viewer groups as the
primary outcome from cable television viewing, while Arousal
was perceived as the least important function of cable
television. These results indicate that Korean students make
use of cable television for instrumental purpose which is
anchored by intentional or selective use of the medium. This
instrumental use of cable television includes getting
information about the world and for daily living. This
pattern of cable television use is understandable in that,
in order to settle down to new culture, Korean students
might want to actively seek for information about the new
country and new cultures. Furthermore, there are limited
sources of information they can count on. In this
circumstance, television provides the easiest access to
those information. Television also can be a venue where
Korean students can relax from their real-life pressures and
problems occurring in the school life. So, television
becomes an important means of "acculturalization" for the

foreign students to the exotic cultures (Kim, 1982).

106

Although it is uncertain which pattern of media use, be
ritualistic or instrumental, facilitates students'
acculturalization processes, an early contact with mass
media may certainly shape their perceptions of the social
reality of the new country. This provides interesting topic
for the future studies regarding the roles and effects of
mass media for the students from different countries.
Cultivation analyses are certainly eligible to examine these
effects.

The next major finding in this study is that
expectancy-value judgment was not a good predictor of

behavioral intention than the general attitude and
gratification sought (EGSi). When each composite of
expectancy~value judgements was related to behavioral
intention to subscribe to cable television with controls of

attitudes, ZGSi, the relationship was statistically

nonsignificant. This finding is consistent with those found
in other studies (e.g., Babrow, 1988; Babrow, 1989; Babrow &
Swanson; 1988). These researchers concluded that a
generalized attitude or gratifications sought was the

stronger predictor of behavioral intention or actual

107

behavior than expectancy-value judgements. One explanation
can be that other functional alternatives (e.g., VCR or
Newspapers) may enable students to obtain those perceived
gratifications (Palmgreen & Rayburn, 1984). In fact, Babrow
(1989) reasoned that popular media content such as soap
operas represents a well-known experience for all and the
perceived expectations and values of various consequences
(i.e., gratifications sought) may be clear and stable. For
this reason, a generalized attitude or gratifications sought
may predict behavior better than expectancy-value
judgements. In a similar fashion, LaRose & Atkin (1991)
contended that previous studies have focused on behaviors
whose consequences were highly predictable. They concluded
that expectancy-value judgements can predict media behavior
under certain media environment where reasoned decision
making processes are needed to select a particular medium or
content among various alternatives. In fact, they found that
expectancy-value judgements directly predicted intention to
consume pay-per-view. In the present study, the most salient
anticipation perceived by Korean students from cable

television viewing was Information, which is considered to

be very intentional and instrumental viewing pattern.

 

 

108

Although this type of viewing pattern requires very
deliberate decision making process for the exposure, it is
not necessarily the case. For instance, Information was
strongly expected across each viewer group. This means that
Korean students' perceived expectations and values of cable
television viewing are very stable and clear. For this
reason, their expectancy-value judgements of cable were less
predictive of intention to subscribe to cable television and
actual viewing behavior than their generalized attitudes and
gratifications seeking from cable television viewing.

Whatever the reason, it seems clear that our belief
systems are far more-complex than expected. These beliefs
have many origins and derive from the sum of an individual's
experiences (direct and indirect) with a particular media
object (Palmgreen & Rayburn, 1984). Fishbein and Ajzen
(1975) distinguished between three kinds of belief systems:
descriptive, inferential, and informational.

"Descriptive" beliefs are formed from the direct
observation of an object. Korean students in this study may
not have experienced much of the cable television. This may
have contributed to the low predictability of behavioral

intentions. "Inferential" beliefs are related more to our

109

personalities than direct observations. "Informational"
beliefs are largely formed by information from outside
sources such as friends or magazines. For some students,
especially new students whose direct experiences with cable
television are limited, beliefs toward cable television may
be formed based on outside sources.

Overall, beliefs are formed over a long period of time.
So, to understand the roles played by belief systems in

media consumption, more longitudinal studies are necessary.

