«

(R v \
a‘ﬁw-‘éié‘a
55* .4 “A “E"
iixgﬁta}: ’3-

 

 

 

 

 

 

mcmam sure on

lI/Il/lI/l/ ll/I/ll/I/llillllllTill/ll

3 /
THESI'S " 1293 01409 1353

 

 

This is to certify that the
thesis entitled
Advertising's Effect
on

Men's Gender Role Attitudes

presented by

Jennifer Garst

has been accepted towards fulfillment
of the requirements for

Master Wdegree in PSXCh0109Y

Lawn-wages

ajor professor

Date W 3|, ms’

 

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

LIBRARY
M'CMQan State
Unlverslty

 

 

 

PLACE ll RETURN DOXbmnmmbchockouMywncud.
1’0 AVOID FINES mun onor [adored-106m.

DATE DUE DATE DUE DATE DUE
49+ APR 2 8 .998] '

 

 

 

 

 

, t
l

 

 

yWZW—Tgﬁ
3132‘ 9 W

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-
_- -
---

muucmmmmmommmlm

 

 

 

ADVERTISING’S EFFECT ON MEN’S GENDER ROLE A'I'I‘ITUDES
By

Jennifer Garst

A THESIS

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

MASTER OF ARTS

Department of Psychology

1995

ABSTRACT
ADVERTISING’S EFFECT ON MEN’S GENDER ROLE ATTITUDES
By

Jennifer Garst

Within the framework of the social judgment and social learning theories, it was
hypothesized that men will only be affected by media portrayals of other men if the
portrayals broaden but do not contradict existing attitudes toward the male role. This
prediction follows from the assumption that men will only be inﬂuenced by male models
who are favorably perceived. Demographic similarity was also thought to increase the
inﬂuence of the media portrayals. More traditional and less traditional men viewed two
magazine advertisements that contained male media models who varied in terms of their
level of masculinity and femininity and whether they were the same age or much older
then the male participants. Results indicated that participants’ liking of the models did
not exert a uniform eﬁ‘ect on their male gender role attitudes. More traditional men
reported more traditional attitudes after they were exposed to advertisements containing
models who had an intermediate level of femininity, whereas less traditional men
reported more traditional attitudes when exposed to the least feminine male models.
Only more traditional participants were sensitive to the age of the media models; they
were aﬂ‘ected more by the younger media models than the older models. Results were
discussed in terms of the long-term, cumulative inﬂuence of media depictions on

individuals’ attitudes.

ACKNOWLEDGMENTS

Through the long and arduous process of completing this thesis, many people
have been instrumental in keeping my spirits, energy, and conﬁdence on an even keel.
First, and foremost, I would like to thank my advisor, Galen V. Bodenhausen, for his
never ending supply of guidance and support. In addition, I would like to thank my
thesis committee members, Norbert Kerr and Linda Jackson, for their valuable advice
and suggestions. Bradley Greenberg, Sandi Smith, and Judy Berkowitz also deserve
thanks for their assistance on earlier drafts of this thesis.

This thesis would not have been possible without the help of a number of
undergraduate research assistants who helped me gather data over two long semesters.
My thanks go out to Lisa Campbell, Brian Curneal, Nickie DeLaRosa, Moonson Elliot,
Corrie Higgins, Leslie Rabaut, Brian Richardson, Nicole Rieger, Christopher Rocco,
Adam Stevenson, Glenn Stutzky, and Patrick Taylor.

Since only those who have had the same experiences can truly understand what
one is going through, I would like to thank my fellow graduate students: Carole, Dale,
Deana, Julie, Mike, Ruth, Susan, Todd, and especially Donna. Lastly, I would like to
thank Regina, my downstairs neighbor, for providing a place to seek reﬂrge when I

desperately needed to take a break and receive encouragement.

iii

TABLE OF CONTENTS

List of Figures ....................................................... v
Introduction ......................................................... 1
Message Discrepancy ............................................. 6
Demographic Similarity ........................................... 9
Overview ..................................................... 13
Method ............................................................ 17
Participants and Design .......................................... 17
Materials ..................................................... l7
Stimulus Magazine Advertisements ............................ 17

Gender Attitude Scale (GAS) ................................ 20

Model Perception Scales (MPS) .............................. 21

Procedure .................................................... 21
Screening ............................................... 21

Main Experiment ......................................... 22

Results ............................................................ 24
Manipulation Checks and Preliminary Analyses ........................ 24
Eﬂ‘ects of Gender Role Attitudes on Perceptions of Media Models .......... 29
Eﬁ‘ects of Advertisements on Subsequent Male Gender Role Attitudes ....... 32
Discussion ......................................................... 47
Summary of Findings ............................................ 56
Implications ................................................... 57

Future Research ................................................ 60
Conclusions ................................................... 61
Appendices ......................................................... 63
Appendix A: Stimulus Magazine Advertisements ....................... 63
Appendix B: Subset of Gender Attitude Survey (GAS) .................. 76
Appendix C: Model Perception Scales (MPS) ......................... 78

List of References ................................................... 81

iv

LIST OF FIGURES

Figure 1: Distribution of average male gender role attitude scores ................ 25
Figure 2: Overall Perception of Models by original attitude and age of models ....... 31
Figure 3: Adjusted Mean GAS Score by original attitude and type of portrayal ...... 35
Figure 4: Moral Superiority of Women by original attitude and type of portrayal ..... 37
Figure 5: Moral Superiority of Women by original attitude and age of models ....... 39

Figure 6: Men’s Role for more traditional men by age of models and type of portrayal . 41
Figure 7: Men’s Role for less traditional men by age of models and type of portrayal . . 42
Figure 8: Work Role for more traditional men by age of models and type of portrayal . 45

Figure 9: Work Role for less traditional men by age of models and type of portrayal . . 46

Introduction

Individuals within each society must learn the social roles that they are expected
to ﬁrlﬁll and the expected behaviors associated with those roles. Many of these role
obligations are determined according to an individual’s sex (e. g., mother, father, sister,
brother, etc.) Sex is used to refer to the biologically based categories of male and
female, whereas gender is used to refer to the psychological aspects associated with the
biological objects of male and female. Overall, it is often thought that individuals rely on
their culture’s conceptions of masculinity and femininity when developing their schemata
about gender role appropriate behavior. This gender role appropriate behavior and
related attitudes may be acquired through observation of models, whether in-person or
observed in the mass media (e.g., Deaux, 1985; Remafedi, 1990; Tan, 1985; Vaughan &
Fisher, 1981). In turn, gender roles have been found to inﬂuence career planning,
intimate relationships, sexual behavior decision making, and attitudes toward parenthood,
to name just a few behaviors and attitudes (Greenberg, 1990; Remafedi, 1990).

The role of the mass media in the United States with regard to attitudes toward
gender roles has become an area of considerable interest and concern in the past quarter
century. Analyses of gender portrayals have found predominantly stereotypic portrayals
of males and females within the contexts of advertisements (print and television),
magazine ﬁction, newspapers, child-oriented print media, textbooks, literature, ﬁlm, and
popular music (Busby, 1975; Pearson, Turner, & Todd-Mancillas, 1991). Unfortunately,
most of the research done to date has focused primarily on the female gender role. A
review of research on men in the media suggests that, except for ﬁlm literature, the topic

of masculinity has not been addressed adequately (F ejes, 1989). One study, a

2

longitudinal content analysis of nine magazines in 1959, 1969, and 1979, did ﬁnd that
advertising featuring men appears to be slowly moving toward decreased gender role
stereotyping. However, in traditionally male magazines such as Esquire and W
Stream, the proportion of advertisements depicting men in “manly” activities did not
decrease as much as they did in traditionally female and general interest magazines
(Skelly & Lundstrom, 1981).

The mass medium that has been the focus of most research and of dominant
concern has been television. Television has become a widespread part of everyday life.
In the average American household, the television is turned “on” for almost seven hours
each day and the typical adult or child watches two to three hours of television per day.
It is estimated that the average child sees 360,000 advertisements by the age of eighteen
(Harris, 1989). Despite some progress, research has consistently shown that television
programming and advertisements portray men and women in imbalanced and stereotypic
manners, with men as more dominant and women as more nurturant (Durkin, 1985a;
Lovdal, 1989; Signorielli & Lears, 1992). Although some changes in portrayals have
been made, women are still seen predominantly at home, male voice—overs are used more
ﬁequently, women are younger than their male counterparts, and women are over
represented in commercials for personal hygiene and cosmetic products. A content
analysis by Lichter, Lichter, & Rothmann in 1986 found that men consistently outnumber
women on television by more than two to one (cited in Pearson et al., 1991). Lehrer
(1989), in an opinion piece, suggests that overall, such shows as “thirtysomething” do
appear to present a new image of manhood but, he laments that sensitive, nurturing male

characters are hard to ﬁnd in television programming. In general, men are portrayed as

3

calm and cool, self-conﬁdent, decisive, and emotionless (Harris, 1989). Although Harris
posits that this may be positive in many ways, it also suggests to a man that he is not a
“real” man unless he is able to deny his feelings.

The sex-typed portrayal of men and women becomes more important in light of
the fact that numerous studies have revealed that the mass media play a very important
role in the socialization process for both children and adults (Greenberg, 1990; Remafedi,
1990; Tan, 1985). Socialization is deﬁned as the process whereby people learn about
their culture and acquire some of its values, beliefs, perspectives, and social norms
(Signorielli, 1990). In essence, socialization is the way in which individuals come to take
on the behaviors and values of the group to which they belong. As such, people are
socialized to consider certain behaviors to be gender appropriate. This learning may be
accomplished via interpersonal or mediated channels.

The social learning theory posits that the media provide models whose behaviors
and attitudes may be imitated by their audiences (Bandura, 1986; Greenberg, 1990; Shaw
& Costanzo, 1982; Tan, 1985). More speciﬁcally, Bandura’s social cognition theory
suggests that observational learning occurs when observers acquire cognitive skills and
new behaviors by observing the performance of others (Bandura, 1986). The attention
given to the modeled activities determines what is observed in the activities and what
information is gained from the modeled events. According to Bandura, one of the factors
inﬂuencing the attention paid to the modeled activities is the “sway of attention” (p. 54).
This factor suggests that observers tend to seek out models who are interesting and
otherwise rewarding, whereas they ignore or reject models who have unattractive

qualities.

4

Atkin and Greenberg (1980) outline the conditions for optimal learning ﬁom the
mass media as: availability of content; frequent and consistent presentation of content;
exposure to content; lack of real world experience with content via other agents of
socialization such as family and friends; and few personal experiences inconsistent with
the media portrayal. A number of these conditions appear to be met in the manner in
which the media portray traditional gender roles. Thus, individuals may learn gender
role appropriate behavior vicariously through observation of mass media models.

The motivation for following what is learned ﬁ'om the mass media can be
explained by what Geis, Brown, Jennings (W alstedt), and Porter (1984b) describe as
“cultural demand characteristics” (p. 513). Although the authors were referring to
television, it can be argued that all mass media portray unarticulated but subtly implied
expectations about appropriate forms of behavior. Mass media carry the demand in the
form of presentations to the viewers of what appears to be social consensus about gender
roles. What makes these demands so powerful, according to Geis et al. (1984b), is that
on a macro level they are broadcast publicly and so appear to represent “everyone.” In
addition, on a micro-level the characters are portrayed as expecting and accepting their
complementary gender roles and thus validate the gender roles within the scenario. This,
according to Bandura’s social cognition theory (1986), is an example of vicarious
reinforcement. Vicarious reinforcement occurs when viewers perform the same behavior
as the model because they expect that they will receive the same reinforcement as does
the model. Likewise, Greenberg (1990) posits that one of the motives for young people
to watch most media is to ﬁnd out what society expects from them, to learn how to act

and dress appropriately, to learn about what could happen to them, and to see how other

5

people deal with issues that arise in various situations.

Research has shown that individuals do, in fact, imitate gender role behavior
based on media portrayals. Correlational studies have consistently shown that there is a
relationship between traditional gender role attitudes and heavy viewing of television by
children (e.g., Frueh & McGhee, 1975; Signorielli & Lears, 1992). In one Canadian
community without television until 1974, Kimball (1986) found that children’s sex-role
attitudes became more sex-typed after the introduction of television and became more
similar to children ﬁ'om a community that already had access to television. Numerous
other studies have shown that children’s gender role attitudes can be inﬂuenced by
counter-stereotypic gender portrayals in ﬁlm (e. g., Flerx, Fidler, & Rogers, 1976; Jeﬂ‘ery
& Durkin, 1989; Johnston & Ettema, 1982; Vaughan & Fisher, 1981). One of these
studies found a signiﬁcant, though smaller effect, when using counter-stereotypic
portrayals in picture books rather than ﬁlm (Flerx et al.).

Unfortunately, relatively few studies have examined media’s inﬂuence on adults’
gender role attitudes. Studies have shown that advertising that portrayed women and
men in reversed gender roles had a positive impact on women’s achievement emphasis
(Geis et al., 1984b), level of independence, and self-conﬁdence (Jennings (Walstedt),
Geis, Florence, & Brown, 1980). Men, on the other hand, did not change their
achievement emphasis when they viewed the gender role reversed commercials (Geis et
al., 1984b). Likewise, Jensen (1989) found that viewing counter-stereotyped images of
males in magazine advertisements did not cause men to report more liberal attitudes
about the male role. However, it can be argued that Jensen’s study may have been

inﬂuenced by demand characteristics since the gender role attitude scales were given to

6

participants at the same time as the advertisements.
Message Discrepancy

Durkin (1985b), in part two of his three-part review of television and sex-role
acquisition, has criticized most of the research on the viewer-media relationship. He
posits that most approaches work under the assumptions that (1) the eﬁ’ects are linearly
associated with either the amount of sex—typed representations and/or the amount of time
spent viewing the medium, and (2) that the content “irresistibly shapes” the perceptions
of the viewer. Durkin’s mq'or criticism of the linear eﬂ’ects approach is that it fails to
differentiate among the messages or images of the medium in terms of their salience to
the viewers. According to Durkin, most research on the impact of sex-roles only looks at
the overall impact of the medium but does not investigate the individual diﬁ‘erence
variables that may moderate the impact. Lending support to Durkin’s view is one study
in the domain of racial attitudes that shows that viewers’ preexisting attitudes moderate
their reactions to a television program. In an investigation of the role of Archie Bunker
in “All in the Family,” it was found that high prejudiced as compared with low
prejudiced white viewers tended to identify more with the character of Archie Bunker,
saw him as winning rather than as ridiculed, and saw nothing wrong with his use of racial
and ethnic slurs (Vidmar & Rokeach, 1974). Likewise, Eisenstock (1984) found that
androgynous boys and girls tended to be as likely as feminine girls and boys and more
likely than masculine children to identify with nontraditional televised models. It has
even been found that preexisting beliefs regarding the persuasiveness of stereotypes in
media depictions of women were conﬁrmed when individuals rated magazine

advertisements (Jones & Miller, 1993, April).

7

Durkin (1985c) posits in part three of his review that counter-stereotyped
portrayals are more effective in bringing about attitude change when they attempt to
build on an existing understanding of the social world rather than contradicting or
undermining it. For example, Geis et al. (1984b) hypothesized in their study that “four
commercials showing subservient males [are] expected to have no measurable impact
against men’s 20 years’ internalization of unambiguous life planning” (p. 515).
Likewise, Jeffery and Durkin (1989) found that the power held by a counter-stereotypic
male character moderated children’s reactions to that character. Less powerful male
characters elicited less positive reactions than powerful characters. The authors posit that
the practical implication of their study is that images of males intended to promote more
egalitarian lifestyles are likely to be more favorably received when the counter-
stereotyped characters are depicted as powerﬁrl.

It can be argued within the social learning paradigm that how people perceive a
portrayed character will mediate the impact that character has on viewers’ attitudes.
Within the gender role domain, it may well be that gender role portrayals will affect
viewers’ attitudes only if viewers favorably perceive the portrayed character. If one ﬁnds
a character interesting or otherwise rewarding, then one is more likely to emulate that
character’s interests or behaviors than if one is shown a character that is not perceived
favorably (e.g., Bandura, 1986).

An explanation for this phenomenon may be found within the framework of the
social judgment theory (Hovland, Harvey, & Sherif, 1957). The social judgment theory
posits that an individual’s attitude change is dependent upon the individual’s assessment

of the position advocated in the persuasive attempt relative to his/her own position.

8

Attitude change, thus, is considered to be a two-part process. First, after exposure, the
individual makes a judgment about the position advocated in the message relative to
his/her own position on the issue presented. Then, after this initial judgment is made,
attitude change may or may not occur. The magnitude of the attitude change will be
dependent upon the degree of discrepancy between the individual’s initial position and
the position advocated by the message.

The individual’s attitude serves as an anchor or reference point ﬁ'om which the
message is judged. For each individual, messages may be considered to fall within one
of three zones, depending on the amount of discrepancy between his/her initial attitude
and the advocated attitude. The ranges, according to the social judgment theory, are: the
latitude of acceptance, the latitude of rejection, and the latitude of noncommitrnent.
Positions that the individual ﬁnds to be acceptable or most similar to his/her initial
attitude would fall into his/her latitude of acceptance. Positions the individual ﬁnds to be
unacceptable or totally discrepant from his/her own attitude would fall into his/her
latitude of rejection while positions the individual considers neither acceptable nor
unacceptable (mildly discrepant) would fall into his/her latitude of noncommitrnent. The
size of these zones is thought to be dependent upon the level of personal signiﬁcance or
ego-involvement the issue has for the individual. That is, the higher one’s ego-
involvement on a speciﬁc issue, the smaller the range of acceptable positions and the
larger the range of unacceptable positions and positions one is ambivalent about.

In addition, social judgment theory incorporates an individual’s perceptual
reactions to the position advocated in the message. These perceptual distortions are

termed assimilation and contrast effects. Assimilation eﬂ‘ects occur when the individual

9

perceives the message to advocate a position closer to his/her own than it actually does.
The individual subjectively minimizes the discrepancy between his/her attitude and the
advocated position. Contrast effects occur when the individual perceives the message to
advocate a position ﬁrrther from his/her own than it actually does. The individual
subjectively maximizes the discrepancy between his/her attitude and the advocated
position. The most attitude change occurs when a message falls into an individual’s
latitude of noncommitment but is close to her/his latitude of rejection.

In order to create attitude change from the standpoint of the social judgment
theory, the message must be discrepant from one’s initial attitude but not so discrepant as
to be rejected. Persuasion will only occur if the message is accepted, since a message
perceived as too discrepant ﬁ’om an individual’s original attitude will be rejected
outright. On the other hand, in order to create any attitude change at all, the message
must differ at least somewhat ﬁom one’s initial attitude. A message that is identical to
one’s present attitude will not provide a reason to change the present attitude. It thus
appears that the most persuasive message would be one that is the most discrepant from
an individual’s initial attitude but which stops just short of being rejected because it is too
discrepant.

Demographic Similarity

The attitude change and attraction literature suggests that observers tend to be
attracted to models who are similar to themselves (e.g., Berscheid, 1985; McGuire,
1985). Similarity is thought to increase persuasion through its impact on source
credibility and attractiveness. A review by Hass (1981) of the impact of source factors in

persuasive messages suggests that similarities irrelevant to inﬂuence attempts result in

10

greater persuasion because of the increased attractiveness of the message source.
Irrelevant demographic similarities (i.e., race, age, and occupation) are thought to
promote persuasion because the audience is able to identify with the message source
(Tan, 1985). Tan suggests that because many diverse characteristics are shared, receivers
perceive the source to be “one of us” or “our kind of people” and thus infer tlmt they
share common needs and goals with the message source (p. 116).