Attitudes, EGSi, Subjective Norms, and Behavioral

Intention

The results in the present study support the
anticipation that students' attitudes will be strongly
related to their intention to subscribe to cable television.
This is consistent with Barrow & and Swanson(1988) finding

that students' attitudes toward news viewing directly

effected both intention and exposure levels. ZGSi, on the

other hand, predicted intention only moderately (.34). The

same results were observed in the present study when each

composite of EGSi was low to moderately related to

110

behavioral intention. While media consumption may be related
to multiple motivations (Palmgreen et al., 1985), the
present study employed a limited set of motivational
schemes, which was developed for studies of traditional
media consumption. In fact, viewers of cable television
tended to seek more differentiated gratifications from cable
television viewing. For instance, Shaver (1983)found certain
unique gratifications sought from cable television. Viewers
tended to subscribe to cable for better receptions of
signals, special programs such as children or religious
programs, and more latitudes of controlling over viewing.
With the consideration of these aspects, the present study
might have been able to explain the relationships between
students' gratifications sought and their behavioral
intention in more predictable ways.

Another explanation may be that such external factors
as work schedule and availability of certain media might
have influenced the viewing (Bogart, 1965). In fact, college
students are very limited in their use of mass media. Their
tight work schedules in classes may limit their actual use
of media sources for various purposes. This contention can

partly be explained by the fact that Relaxation was a strong

 

111

predictor of behavioral intention to subscribe to cable
television. Still another possibility is that Korean
students are grown up in different media environment where
television is free. They are not accustomed to paying
monthly bill for what they used to get free-of-charge.
Although these issues were not discussed in the present
study, they deserve further studies.

The present study also supported the suggested
importance of the perceived social pressures in television
viewing decision. This emphasizes the notion that the media
consumption should be understood in relation to the social
contexts in which the media consumptions occur (Blumler,
1985). In a study of television news viewing, Babrow &
Swanson, 1988) suggested that normative perceptions play a
key role in determining audience exposure levels.

Finally, the present study predicted a positive
relationship between students' behavioral intention and
exposure levels of cable television and subscribership. Both
correlation and regression analyses indicated a positive
relationship between two variables. In consistent with the
assumption of expectancy-value theory, this Study showed

that behavioral intention is a direct cause of actual

112

behavior.

Given the characteristics and size of the sample used
in this study, it is very difficult to generalize the
results to other population. Nevertheless, the present study
provides an evidence that integration of uses and
gratifications and expectancy-value model could help
understand the complexities of decision making processes in
media consumption behavior. More specifically, the present
study shed light on the importance of television viewers'
cognitive and affective involvements in understanding media

consumption behaviors.

REFERENCES

Atkin, D., Greenberg, B.S., and Baldwin, T. (1991). The home
ecology of children's television viewingzparental
mediation and the new media environment. Journal of
eemmenieagien, 41, pp. 40-53.

Agostino, D., (1980). Cable Television's Impact on the
Audience of Public Television. Qeernel of Breadcasting.
24(3), pp. 347-363.

Ball-Rokeach, S., & DeFleur, M.L.(1976). A Dependency Model
of Mass Media Effects. Cemmenieegien Reeeerch, 3, pp.
3-21.

Becker, L.B., & Whitney, D.C.(1980). Effects of Media
Dependencies:Audience Assessment of Government,
Qemmenieeeien Reeeereh,7, pp. 95-120.

Becker, L.B., Sobowale, I.A., & Casey,W.E.,Jr.(1979).
Newspaper and Television Dependencies: Effects on
Evaluations of Public Officials, geernel ef
Breeeceeting & Elecgrenie Medie,23, pp. 465-475.

Becker, L.B.(1979). Measures of uses and gratifications.
gemmenieegien Reeeereh, 6, pp. 54-73.

Becker, L.B., Dunwoody. S.,& Rafaeli, S.(1983). Cable's

113

114

impact on use of other news media. Journal of
Breaecesting, 27, pp.127—140.

Blumler, J.G.(1979). The role of theory in uses and
gratifications studies. Communication Research, 6, pp.
9—36.

Blumler, J.G., & Katz, E.(Eds.).(1974). The usee of mase

eommenieations: Curren; perepeetives on greeification

reeeereh, Beverly Hills: Sage.

Bower, R.T. (1985). The ehanging television eeeience in
Ameriea= New York:Columbia University Press.
Bradbury,D.I.(1990). A eses ene greeifieeeiene eeeey of
ehree eeeieneee: ceele Qeelinere, eeeie eeele
W Unpublished

Ph.D. Dissertation, Temple University.

Broadcasting & Cable. (1994). Mar. 21, p.73.

Bryant, T. (1986) The road most traveled:Yet another
cultivation critique. Qeernel ef Breedceeeing &
Electronie Media, 30, pp. 231-235.