Unfortunately, the demographic similarity-persuasion relationship appears to be
much more complex than the above arguments indicate. For instance, it has been found
that the dimensions of humor, attractiveness, strength, and activity are strong predictors
of children’s desires to ‘Vvant to be like” and “want to do like” a television character,
while the demographic dimensions of sex and age are not (Reeves & Greenberg, 1977).
Furthermore, a review by Sirnons, Berkowitz, and Moyer (1970) points out that there are
“an inﬁnite number of dimensions” of similarity-dissunilarity (p. 3). Sirnons et al. posit
that since similarities or dissirnilarities to another person in regard to dimensions such
age, occupation, income, or education will not have uniform eﬁ’ects across all people,
generalizations about “similarity” as one construct are impossible. A second point
Sirnons et al. (1970) and Hass (1981) make is that similarities probably do not inﬂuence
persuasion directly. Rather, similarities only inﬂuence persuasion indirectly through
their impact on the receiver’s liking for the communicator or the receiver’s perception of
the communicator’s credibility. In the case of demographic similarity, the impact on
persuasion has been suggested to be moderated by liking for the communicator, so that
the demographic similarity-persuasion relationship will tend to be weaker and more

easily overwhehned than would a direct link.

11

Although the importance of the attitude similarity-attraction relationship has been
veriﬁed in laboratory situations, ﬁeld studies have found that demographic similarity is
an important component of attraction (Berscheid, 1985). Berscheid cites a study by Hill,
Rubin, and Peplau (1976) that found that, in comparison to couples who broke up over a
two-year period, couples who remained together were more similar in respect to the
demographic characteristics of age, intelligence, and physical attractiveness but not in
regards to attitudes. A large body of literature on homogamy documents this trend (e. g.,
Qian & Preston, 1993).

Likewise, research on the determinants of social distance has found that race
rather than belief structure is a more important determinant of social distance (Triandis,
1961). In fact, Triandis found that race was four times as important as philosophy (belief
system) in determining the social distance for sixteen stimulus persons. Other research
has found that while ideological similarity tends to be more important for abstract
evaluation of people, demographic similarity is more important for less abstract
behavioral acceptance (i.e., eating with and marrying) (Robinson & Insko, 1969).
McGuire suggests in his 1985 review of attitudes and attitude change that these results
derive, at least in part, from the fact that people infer that sources demographically
different from themselves also differ ideologically.

Studies examining the reduction of avoidant behaviors provide some direct
support for the persuasive impact of demographic similarity. Elementary school
children’s refusal to touch a snake and their attitudes toward snakes were more
signiﬁcantly affected by viewing ﬁlms depicting a same-age model handling a snake than

by a ﬁlm with an adult model (Kornhaber & Schroeder, 1975). Likewise, undergraduate

12

university students were more affected by imagining models of the same age and sex
handling snakes versus much older, opposite sex models in terms of the students’
approach behaviors, arousal, anxieties, and attitudes toward snakes. These results were
found both immediately at posttreatment and two to three weeks later (Kazdin, 1974).
Contrary to these ﬁndings, a study by Bandura and Barab (1973) found that for snake-
phobic adults watching either a mixed-sex group of three very young children or three
adults did not diﬁ‘erentially impact the reduction in phobic behavior and fear arousal.
The authors suggested that there were different mechanisms mediating the reaction to
children and adult models. They point to the fact that adult models appeared to help the
participants extinguish their fears whereas child models led participants to only partially
extinguish their fears. Bandura and Barab posit that while the changes due to the adult
model are interpretable as an extinction process, the results from child modeling are more
attributable to motivational inducements.

In a variety of other domains it has also been found that demographic similarity
increases the impact of persuasion. In terms of observational learning of aggressive
behaviors, male peers have the most immediate inﬂuence on boys and girls’ aggressive
behaviors toward a plastic doll (Hicks, 1965). Likewise, the diﬁirsion of innovation
literature has found that increases in the degree to which a source-receiver pair is similar
in certain attributes are associated with increases in the success of diﬂ‘usion. In his book
WWW Rogers (1973) claims that there is a
strong tendency for individuals to exchange information with those who are most similar
to themselves. For instance, Rogers cites research that shows that (1) age and social

status make a difference in political discussions in the U. 8. presidential election;

l3

(2) Chicago ghetto dwellers were more likely to discuss family planning ideas with
others of like social status, age, marital status, and family size and; (3) Indian peasants
mostly interact with other villagers of similar caste, education, and farm size. Rogers
posits that when an individual forms a positive attitude toward an innovation, sources and
channels similar to the individual are used most frequently because they are enough like
the individual to serve as a comparable model and thus have “safety credibility” (p. 123).
Therefore, this safety credibility has a positive impact on the diﬂirsion of innovation.
Overall, this study and the other research cited are generally consistent with the thought
that demographic similarity increases the impact of persuasion and, more speciﬁcally, of
modeled behavior.

Overview

A direct test of the link between stereotyped advertisements and gender role
attitudes would ideally sample a group who had never been exposed to gender role
stereotypes in advertisements or other media sources. Since such a sample is not
possible, an alternative hypothesis will be tested. The idea is that if stereotyped
advertisements serve as a script for traditional gender role attitudes, advertisements that
break the stereotype will lessen traditional stereotypical gender role attitudes.

It could be argued that the few attempts made to change men’s gender role
attitudes have failed because the researchers have created stimuli that are too discrepant
ﬁom mens’ initial attitudes. Admittedly, it might be diﬁcult to create nontraditional
stimuli that are acceptable to men because the stereotype of men is very narrowly
deﬁned. For instance, Hort, Fagot, and Leinbach (1990) found that the male stereotype

was described by both men and women in much more stereotypic terms than was the

14

female stereotype. Likewise, performing “appropriate” gender role behavior is probably
viewed as much more signiﬁcant to men than women. For example, it has been found
that children in elementary school react more negatively toward boys than girls
exhibiting cross-gender ﬁiend and toy preferences (Carter & McCloskey, 1983-1984).
Interestingly, even though most children indicated that cross-gender behavior in their
peers was not wrong, they also indicated that they would prefer not to associate with
children who violated sex-role norms. Lastly, it has been found that men who were
shown in low-status roles were seen as having negatively viewed traits usually associated
with women—emotional, dependent, submissive, content, and being followers rather than
leaders (Geis, Brown, Jennings (Walstedt), & Corrado-Taylor, 1984a). In order to be
eﬁ‘ective, the stimuli used to change the gender role attitudes about men must vary
somewhat but not too much from the rather narrow male stereotype and, in addition,
maintain the power usually associated with the male role. In fact, Kanner (1990, May
21) posits in a magazine article addressing men in advertising that men are simply not
willing to give up their traditionally masculine style to become “domesticated wimps” (p.
20). To support her argument she cited Barbara Durham, associate director of consumer
resources of F CB/Chicago, a large advertising ﬁrm, as saying “role are changing, but we
found it’s important not to rob men of their masculinity” (p. 20).

Taken together, the available evidence on demographic similarity and persuasion
suggests that demographic similarity may increase the observational learning effects of
mass media portrayals of counter-stereotypic male behaviors and interests. It is likely
that a model who is demographically similar to university students (age early twenties

versus late sixties) will be perceived as more similar to the university student

15

participants, will be more liked, will be seen as a more attractive model, and ﬁnally will
promote greater acceptance of the modeled behavior. Thus male models displaying
somewhat discrepant male role behaviors and interests would have a greater impact on
the male gender role attitude of participants if the male models were of the same age as
the participants than if the models were much older.

Like the work by Geis et al. (1984b), Jennings (Walstedt) et al. (1980), and
Jensen (1989), this study will examine the effects of implicit gender role portrayals in the
media on the gender role attitudes of adults. This study and Jensen’s (1989) study also
both differ ﬁ'om previous studies, because they focus on the eﬁ‘ects of gender role
portrayals on men’s rather than women’s attitudes. In addition, the current study is
unique in that it examines the implicit persuasive effects of character portrayals in mass
media sources using the social judgment theory as a ﬁamework.

In this study it is hypothesized that men can be affected by media portrayals of
other men only if the portrayals broaden but do not contradict existing attitudes toward
the male role. This prediction suggests that men will only be inﬂuenced by male models
whom they perceive in a favorable light with respect to their existing male gender role
attitudes. In addition, demographic similarity should increase the impact of the media
portrayals. More speciﬁcally, it is hypothesized that:

1) The traditional attitudes regarding men’s interactions with women held by
traditional men will be decreased by gender role portrayals that are mildly discrepant
from their initial attitudes (androgynous). The attitudes of traditional participants should
not be signiﬁcantly inﬂuenced by traditional portrayals (masculine) because they are

already in agreement, nor will highly discrepant nontraditional portrayals (feminine)

l6

create a change because traditional participants will reject them.

2) This study will also explore the reactions of less traditional men. It is expected
that nontraditional (feminine) male portrayals, highly discrepant from a traditional man’s
initial attitude, will be less discrepant for a less traditional man and thus more acceptable.
However, even the nontraditional (feminine) portrayals are expected to be somewhat
discrepant for the less traditional participants because the portrayals will not contain any
behaviors and interests generally considered masculine. The androgynous portrayals
most likely will not be perceived as discrepant for less traditional men because they will
contain both masculine and feminine qualities. Therefore, it is tentatively predicted that
the androgynous portrayals will not signiﬁcantly aﬁ'ect less traditional participant
because they are not discrepant ﬁ'om the participants’ initial attitudes. Nontraditional
(feminine) portrayals may signiﬁcantly decrease the level of stereotypicality in the
attitudes of less traditional males because they are only mildly discrepant. Support for
this is derived from a study that showed that boys and girls with feminine preferences
tend to identify more with a male counter-stereotypical portrayal than do masculine
children (Eisenstock, 1984). It is expected that less traditional participants will not be
inﬂuenced by traditional (masculine) portrayals even though the portrayals will probably
be mildly discrepant ﬁom the participants’ initial attitudes. Although no clear evidence
was encountered regarding this issue, this prediction is made because it is likely less
traditional participants are constantly inundated with traditional (masculine) images in
the present-day media and may have developed resistance or defenses against these
traditional images.