Buerkel-Rothfuss, N., & Mayes, S.(1981). Soap opera viewing:
The cultivation effect. Qeernel ef gemmenieetien,
31(1), pp.108—115.

Collins, J., Reagan, J., & Abel, J.(1983). Predicting cable

115

subscribership: Local factors. Journal of
Br a in 27, pp. 177-183.
Culbertson,H., & Stempel,G.(1986). How media use and

reliance affect knowledge level. communicatien

Reeeerch,13, pp. 579-602.

Davison, A.R. & Jaccard, J. (1979). Variables that moderate

the attitude-behavior relation: Results of a

longitudinal survey. ourn l f Per n li 'n ocial
nsxsholesx. 37. pp- 1364-1376-

DeFleur, M.L., & Ball-Rodeach, S.J.(1982). Theeriee of Mees
gemmeniceeien (4th ed.). New York: Longman.

Dimmick, J., & McCain, T.(Eds.).(1979). Use of mass media:

Patterns in the life cycle. Amerieen Behevieral
Seientieg, 23(1).

Dobos,J., & Dimmick,J.(1988). Factor analysis and
gratification constructs, rn l f r in &
eleegrenie medie, 32(3), pp. 335-350.

Donohew,L., Palmgreen,P., & Rayburn,J.(1987). Social and
Psychological origins of media use: A life-style
analysis, 0 rn l of Br c in El ronic
Media.31(3). pp. 255-273.

Ducey,R., Krugman, D.M., & Eckrich,D. (1983). Predicting

116

Market Segments in the Cable Industrszhe Basic and Pay
Subscribers. Journal of Broadcasting, 27(2), pp. 155-
161.

Ferguson, M.(Ed.).(1986). New communication technologies and
th 1i interes - C m rative ers ectives on olic
ene reeeareh, London: Sage.

Fishbein,M.,& Ajzen,I.(1975). Belief, Ageituee, Intenuion
ene Behevior, Reading,MA:Addison-Wesley.

Fishbein,M.(1963). An investigation of the relationships
between beliefs about an object and the attitude toward

that object, Human releuione,16, pp. 223-240.

Furno-Lamude,D., & Anderson,J.(1992). The uses and
gratifications of rerun viewing, Qeur eliem Quarterly,
69(2), pp. 363-372.

Galloway, J.J., & Meed, F.L.(1981). Audience uses and

gratifications: An expectancy model. gemmunieeeien
Reeeerch, 8, pp. 435-449.

Goodhart, G.J., Ehrenberg, A.S.C., & Collins, M.A.(1975).
The Lelevieien eueienee: Peuterne of viewing,
Lexington, MA:Lexington Books.

Greenberg,B.S.(l974). Gratification of television viewing

and their correlates for British children, in

117

Blumler,J.G., and Katz,E.(eds.). The use of mass
communicetions:current perspectives on gratification
reeeereh, Beverly Hills,CA:Sage.

Heeter, C.(1988). Implications of New interactive
technologies for conceptualizing communication. In J.
Salvaggio & J. Bryand(Eds.). Megia use in ehe

infermetion ege; Emerging patterns pf eeoppien and

eeneumer use, Hillsdale, NJ:Erlbaum.
Henke,L.L.,(1985). Perceptions and Use of News Media by
College Students. Qournel pf Breaegesting e Eleetrenic
Meeie,29, pp. 431—436.
Jeggres, L.(1978). Cable TV and Viewer Selectivity. Jeurnal
pf Breaeeeeping, 22, pp. 167-177.
Kaplan, S.J.(1978). The impact of cable television
services on the use of competing media. Spurnel of

Breeeeaeping, pp.155—165.

Katz, E., Blumler, J.G., & Gurevitch, M.(1973-1974). Uses

and gratifications research. Publie Qpinien QuerEerly,

37, pp.509-523.

Katz, E., Gruevitch, M., & Haas, H.(1973) On the use of mass

media for important things. Amerieen Soeielpgieel

ReviewI 38, PP. 164-181.

118

Kim, Y.Y. (1982). Communication and acculturation in L.A.
Samovar and R.E. Porter (eds). Iupepeuluupe;_
.gemmuuieeuipu. Belmont, Calif.: Wadsworth.

Kippax, S.,& Murray,J.(1980). Using the mass media:need
gratification and perceived utility, gpmmuuieepieu
reeeereh, 7, pp. 335-360.

Krugman, D.M., & Eckrich, D.(1982). Differences between
cable and pay-cable audiences. Qeurnel pf Agvertieing
Researchl 22, pp. 23-29.

Krugman, D.M. & Rust, R.T. (1993). The impact of cable and
VCR penetration on network viewinngssessing the
decade. Qeurnel pf Agvepeieing reeeareh, Jan/Feb, pp.
67-73.

Kubey, R.W.(1986). Television use in everyday life: Coping
with unstructured time. figurnel pf gemmunieeeien,
36(3), pp. 108-123.

LaRose,R.,& Atkin,D.(1988). Satisfaction, demographics, and
media environment predictors of cable subscription,
Jeuruel pf breadeeeping u eleeurenie megie, 32(4), pp.
403-413.

Lin, C.A. (1993). Adolescent viewing and gratifications in a

new media environment. Meee eemm, Review, 20, pp.39-50.

119

Lometti, G.E., Reeves,B., & Bybee,C.R.(1977). Investigating
the assumptions of uses and gratifications research,
gemmunieeeien Reeeereh, 4, pp. 321-338.

Mazino,S.(1987). An exploratory analysis of the impact of
VCR movie cassette use and pay-TV movie subscription on
movie theater attendance, Meee epmmunicepien reeeerch,

pp. 30-35.

McGuire, W.(1974). Psychological motives and communication

gratification. In J.G. Blumler & E. Katz(Eds.), The

ueee pf meee eemmunieeeieue; gurren; perepeeuivee en
gretificeeipns reseereh (pp. 167-196). Beverly Hills:

Sage.

McLeod, J.M., & Becker, L.B.(1974). Testing the validity of
gratification measures through political effects
analysis. In J.G. Blumler & E. Katz(Eds.), Ipe_ueee_pf
W:W
ggeuifieaeipne reeeereh, (pp. 137-164). Beverly Hills:
Sage.

McLeod, J.M., & Becker, L.B.(1981). The uses and
gratifications approach. In D. Nimmo & K.
Sanders(EdS.), k 1i i l ni i n

(pp.67-99). Beverly Hills: Sage.

120
McQuail,D.(1983). Mass Communicauien Theory. Beverly
Hills,CA:Sage.
Miller,M.M., & Reese, S.D.(1982). Media Dependency as
Interactioanffects of Exposure and Reliance on
Political Activity and Efficacy. Qemmunieepien Reeearch,
9, pp. 227-248.

Miyo,Y.(1983). The Knowledge-gap Hypothesis and Media
Dependency. In Bostrom,R.(ed.). gemmunieeeien Yeerpeek,
7, Beverly Hills,CA:Sage.

Morgan, M., Shanahan, J., & Harris,C. (1990). VCRs and the

effects of television: New diversity or more of the
same? In J.R. Dobrow (Ed.), Social_and_cultural_
eepeepe_ef_ygg_uee,pp. 107-123, Hillsdale,NJ:Lawrence

Erlbaum.

Nielsen Media Research. (1980, 1990). Nielsen Television
Index(NTI):National audience demographics report,
Estimates of Average Hours of TV usage per week,
November, 1980, 1990.

Palmgreen, P., & Rayburn, J.(1985). An expectancy value
approach to media gratifications. In R.E. Rosengren,
L.A. Wenner, & P. Palmgreen(Eds.), M i r ifi ions

reeeereh (pp.61-72). Beverly Hills, CA : Sage.

121

Palmgreen,P.(1984). Uses and gratifications: a theoretical
perspective, in Bostrom,R.N (ed.), QQEEEQIQQLIQQ
yeeppppk, 8, Beverly Hills, CA : Sage.

Palmgreen, P., & Rayburn, J.D.(1982). Gratifications sought
and media exposure:an expectancy value model,
gemmunigepien Reeeereh, 9, pp. 561-580.

Palmgreen,P., & Rayburn,J.D.(1979). Uses and gratifications
and exposure to public television: a discrepancy
approach, gemmuuieeuipn reeeereh, 6, pp. 155-180.

Palmgreen,P., Wenner,L.A., & Rosengren,K.E.(1985). geee end

r ifi i n R r h:Th P T n Y r , in

Rosengren et al.,(eds), Media Gratifications
Research.Beverly Hills, CA : Sage.