3) It is predicted that the demographic similarity of the models in the

17

advertisements will also help to determine the degree of attitude change. This will occur
because when the models are similar to the participants in terms of age, they will be
perceived as more similar, will be more identiﬁed with, will be seen as more attractive,
and, ﬁnally, will promote greater acceptance of the modeled behavior than will dissimilar
models.

Method
Participants and Design

A total of 267 male undergraduates ﬁ'om introductory psychology classes
participated in the experiment in exchange for course credit. Responses of 55
participants were excluded ﬁom the data analysis because they could not be matched
with a previously completed screening survey. This left a total of 212 participants,
whose ages ranged ﬁ'om 17 to 46 with a mean age of 19.4 years.

Participants were randomly assigned to one of six experimental conditions,
yielding a 3 (type of media portrayal: feminine, androgynous, or masculine) X 2
(participants’ demographically similarity to the models: low versus high) X 2 (current
gender attitudes: traditional versus less traditional) between-subjects factorial design.
Materials

Stimulus Magazine Advertisements. Stimulus materials were color magazine
advertisements that pictured a typical male user of the product being advertised along
with a written biographical proﬁle of the product user. Advertisements for Dewar’ s®
scotch that can be found in currently circulating magazines were the models for the
stimulus advertisements. Advertisements featuring two different products were

developed for each of six conditions (See Appendix A). One advertisement for each

18

condition pictured a cup of coﬂ‘ee being ﬁlled with Delvecchio’s gourmet coffee
(Scenario 1) while the other advertisement featured a computer represented to be an EPS
Technologies personal computer (Scenario 2). Both advertisements listed the typical
male product user’s name (e. g., John Hadon), home (e.g., Chicago, Illinois), hobbies
(e.g., enjoys going to concerts), recent achievement (e. g., received a service award ﬁ'om
the social club he belongs to), name of the product he uses (e.g., Delvecchio’s gourmet
coffee), outlook on relationships, philosophy regarding marriage, age, and profession.
The makeup of the category of outlook on relationships as well as the category of
philosophy regarding marriage were varied across the feminine, androgynous, and
masculine media portrayals for each of the two product advertisements. Four
interest/role behaviors were used to form both the outlook on relationships and
philosophy regarding marriage categories. For the masculine media portrayals, each
category was largely based on role behaviors and interests which have been identiﬁed by
Orlofsky, Ramsden, and Cohen (1982) in the Revised Sex-Role Behavior Scale (SRBS-Z)
as more typical of men but considered desirable by both sexes. The feminine media
portrayals used role behaviors and interests based on those identiﬁed by Orlofsky et al. as
more typical of women than men but considered desirable by both men and women.
Lastly, the categories of outlook on relationships and philosophy regarding marriage for
the androgynous media portrayals each generally contained two behaviors/interests ﬁom
the masculine media portrayals for that particular product and two from the feminine
advertisement for that product. This was done because past research has found that
androgynous persons of either sex were perceived as having the greatest overall

employment potential compared to sex-typed men and women (Jackson, 1983).

19

Likewise, men and women who were described as relatively high in both expressive
qualities usually associated with women (kindness, sensitivity to others, emotional
responsiveness, and need for aﬁliation) and instrumental qualities associated mostly
with men (independence, assertiveness, and decisiveness) were perceived as more
attractive than individuals high in only one category (Spence, Deaux, & Helmreich,
1985)

The demographic similarity of the males portrayed in the advertisements,
compared to the participants, was also varied. Half the participants saw advertisements
that pictured a young man represented to be a college student either 18 or 20 years of age
(high demographic similarity to the college students who served as the participants). The
other half examined advertisements that featured a photograph of an older man who was
represented to be either 65 or 66 years old and retired (low demographic similarity to the
participants). The photos of the older and younger man associated with each product
advertisement were selected so that their perceived level of masculinity was comparable
based on pilot testing. Both the demographically similar and dissimilar portrayals
implied that the models were unemployed in order to avoid confounds created by power
diﬁ'erentials of the models. Past research has found that the power of the model
signiﬁcantly impacts the observational learning gained from media portrayals (Jeffery &
Durkin, 1989).

All attributes other than outlook on relationships, philosophy regarding marriage,
age, and profession were held constant across the different magazine advertisements for

each product.

20

Gender Attitude Scale (GAS). A total of thirty-three slightly modiﬁed items
were selected ﬁom the 109 items of the Gender Attitudes Survey (GAS; Ashmore, Del
Boca, & Bilder, 1992; see Appendix B) to measure participants’ attitudes toward the
relationships between men and women and the roles played by men in contemporary
American society. Participants’ responses were assessed using a 7—point scale with “1”
signifying “agree strongly” and “7” signifying “disagree strongly.” The items were
worded so that a high score sometimes reﬂected traditional attitudes and sometimes
reﬂected less traditional attitudes. The items were recoded so that high values indicate
less traditional attitudes. Items ﬁ'om six of the fourteen subscales identiﬁed by Ashmore
et al. were used: date initiative (e. g., the initiative in dating should come ﬁ'om the man),
women are superior (e. g., women are morally superior to men), traditional stereotypes
(e. g., compared to women, men tend to be cruel), chivalry (e. g., chivalrous gestures
toward women on the part of men should be encouraged), family roles (e. g., the husband
should have primary responsibility for the support of the family), and work roles (e.g., all
occupations should be equally accessible to both men and women). Participants also
answered thirteen items from these subscales that focused exclusively on women’s
gender roles but these items were excluded ﬁom all data analyses. In addition, items
ﬁom eight of Ashmore et al. ’3 subscales that did not deal with the interaction and
relationship between men and women were not used (e. g., homosexuality, female sex
initiative, male violence, political leadership, social change, day care, abortion, and

individual change).

21

Model Perception Scales (MPS). In order to measure participants’ perceptions
of the males portrayed in the magazine advertisements, a total of nine 5-point Likert-type
items were used (see Appendix C). The measures focused on participants’ liking of each
model (e. g., how much do you think you would like the person in the advertisement?),
similarity to each model (e.g., how similar do you perceive the person in the
advertisement to be to you?), and the social attractiveness of each model (e.g., how well
liked to you think this person is by others?) In addition, a series often 5-point semantic
differential scales were used to directly assess participants’ perceptions of each model
(e.g., masculinity, femininity, similarity, attractiveness, power, competence, and
normality).

Procedure

Screening. A total of 738 male introductory Psychology students were given the
thirty-three item subset of the GAS during the ﬁrst week of class. The participants were
divided by using a median split of their average GAS scores (Mdn = 4.46 on 7-point
scale); those in the upper halfwere classiﬁed as “less traditional” while those in the
lower halfwere classiﬁed as “traditional.”

Because multiple measures of participants’ gender role attitudes were collected, a
principle-axis factor analysis was performed on the GAS items (retaining factors with
eigenvalues > 1, using a varimax rotation) in order reduce the data into meaningful
components for the particular population under study. A total of six composites resulted
that generally paralleled Ashmore et al.’s (1992) original subscales (see Appendix B). A
ten-item composite labeled “traditional stereotypes” dealt with beliefs such that men are

more competitive, conﬁdent, and adventurous then women. A second composite, entitled

22

“women are superior” contained six items dealing with women’s moral superiority to
men (e.g., women are morally superior to men). A four-item “chivalry” composite
related to the appropriateness of male chivalry, while the “work role” composite
contained four items which addressed whether men and women have the similar work
abilities. The ﬁfth composite, entitled “men’s roles,” combined eight items from both of
Ashmore et al. ’5 family roles and date initiative composites. This composite dealt with
the role men play in both dating and family situations. Lastly, the factor analysis
revealed that there is one distinctive item called “child care” that addressed child care
responsibilities in a family (e.g., care of children should be shared equally by both
spouses).l

Main Experiment. Participants reported to the laboratory for a study on
magazine advertising. They were unaware of any connection with the previous screening
session. The sessions included groups of up to 15 males with male and/or female
experimenters. After being greeted by the ﬁrst experimenter, participants were told that
due to the short length of the study, they will also be asked to ﬁll out a questionnaire
needed as part of a graduate student’s dissertation. The experimenter then explained that
the ﬁrst study concerned magazine advertising and that they will be asked to look over
and provide their perceptions of and reactions to two advertisements. Participants were

told that the purpose of the study was to determine which features of magazine

 

' Examination revealed that most of the scores on the GAS subscales were highly
correlated with participants’ overall GAS scores. When the division between traditional
and less traditional participants was reexamined in the absence of the subscales scores
that had low correlation with the overall GAS scores (i.e., child care and chivalry
subscales), fully 97.2% of the participants retained their original designation as
traditional or less traditional.

23

advertisements most effectively promoted purchase intentions. Participants were then
asked to read and sign a consent form and to begin going through the experimental
booklets.

Each participant received an experimental booklet and a randomly selected folder
containing a set of two color stimulus magazine advertisements that had been randomly
sorted within the folder. Each folder contained advertisements with either feminine,
androgynous, or masculine biographical proﬁles of models who were either
demographically similar or dissimilar to the participants. Because experimental
conditions cannot replicate the lifetime of exposure to gender depiction in the mass
media, the problem was how to amplify the small impact expected to be gained ﬁ'om two
magazine advertisements. The solution used in this experiment was to ask people to
answer questions regarding each media depiction after viewing it. This approach was
adapted from a procedure used by Geis et al. (1984b) in their study of the impact of
television commercials as achievements scripts for women. Therefore, after viewing
each advertisement, participants were asked to ﬁll out the Model Perception Scale items.
In order to hide the true intent of the experiment, the participants were also asked to
answer questions that assessed their purchase intentions of the product (e.g., assuming
you had the money, how likely would it be for you to purchase the advertised product?),
their reactions to the product (e.g., overall, how much do you like the advertised
product?), and their reactions to the advertisement itself (e.g., overall, do you think the
magazine advertisement you just viewed is of high quality?) After participants answered
the questions regarding each of the two advertisements, they completed standard

demographic items. When all participants had ﬁnished the questionnaire, the

24

experimenter thanked them for participating in the advertising study, introduced them to
a diﬂ’erent experimenter who was in charge of the supposedly unrelated second study,
and then left the experiment room. The second experimenter had entered the room after
the ﬁrst experimental session had already begun.