Perse, E.M. & Courtright, J.A.(1993). Normative images of
communication mediazMass and interpersonal channels in
the new media environment. Humeu gemmupieeeien

Reeeereh, 19. pp. 485-503.

Reagan, J.(1982). Ef f l 1 vi i n n w se.
Unpublished manuscript, University of Michigan.
Rosenberg, M. & Hovland, C.I. (1060). Cognitive, affective,
and behavioral components of attitudes. In M.J.

Rosenberg, C.I. Hovland, W.J. McGuire, R.P. Abelson,

122

and J.W. Brehm(Eds), Attitudes organization and change.

New Haven, Conn,:Yale University Press.

Rosengren,K.E., & Windahl,S.(1977). Mass media use:causes

and effects. gommunicauipne:International Qeurnal of

eemmuniceeien reeeereh, 3, pp. 336-351.

Rothe, J., Harvey, M., & Michael, G.(1983). The impact of
cable television on subscriber and non-subscriber
behavior. Qeurnelef Aeverpieing Reeeareh, 23(4), pp.
15-23.

Rubin, A.(1984). Ritualized and instrumental uses of
television. Qpuppe;_pfi_gpmmuuieeuippp 34(3). PP. 67-77.

Rubin,A.M., & Rubin,R.B.(1981). Age,context, and television

use. Journal_ef_8readcastins. 25(1). pp- 1-13-

Sachar, E.B. (1982). Reflection on the future of the daily

newspaper. Luzeeemeuupgeeeezeh, New York: Goldman Sach
Corporation.
Severn. W.J..& Tankard.Jr. (1988). Communicatien_lheeries.

2nd ed., White Plains, NY: Longman.

Shaver,J.L.(1983). The uees pf eeple TV, Unpublished
Master's Thesis, University of Kentucky. Lexington,
KY.

Sparkes,V.M.(1983). Public Perception of and Reaction to

123

Multi-Channel Cable Television Service. Journal of
greedeeeting, 27(2), pp. 163-175.

TV dimensions '94. (1994). Media Dynamics, inc. New York,
NY.

Umphrey, D.,(1988). Segmenting the Cable Audience by Reason

for Subscribing, qurneliem Queruerly, 66, pp. 973-976.

Wakshlag, J., & Webster, J.(1986). Qn euplein giffereneee:
Ielevieien viewing iu preeeeeet, peeie ene pey eeple
hpueehpleep Unpublished manuscript.

Walker, J.R. (1988). Inherence effects in the new media
environment. Jeupnel pf Ereeeeeeeing e Eleeprenic
Megie, 32, pp. 391-401.

Webber, J.G. & Wakshlag, J. (1985). Measuring exposure to
television. In D. Zillmann & J. Bryant(Eds), ive
exposure_t9_senmucatien_(pp. 35-62). Hillside, NJ:
Lawrence Erlboam Associates.

Webster,J.G., & Wang,T.Y.,(1992). Structural Determinants of

Exposure to TelevisionzThe Case of Repeat Viewing.

Ieurnal_9f_BrnadSestins_e_Elestrenic_Media, 36(2). pp-
125-137.

Webster,J.G., & Newton,G.D.(1988). Structural Determinants

of the Television News Audience. qupue; of

124
Breedeaeting & Electrenic Media, 32(4), pp. 381-389.

Webster, J.G.,(1986). Audience Behavior in the New Media

Environment. Qeurnel pf gommunication. 36(3), pp.77-91.

Webster, J.(1983). Th im f c bl an a ca le
eelevieieu pp leeel seepion eueiencee, Report for the

National Association of Broadcasters, Washington, D.C.

Webster, J., & Wakshlag, J.(1983) A theory of program

choice. Qemmunieegipn Reeeereh, 10(4), pp. 430-447.

Wenner,L.A.(1983). fireeifieeuipns epugh; end epueineg; meeel
apecifisatien_and_theereti§al_dexelepment. Presented at

the annual meeting of the international communication
association, Dallas.
Zillmann, D., Hezel, R.T., & Medoff, N.J.(1980). The effect

of affective states on selective exposure to televised

entertainment fare. Ieurnal_ef_8pplied_§921al_

.EEYQQQIQSX. 10. PP-323-339-

NSTQT EU

"IIIIIIIIIIIZIIIIIIII IIII IIII IIII IIII III IIIII II