The second experimenter then thanked the participants for helping out the
graduate student with her study and explained that the second questionnaire concerned
basic attitudes and values currently held by college students on a wide variety of issues.
The experimenter then administered new consent forms after which he/she handed out a
questionnaire that used a different format and type face than used for the advertisement
study. The new questionnaire contained the same subset of the GAS items ﬁlled out by
the participants previously. The GAS questions were interspersed with 21 gender
unrelated attitude measurement items. After the participants completed the
questionnaire, they were debriefed, thanked, and excused.

Results
Manipulation Checks and Preliminary Analyses

The appropriateness of viewing the traditional and less traditional gender role
attitude categories as signiﬁcantly different from each other was conﬁrmed by a one-way
analysis of variance (ANOVA). Unsurprisingly, those classiﬁed as less traditional
displayed signiﬁcantly less traditional male gender role attitudes than those who were
classiﬁed as traditional, [Ms = 5.02 versus 3.86, respectively, E(l, 210) = 350.23, p <
.001]. Given the fact that the distribution of participants’ average male gender role
attitude scores is slightly skewed to the more liberal side of the scale, it may be more

accurate to refer to traditional participants as “more traditional” rather than “traditional”

25

(See Figure 1).

DIstrlbutIon of average male gender role attitude scores
40

 

(a)
O
l

Froquencles
N
0
j

10l—

 

 

 

 

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
AverageGASseores

Figure 1: Distribution of average male gender role attitude scores

26

To determine whether the participants thought that there was a continuum of
masculinity and femininity between the feminine, androgynous, and masculine media
portrayals, and to ascertain whether participants perceived the younger models to be
more similar to themselves than the older models, repeated-measures AN OVAs were
used to examine participants’ perceptions of the masculinity, femininity, and similarity of
the two media portrayals as a function of the type of media portrayal, age of the models,
and original gender attitudes held. As expected, participants differentiated between the
types of media portrayals as a function of their femininity, E(2, 199) = 10.28, p < .001.
Participants perceived the feminine media portrayals to be signiﬁcantly more feminine
than the androgynous portrayals [Ms = 2.49 versus 2.15; E(1, 140) = 6.23, p < .05] and
the androgynous portrayals, in turn, were seen as signiﬁcantly more feminine than the
masculine portrayals [Ms = 2.15 versus 1.88; E(1, 139) = 3.96, p < .05]. Participants did
not, however, differentiate among the three types of media portrayals in terms of their
perceived masculinity, E(2, 200) = 1.40, p > .15 (Ms = 3.26 for feminine; 3.31 for
androgynous; 3 .46 for masculine). Rather, there was a two-way interaction between the
age of the models and the participants’ original gender role attitudes, E(1, 200) = 8.03, p
< .01. Whereas less traditional participants did not diﬂ’erentiate between the masculinity
of younger and older models [Ms = 3.31 and 3.52, respectively, RU, 108) = 2.24, p, >
. 15], more traditional participants perceived the younger models to be signiﬁcantly more
masculine (M = 3.44) than the older models [M = 3.06, E0, 100) = 5.50, p < .05].

Also, as expected, participants perceived the younger models to be more similar
to themselves than the older models, with means of 2.64 versus 2.34, E(1, 200) = 5.67, p

< .05. However, this was qualiﬁed by an unexpected interaction between the age of the

27

models and the original gender role attitude held by the participants, RU, 200) = 6.94, p <
.01. Although more traditional participants perceived the younger models to be
signiﬁcantly more similar to themselves than the older models [Ms = 2.82 versus 2.17,
respectively, EU, 1, 100) = 15.50, p < .001], less traditional participants did not
differentiate between the younger and older models’ similarity to themselves [Ms = 2.47
versus 2.49, respectively, E(1, 108) = 0.01, p > .15]. It was also found that participants
differentiated between the feminine, androgynous, and masculine portrayals in terms of
their perceived similarity, E(2, 200) = 3.03. p < .05. Participants rated the masculine
portrayals (M = 2.28) as signiﬁcantly less similar to themselves than the feminine
portrayals (M = 2.64) and as marginally less similar than the androgynous portrayals, (M
= 2.56), E(l, 138) = 5.51, p < .05 and E(l, 140) = 3.54, p = .062, respectively. There was
no difference between the perceived similarity of the feminine and androgynous
portrayals, E(1, 140) = 0.25, p > .15.

To discover whether the stimulus materials were able to avoid possible confounds
created by perceived power diﬁ‘erentials between young and old models, the perceived
power of the models was examined in terms of the age of the models, type of media
portrayal, and original attitudes held by the participants. Overall, participants tended to
perceive the older models to be more powerful than the younger models, with means of
3.01 versus 2.80, respectively, E(l, 200) = 3.12, p = .079.

Unexpectedly, participants appeared to respond diﬁ‘erently to the models in the
advertisements that featured Delvecchio’s coffee (Scenario 1) versus EPS computers
(Scenario 2). Participants’ direct ratings of the models revealed that they perceived the

models used in the coffee advertisements as more masculine and powerﬁrl and less

28

feminine than the models used in the computer advertisements (all as are < .01). These
scenario eﬁ‘ects were further qualiﬁed by scenario by age of models interactions (all ps
are < .001 ). Whereas participants saw the younger model in the coffee advertisements as
more masculine and powerful and less feminine than the younger model in the computer
advertisements, they did not diﬁ'erentiate between the older models used in the two
advertisements. In terms of the perceived similarity of the media models, participants
perceived the models in the coffee advertisements to be only marginally more similar to
themselves than the models used in the computer advertisement (p = .069). Generally,
the fact that participants tended to discriminate between the characteristics of the younger
but not the older models suggests they paid more attention to the younger models than
the older models.

The validity of using a single model perception composite measure for the 9-item
model perception scale (MPS) was veriﬁed by a principle-axis factor analysis (retaining
eigenvalues > 1, oblique rotation) on each of the two scenarios separately. The obtained
component for each scenario focused on participants’ liking for the model, the perceived
similarity of the model to the participants, and the social attractiveness of the model (see
Appendix C). Reliabilities of the composites were high, with Cronbach’s a = 0.91 for
the advertisement for Delvecchio’s coffee (Scenario 1) and Cronbach’s a = .90 for the
computer advertisement (Scenario 2). A higher average score value on the MPS
composite indicates that the participants perceived the models in a more favorable light
(i.e. ., they liked the models more, found them more similar to themselves, and found the
models more attractive) compared to those who had lower scores on the average MPS

composite.

29
Effects of Gender Role Attitudes on Perceptions of Media Models

A central assertion of the present study was that men will be most affected by
male media portrayals that they perceive in a favorable light. More speciﬁcally, it was
thought that gender role portrayals that broaden but do not contradict existing attitudes
toward the male role will have the most inﬂuence on men’s gender role attitudes. In
addition, it was thought that younger models within the media portrayals would have
more impact than older models because they would be more similar.

In order to assess whether participants’ perceptions of the advertisements were
inﬂuenced by their original male gender role attitudes as well as to determine whether
the media models’ ages and type of media portrayal had any inﬂuence, participants’
average MPS composites for both of the advertisements were examined using repeated-
measures analysis of variance (AN OVA). The analyses revealed that participants’
overall perceptions of the models varied as a function of the type of media portrayal
viewed, E(2, 196) = 12.58, p < .001. All participants displayed signiﬁcantly more
favorable views of the feminine (M = 3.13) and androgynous (M = 2.98) media
portrayals than the masculine portrayals [M = 2.58; E(1, 136) = 24.88, p < .001 and E(1,
137) = 11.89, p < .001, respectively]. The perceptions of the feminine and androgynous
portrayals did not signiﬁcantly differ, E(1, 137) = 1.71, p > .15. Contrary to what was
hypothesized, the effect for the type of media portrayal viewed was not qualiﬁed by an
interaction with the original male gender role attitude held by the participants, E(2, 196)
= .86, p > .15.

However, the original male gender role attitudes of the participants did interact

with the age of the models viewed in a pattern close to the one found when participants’

 

 

mod

 

pm

3.45

30

direct ratings of the models’ similarity were examined, E(l, 196) = 4.23, p < .05 (see
Figure 2). Whereas less traditional participants’ perceptions of the younger and older
models did not differ [E(l, 104) = 1.03, n > .15], more traditional participants tended to
perceive the younger models in a more favorable light than the older models, E(1, 100) =

3.45, p = .066.

31

Overall Perception of Models
By original attitude and age of models

 

U
l

Favorablsness of perception
- N
l ' ‘1

 

 

 

Less traditional More tradiionai
Original gender role attitude

Young I Old

Figure 2: Overall Perception of Models by original attitude and age of models

32

Again, participants appeared to respond differently to the advertisements for
Delvecchio’s coﬁ‘ee versus EPS computers. When participants’ overall perceptions of
the advertisements were examined in terms of the type of media portrayal, age of models,
and original male gender role attitudes of the participants, there was a trend for an
interaction between scenario and models’ ages, E( 1, 196) = 3.62, p = .059. Whereas the
perceptions of the older models remained constant across the two versions of the
advertisements, participants had signiﬁcantly more favorable perceptions of the younger
model in the coffee advertisement (Scenario 1) than in the computer advertisement
(Scenario 2). Direct ratings, discussed earlier, revealed this same pattern. Participants
saw the younger model in the coﬁ‘ee advertisement as more masculine, less feminine, and
more powerﬁrl than the younger model used in the computer advertisement while there
was no diﬂ'erentiation for the older models.

Effects of Advertisements on Subsequent Male Gender Role Attitudes

Determining whether participants were differentially inﬂuenced by the type of
media portrayals used in the advertisements (feminine, androgynous, or masculine) based
on their prior male gender role attitudes was of central interest in this study. This issue is
of particular interest given that participants’ perceptions of the diﬁ’erent types of models
used in the advertisements were not inﬂuenced by the original male gender role attitudes
they held. The fact that participants had the most favorable perceptions of the feminine
and androgynous models versus the masculine models suggests that participants’
attitudes may be least inﬂuenced by the masculine portrayals.

Also of interest was the extent to which participants’ gender role attitudes were

inﬂuenced by the similarity of the models’ ages to their own age. The most attitude

33

change was expected to occur when the participants could relate to a model whose age is
most similar to themselves. Unfortunately, the MPS composite and the single item
similarity rating suggest that demographic similarity does not uniformly affect all
participants but is moderated by participants’ original male gender role attitudes.
Whereas the less traditional participants’ perceptions of the models did not differ based
on the age of the models, more traditional participants had more favorable reactions to
the younger models versus the older models.

Because of regression to the mean of more traditional and less traditional
participants’ gender role attitudes ﬁom pretest to posttest, the interpretation of the
difference scores between pre and post attitude measures was complicated. To control
for the eﬁ‘ect of participants’ original attitudes on their post experimental male gender
role attitude responses, participants’ male gender role attitudes after viewing the
advertisements were analyzed using analysis of covariance (ANCOVA) with pretest
scores as a covariate. The average GAS score and each subscale (i.e., traditional
stereotypes, women are morally superior, chivalry, work role, men’s role, and child care)
were examined as a ﬁrnction of original gender role attitudes (more traditional versus less
traditional), type of media portrayal (feminine, androgynous, and masculine), and age of
the media models (young versus old), with the participants’ original male gender role
attitudes covaried out. Unsurprisingly, for each analysis participants’ original male
gender role attitudes were signiﬁcant covariates (gs < .001) of the gender role attitudes
they expressed after they had viewed the two advertisements.

As expected, more traditional and less traditional participants’ average posttest

GAS scores were diﬁ’erentially inﬂuenced by the type of media portrayals they viewed,

34

E(2, 199) = 4.31, p < .05 (see Figure 3). While the difference of less traditional
participants’ responses to the different portrayals was signiﬁcant [E(2, 107) = 4.80, p <
.01], there was only a marginal difference in more traditional participants’ responses to
the media portrayals [E(2, 99) = 2.37, p = .099]. As predicted, the feminine portrayals, as
well as the androgynous portrayals, caused less traditional participants to display less
traditional average GAS scores as compared to those who viewed the masculine
portrayals, F( 1, 72) = 7.97, p < .01 and F(1, 70) = 6.62, p < .05, respectively. The
average GAS scores for less traditional participants who saw the feminine and
androgynous media portrayals did not differ, F(1, 72) = .24, p > .15. However, contrary
to what was predicted, more traditional participants who saw the androgynous portrayals
revealed signiﬁcantly greater traditionalism than those who saw the feminine portrayals
[F(1, 66) = 4.12, p < .05] and revealed marginally greater traditionalism than those who
viewed the masculine portrayals, F(1, 68) = 3.58, p = .063. There were no diﬁ‘erences
between more traditional participants who saw the media portrayals that were sex-typed

as feminine and masculine, F(1, 64) = .04, 32> .15.

35

Adjusted Mean GAS Score
By original attitude and type of media portrayal
With pretest score as a covariate

 

 

 

 

 

 

 

 

0.2
0.09

I
in,
E
inc-2) . or

(0.3) 1 1

Less traditional More traditional

Original gender role attitude

 

Femin. I Androg. Cl Masc.

 

 

 

Figure 3: Adjusted Mean GAS Score by original attitude and type of portrayal

36

In terms of participants’ attitudes toward the moral superiority of women, more
traditional and less traditional participants showed a tendency for differential responses
to the three types of media portrayals, E(2, 199) = 2.67, p = .072 (see Figure 4). In a
pattern somewhat similar to their average male gender role aptitudes, less traditional
participants revealed marginally different responses in regards to women’s moral
superiority based on the type of media portrayals, E(2, 107) = 2.87, p = .061. More
speciﬁcally, less traditional participants who viewed a masculine portrayal revealed
signiﬁcantly more traditional attitudes than those who viewed the feminine portrayals,
F(1, 72) = 5.67, p < .05. There were no differences between less traditional participants
who viewed the androgynous versus either the feminine [F(1, 72) = .85, p > .15] or the
masculine portrayals, F(1, 70) = 2.06, p > .15. Although more traditional participants’
responses regarding the moral superiority of women did not depend on the type of media
portrayals they viewed [E(2, 99) = 2.09, p > .10], the pattern of means suggests that their
responses were more traditional after they viewed the androgynous versus either the

feminine or masculine portrayals.

37

Moral Superiority of Women
By orig‘sral attitude and type of media portrayal
With pretest score as a covariate

 

0.3
0.2 r

0.1 -

 

 

l

(0.1)

Level of Nentradltlonalism

 

 

(02) r

-0.23 -O23
L

 

 

 

(0.3) 1
Less traditional More tradlionai

Original gender role attitude

 

Femin. I Androg. El Masc.

 

 

 

Figure 4: Moral Superiority of Women by original attitude and type of portrayal

38

Similar to what was found in regards to the model perception scores, more
traditional and less traditional participants’ attitudes toward the moral superiority of
women were differentially inﬂuenced by the age of the media models, F(1, 199) = 1.69, p
< .05 (see Figure 5). As found previously, less traditional participants were consistent in
their attitudes regarding women’s moral superiority regardless of the age of the media
models, F(1, 108) = .08, p > .15, whereas more traditional men displayed signiﬁcantly
more traditional attitudes toward the moral superiority of women when they saw the

younger models versus the older models, F(1, 100) = 5.89, p < .05.

39

Moral Superiority of Women
By original attitude and age (1 models
With pretest score as a covariate

 

 

 

 

 

0.4

E

.2

g

I

b .,_.-. _

s 722:;227/;;/////////z '9

2

‘5

E (0.1)

3 ,.
(02) are
(0.3) 1 1

Less traditional More traditional

Original gender role attitude

 

Young I Old

 

 

 

Figure 5: Moral Superiority of Women by original attitude and age of models

40

In terms of men’s perceived roles, more traditional men displayed signiﬁcantly
less traditional attitudes than did less traditional men after viewing the advertisements
[Ms = .14 and -. 13, respectively, F(1, 199) = 4.58, p < .05]. However, this was qualiﬁed
by a three-way interaction between original gender attitude, age, and type of media
portrayal, E(2, 199) = 4.45, p < .05 (see Figures 6 & 7). When only more traditional men
were examined (see Figure 6), there was a signiﬁcant interaction between the age of the
media models and the type of media portrayal viewed, E(2, 96) = 4.40, g < .05. Although
more traditional participants did not diﬂ‘erentiate between the type of media models when
the models were older [E(2, 46) = 1.55, p > .15], they responded differently to the types
of media portrayals when the models were younger, E(2, 50) = 3.89, p < .05. More
speciﬁcally, more traditional participants’ attitudes regarding men’s role were
signiﬁcantly less traditional when they viewed the masculine, younger models versus the
feminine, younger models [F(1, 33) = 5.08, n < .05] and androgynous, younger models,
F(1, 33) = 5.60, p < .05. No differences were found between the reactions of more
traditional participants to the feminine and androgynous younger models, F(1, 34) = .02,

p>.15.

41

Men's Role - More Traditional Men
By age of models and type of media portrayal
With pretest score as a covariate

.0
a:

 

P
.5

.°
N

 

 

O

 

Level of Nontraditionalism

A
p
N

v

-0.19

 

 

 

(0.4) 1 4
Young Old
Age of models

 

 

 

Femin. I Androg. [:1 Mass.

 

Figure 6: Men’s Role for more traditional men by age of models and type of

portrayal

42

Men's Role - Less Traditional Men

By age of models and type of media portrayal
With pretest score as a covariate

 

 

 

 

 

 

E
s
g (0.1) ///////%§§
: - ~
'3
3
(0.5) J J
Young Old
Age of models

 

 

Femin. I Androg. [3 Mesa.

 

 

Figure 7: Men’s Role for less traditional men by age of models and type of portrayal

43

For less traditional participants, their attitudes regarding men’s roles were not
inﬂuenced by both type of media portrayal and age of models [E(2, 104) = .94, p > .15]
but were only marginally inﬂuenced by the type of media portrayal they viewed, E(2,

104) = 2.86, p = .062 (see Figure 7). In a pattern consistent with their average male
gender role attitudes, less traditional participants who viewed masculine portrayals (M =
-.30) were signiﬁcantly more traditional in their attitudes regarding men’s role than when
they viewed the androgynous portrayals [M = .05; F(1, 70) = 5.11, p < .05] and were
marginally more traditional than when they viewed the feminine portrayals [M = .03; F(1,
72) = 3.77, p = .056]. Less traditional participants’ attitudes regarding men’s roles did
not diﬁ‘er when they viewed the feminine and the androgynous models, F(1, 72) = .01, p
> .15.

Participants’ attitudes about the expected work role of men were signiﬁcantly less
traditional for those who viewed the younger models versus the older models [Ms = .16
versus -.16, E(1, 199) = 6.80, p < .01]. This effect for age was qualiﬁed by a second-
order interaction between original gender role attitude and type of media portrayal, E(2,
199) = 3.30, p, < .05 (see Figures 8 & 9). More traditional participants displayed
signiﬁcantly less traditional attitudes when they saw the younger versus the older models,
with means of 0.24 versus -0.20, F(1, 96) = 5.72, g < .05, but this was qualiﬁed by an
interaction with the type of media portrayaL E(2, 96) = 3.92, p < .05 (see Figure 8).
Although more traditional participants did not respond differently to older models in the
three types of media portrayals [E(2, 46) = 1.13, p > .15], they did react differently to the
younger models in the portrayals, E(2, 50) = 3.28, p < .05. In a pattern similar to the one

found for attitudes regarding men’s roles, the masculine, younger portrayals caused the

44

more traditional participants to report less traditional work role attitudes than did the
androgynous, younger portrayals, EU, 33) = 5.58, p < .05. However, diﬁ‘erences were
not found when comparing those who viewed the feminine versus either the masculine or

androgynous portrayals, F(1, 33) = 1.62, p > .15 and F(1, 34) = 2.13, p > .15,

respectively.

45

Work Role of Men - More Traditional Men
By age of models and type of media portrayal
With pretest score as a corrariate

 

0.8 b
0.6
0.4 i
0.2

7/////////////% -”'

(02)
(0.4)
(0.6)

Level of Nontraditionailsm

 

 

 

 

Young Old
Age of models

 

 

 

Femin. I Androg. El Masc.

 

Figure 8: Work Role for more traditional men by age of models and type of

portrayal

46

WorkRoleofMen- LeuTradltlonal Men
By age of models and type of media portrayal
With pretest score as a covariate

 

.° P
as a:

.°
N

.-.._. ,
I: 1‘ "“3 $.-

_.
‘2
..

I
a

Level of Nontraditionailsm
O

A A
.0 .°
5 N
V V

 

 

 

 

(0.6) 1 ‘
Young Old

Age or models

 

 

 

Femin. I Androg. El Masc.

 

Figure 9: Work Role for less traditional men by age of models and type of portrayal

47

Again, less traditional participants’ attitudes were only inﬂuenced by the type of
media portrayal they viewed, E(2, 104) = 3.42, p < .05 (see Figure 9). In an identical
pattern to that found for men’s roles, less traditional participants who saw the masculine
media portrayals (M = -.33) were signiﬁcantly more traditional in their male work role
attitudes than those who saw the androgynous portrayals [M = .17; F(1, 70) = 6.93, p <
.01] and tended to be less traditional than those who saw the feminine portrayals [M =
.05; F(1, 72) = 3.56, p = .063]. There were no diﬁ’erences between those who saw the
feminine and androgynous portrayals, F(1, 72) = 0.39, p > .15.

No signiﬁcant main effects or interactions were found in terms of the traditional
stereotypes composite measure, the chivalry composite measure, or the child care
measure, although participants’ original male gender role attitudes continued to be
signiﬁcant covariates (gs < .001).

Discussion

Within the ﬁ’amework of the social judgment theory, attitude change is most
likely to occur when an individual likes a portrayed character. Contrary to what was
hypothesized, more traditional and less traditional participants did not have diﬁ’erent
reactions to the feminine, androgynous, and masculine media portrayals. Rather, both
more traditional and less traditional participants indicated that they liked the feminine
and androgynous media models more than they liked the masculine media models.
Participants’ perceptions of the similarity of the media models to themselves followed
this same general pattern. This pattern may have occurred because of the perceived
masculinity and femininity of the media models. Whereas the perceived masculinity of

the media models did not diﬂ’er across the portrayals, participants’ perceptions of the

48

femininity of the models increased ﬁom the masculine models to the androgynous
models to the feminine models. These ﬁndings suggest that contrary to what was
expected, the androgynous portrayals were actually perceived to be only somewhat
androgynous since they were perceived as more masculine than feminine, whereas the
feminine portrayals would be closer to the deﬁnition of androgyny in that they were
perceived to be both masculine and feminine. Due to this interpretation, it would be
more accurate to refer to the feminine media portrayals as “highly androgynous,” the
androgynous media portrayals as “mildly androgynous,” and the masculine media
portrayals as “traditionally masculine.” The fact that participants expressed greater
liking for the highly androgynous and the mildly androgynous models versus the
traditionally masculine models, is therefore consistent with research that has found that
people ﬁnd men and women who are described as relatively high on expressive qualities
usually associated with women as well as instrumental qualities associated mostly with
men as more attractive than individuals high in only one category (Spence et al., 1985).
Unfortunately, because the study failed to provide media models that the participants
perceived to be mostly feminine, the hypotheses regarding men’s reactions to
feminine/nonmasculine media portrayals was not tested.

Also contrary to what was predicted, the college age participants did not perceive
the younger models to be more likeable, similar, and attractive than the older models.
Rather, the participants’ perceptions of the younger and older models were qualiﬁed by
the gender role attitudes of the participants. Whereas less traditional participants did not
diﬂ’erentiate between the younger and older models, more traditional participants tended

to perceive the younger models in a more favorable light than the older models. It is

49

possible that this result is due to the fact that more traditional participants may be more
sensitive to issues regarding age than are less traditional participants and, more
speciﬁcally, that more traditional participants may be more likely to display a bias toward
youth, or ageism, than are less traditional participants. Research has shown, for example,
that voters generally display greater ageist versus sexist or racist biases against political
candidates, but unfortunately, this line of research has not examined the inﬂuence of
voters’ prior attitudes on their voting choices (e.g., Piliavin, 1987; Sigehnan & Sigehnan,
1982). However, research has found that traditionalism or conservatism is tied to
authoritarian personality characteristics and that, in turn, authoritarianism is tied with
prejudiced attitudes toward many social groups (Dillehay, 1978). Thus, it may be that
more traditional individuals also tend to be more ageist.

Altogether, participants’ perceptions of the media portrayals suggest that they like
the highly androgynous and mildly androgynous media portrayals better than the
traditionally masculine media portrayals. Given that people are generally inﬂuenced
most by media models they they favorably perceive, this suggests that the participants
will be inﬂuenced most by the highly androgynous and mildly androgynous portrayals
and least by the traditionally masculine portrayals. In addition, the fact that more
traditional participants tended to have more favorable perceptions of the younger versus
older models suggests that they will be more inﬂuenced by the younger models than the
older models. Due to their perceptions of the models, less traditional participants would
not be expected to respond differently to the younger versus the older models.

As predicted and as expected ﬁ'om the pattern of responses in regards to all

participants’ perceptions of the models, less traditional participants displayed less

50

traditional overall gender role attitudes when they viewed the highly androgynous
portrayals versus the traditionally masculine portrayals. This same pattern was also
found in terms of the perceived moral superiority of women and, to a lesser extent, in
regards to the work role of men and men’s role. It can be argued that this is due, at least
in part, to the inﬂuence of the traditionally masculine media portrayals. Across the four
subscales, the traditionally masculine portrayals consistently caused less traditional
participants to display greater traditionalism as compared to the grand mean for all
participants in the study. In fact, the magnitudes of these diﬂ‘erences were consistently
equal to or greater than the extent to which less traditional participants showed less
traditional attitudes to the highly androgynous portrayals as compared to the overall
sample.

The ﬁnding that less traditional men displayed greater traditionalism when they
viewed the traditionally masculine models versus the highly or mildly androgynous
models is contrary to what was predicted in this study. This is particularly true given that
the participants rated the traditionally masculine portrayals unfavorably as compared to
the two other portrayals. Participants’ ratings of the models suggest they should be less
inﬂuenced by the traditionally masculine portrayals versus the highly androgynous and
mildly androgynous portrayals. It was expected that less traditional men would be
immune to the traditionally masculine portrayals because they had developed resistance
to or defenses against traditional images through repeated exposures to traditionally
masculine images in the media. In fact, it appears that the traditionally masculine
portrayals may inﬂuence less traditional men because the portrayals provide models of

behaviors and interests that are discrepant from less traditional men’s attitudes and

51

because the behaviors and interests used to form the traditionally masculine portrayals
are considered desirable by both sexes, even if more typical of men. Given the fact that
men are generally portrayed in a stereotypic fashion in the mass media (e. g., Busby,
1975; Pearson et al., 1991), this ﬁnding does not bode well for less traditional men’s
ability to maintain their less traditional male gender role attitudes. Rather than being
immune to the inﬂuence of traditional images, less traditional men may be particularly
vulnerable, at least in the short term, to the inﬂuence of traditionally masculine images on
their male gender role attitudes.

It appears that both the highly androgynous and traditionally masculine media
portrayals were equally inﬂuential on less traditional men in regards to participants’
overall gender role attitudes and attitudes toward the moral superiority of women. As
compared to the overall sample, the extent to which less traditional participants expressed
less traditionalism after viewing the highly androgynous media portrayals was
comparable to the extent to which they expressed greater traditionalism after viewing the
traditionally masculine media portrayals. The reason why the highly androgynous
portrayals may have caused less traditional participants to express less traditionalism in
their overall gender role attitudes and the women are morally superior subscale, but not
in the subscales of the work role of men and men’s role, could be due to the extent to
which the measures are subject to social desirability biases. The subscales dealing with
men’s role and the work role of men address issues that are commonly discussed within
today’s society and which generally have answers that are considered more socially
desirable in that they express gender equality (e.g., “All occupations should be equally

accessible to both men and women” and “The initiative in dating should come ﬁ'om the

52

man”). The women are morally superior subscale and the total GAS score, on the other
hand, contain items that may generally elicit less social desirability bias (e.g., “Women
are morally superior to men” and “Men are more competitive than women”). These
items do not invoke thoughts about gender equality, so they may be less likely to promote
social desirability biases. In the presence of social desirability biases (i.e., the men’s role
and work role of men subscales), the extent to which participants can express even less
traditional attitudes is limited by the fact that their responses are already artiﬁcially less
traditional due to pressures on them to express gender equality. In the absence of social
desirability biases, the highly androgynous models are more likely to promote less
traditionalism because less traditional participants’ attitudes are not already artiﬁcially
less traditional.

Interestingly, less traditional participants’ responses to the mildly androgynous
portrayals generally follow the same pattern as for the highly androgynous portrayals. It
appears that less traditional participants do not greatly diﬂ‘erentiate between portrayals
that have at least some minimal level of femininity. This implies that there may be some
threshold at which less traditional men differentiate between traditionally masculine and
mildly androgynous media portrayals. In this particular study, the threshold appears to
be at a point somewhere between the mildly androgynous and traditionally masculine
media portrayals.

In a pattern opposite to what was originally predicted, the results suggest that
more traditional participants tend to make more traditional overall gender role attitude
responses when they view mildly androgynous media models versus highly androgynous

models and, to some extent, versus when they view traditionally masculine media

53

models. In addition, the means for the women are morally superior subscale fall in this
same pattern. These ﬁndings suggest that more traditional participants respond in a more
traditional manner when the gender role portrayals are perceived to have an intermediate
level of feminine behaviors and interests. It is possible that even though they may like
the particular media model, more traditional men react against media models that do not
ﬁt into their pre-established male gender role categories. Since more traditional
participants may not be able to classify the mildly androgynous models as prototypes of
either “manly” masculine men or “with it” androgynous men, they may feel uncertain or
uncomfortable about assimilating the attitudes expressed by the male gender roles of the
models into their own lives. This uncertainty may cause them react in a more traditional
manner since more traditional attitudes are probably more safe and comfortable to the
more traditional participants than are less traditional attitudes. In the same vein, more
traditional participants may feel more comfortable expressing less traditional attitudes
when they view media models that can be classiﬁed and understood to represent a
speciﬁc type of masculine role appropriate for men (i.e., masculine and androgynous).
Since traditional or conservative thought is associated with authoritarian personality
characteristics (Dillehay, 1978), this line of reasoning is consistent with research by Katz
and his colleagues (1960) that found that ego defensiveness is a major obstacle to attitude
change. In fact, Wagrnan (1955) found that counter attitudinal information had a
boomerang effect of making authoritarian people more prejudiced. Authoritarian
individuals tend to be uncomfortable and act defensively when confronted with
ambiguity since they prefer rigid, clearly deﬁned situations. This might explain why

more traditional participants displayed more traditional attitudes when the media models

54

displayed an intermediate level of femininity versus high or low levels of femininity.

As expected ﬁ'om the pattern of perceptions to the media models, but contrary to
what was predicted, the age of the media models did not inﬂuence men’s overall gender
role attitudes. In a pattern similar to that found for participants’ perceptions of the
models, more traditional participants responded diﬁ‘erently to the older and younger
models on several subscales while the less traditional participants did not respond
differently. More speciﬁcally, more traditional participants’ attitudes toward men’s roles
and the work role of men were diﬁ‘erent for the three types of media portrayals when
young men were the media models but not when older men were the media models. This
pattern suggests that more traditional participants are biased in their perceptions of the
media models in that they pay more attention to and are more sensitive to the portrayals
of the younger models versus the older models. This conclusion follows from the
generally more favorable perceptions that more traditional participants have of younger
media models. More traditional participants may feel like they have more to gain from
modeling the behavior of the younger models versus older models because they like the
younger models more than the older models.

In regard to the work role of men subscale and the men’s role subscale, more
traditional men displayed less traditional attitudes when they saw young, traditionally
masculine media portrayals versus young, mildly androgynous portrayals. These
ﬁndings suggest that more traditional men who see young, traditionally masculine media
portrayals may want to differentiate themselves ﬁ'om traditionally masculine portrayals in
order to seem more socially “with it.” This notion ties in with the idea that the men’s

role and the work role of men subscales are inﬂuenced by a social desirability bias that

55
encourages participants to display more gender equality. More traditional participants
may be especially sensitive to this desirability bias when they see young men displaying
traditionally masculine behaviors so they try to “distance” themselves ﬁ'om the portrayals
through more liberal assessments of the behaviors and traits that are appropriate for men.
Since, according to Dillehay, an authoritarian “endorses and adheres to conventional
norms” (p. 101), a less traditional response may simply reﬂect more traditional
participants desire to display gender role attitudes that are socially desirable in a
student/academic population.

More traditional participants again showed more sensitivity to the age of the
media models in terms of their perceptions of the moral superiority of women. For this
subscale, older models caused more traditional participants to display less traditionalism
than did those who saw the young models. This pattern again may be explained by the
idea that more traditional participants display more ageist tendencies than do less
traditional participants. More traditional participants may follow the edict to “respect
your elders” so that they perceive their male elders to be more moral than their peers. If
more traditional participants think that elderly men are more moral than are younger men,
it would explain why more traditional participants respond to the older media models by
displaying less traditional attitudes about the moral superiority of women (i.e., they think
older men are equally as moral as elderly women) while they are more traditional when
they view the younger media models (i.e., they think young men are less moral than
young women).

Although participants had more favorable perceptions of the young model in the

coﬂ‘ee advertisement versus the young model in the computer advertisement, the reason

56

why participants’ perceptions of the two scenarios differed is unclear. The differences
could be due to the photographs used in the advertisements, the products advertised, the
actual descriptions of the typical product user, or even due to the name assigned to each
of the media models.

Summary of Findings

The major assertion of the current research endeavor was that male gender role
portrayals in magazine advertisements can aﬁ’ect men’s gender role attitudes. This
proposition was indeed found to be true, but it is also very clear that there is not a simple
answer to how and when male gender role portrayals in magazine advertisements will
inﬂuence men’s gender role attitudes. The inﬂuence of male gender role portrayals
depends on (1) the speciﬁc male gender role attitude being assessed; (2) the original
gender role attitude of the viewer of the advertisement; (3) the age of the media model;
and (4) the extent to which the male media models display feminine characteristics.

In general, less traditional men appear to be particularly susceptible to the
inﬂuence of traditionally masculine media portrayals. Regardless of their speciﬁc
perceptions of the media models, less traditional men display more traditional gender role
attitudes when they see traditionally masculine media portrayals versus mildly or highly
androgynous media portrayals. More traditional men, on the other hand, tend to be more
discriminating in their response to male media portrayals. For instance, they appear to
react defensively, with more traditional attitudes, when they are exposed to media images
that have an intermediate level of perceived femininity versus high or low levels of
perceived femininity. It appears that mildly androgynous media portrayals may not ﬁt

into more traditional men’s preestablished deﬁnitions of appropriate male gender roles.

57

In addition, more traditional men appear be more sensitive to young male media models
versus old male media models. When the issue under examination is prone to social
desirability pressures, more traditional men appear to be concerned with expressing more
socially desirable attitudes when the media model is young but not when the media
model is old.
Implications

Overall, the ﬁndings of the present research are inconsistent with the premise that
observers tend to seek out and emulate models that are interesting and rewarding and
reject or ignore models who possess unattractive qualities. Even though the men in this
study were more favorable about the highly androgynous and mildly androgynous media
models than the traditionally masculine media models, there was not a consistent
tendency to emulate the behaviors and interests displayed by the better liked media
models. In fact, the traditionally masculine images appeared to reinforce masculine
cultural norms for more liberal participants while more traditional men appeared to reject
or respond defensively to images that did not ﬁt their deﬁnitions of what are appropriate
role behaviors and interests for men. In sum, the argument that people emulate media
portrayals that they ﬁnd attractive may be too simplistic within the domain of men’s
gender role attitudes. Even if a man likes a media model, this aﬂinity does not
necessarily indicate that he is willing to espouse the attitudes portrayed by the model.

One of the more important implications of this study is that male gender role
attitudes are complex and that it is diﬁicult to create less traditional attitudes in all men
using the same methods. Since research has found that the stereotype of the male gender

role is narrowly deﬁned (Hort et al., 1990) and that there are sanctions for violating the

58

socially prescribed male gender role (Carter & McCloskey, 1983-1984), the results of the
current study are not totally surprising. When men see traditionally masculine images in
the media, the process by which they are inﬂuenced by those images is more complex
than whether the mass media image is likeable. In line with Durkin’s argument that
individual diﬂ‘erence variables moderate the impact of mediated models on gender role
attitudes (1985b), the present research suggests that the inﬂuence of male media images
on male gender role attitudes is contingent not only on men’s prior gender role attitudes
but also on prescribed cultural norms regarding men’s roles and the individual
characteristics of the receiver. For instance, there is some suggestion that more
traditional men have rather inﬂexible male gender role attitudes that cause them to set up
strict boundaries or deﬁnitions of what are appropriate interests and behaviors for males.
Anything that conﬂicts or contradicts this deﬁnition appears to be rejected by more
traditional men in terms of their own personal attitudes. Thus, the present research casts
some doubts on the feasibility of creating dramatic changes in more traditional men’s
gender role attitudes through mass media channels because more traditional men appear
to reject images that do not ﬁt into their framework for appropriate male roles.

Surprisingly, it appears that traditionally masculine images promote less
traditional men to endorse more traditional male gender role attitudes. It appears that
even “liberal” men are not immune to the inﬂuence of images that reinforce the cultural
norm of strong, highly masculine men. Granted that the inﬂuence of traditionally
masculine images may only be short-term, this still makes one wonder at the malleability
of men’s gender role attitudes. In fact, this research in conjunction with research

showing that a single exposure to well liked Aﬁican-American exemplars such as Oprah

59

Winfrey and Michael Jordon can inﬂuence racial attitudes (Bodenhausen, Schwarz,
Bless, & Wanke, 1995), suggests that attitudes may be quite ﬂexible and highly context
dependent (cf. Wilson & Hodges, 1992). However, Bodenhausen and his colleagues
posit that this malleability may be limited to the area within an individual’s latitude of
acceptance. From this premise one can speculate that more traditional men’s gender role
attitudes may be less malleable than less traditional men’s because more traditional
men’s latitude of acceptance will be rather narrow due to their apparent need for more
rigid deﬁnitions and boundaries (e. g., Dillehay, 1978). Likewise, less traditional men’s
male gender role attitudes may be considerably more malleable because of larger
latitudes of acceptance, due to their greater ﬂexibility. In fact, it is interesting to
speculate whether more traditional men’s gender role attitudes may be more stable than
less traditional men just because they do tend to reject more images of masculine
behaviors and interests, since they do not fall within their rather narrow latitude of
acceptance.

This study, like many of the other studies dealing with the inﬂuence of similarity
on persuasion, has shown that the inﬂuence of similarity is not a simple process. It is
hypothesized that demographic similarity causes recipients to like the mediated model
more and due to this they will be more likely to emulate the behaviors and interests of the
mediated model. Overall, the present research indicates that this process is not so
straightforward. First, it is clear that personality characteristics can inﬂuence even the
extent to which you like a mass media model. Less traditional participants did not like
the younger, more demographically similar models more than they liked the older,

demographically dissimilar models. Second, even if people perceive a demographically

60

similar model in a favorable light, this does not necessarily indicate that they will
emulate that model’s behaviors and interests. More traditional participants liked the
younger models somewhat better and paid more attention to them, but they did not
directly imitate those models’ behaviors and interests. Overall, the present research
suggests that to fully understand the inﬂuence of demographic similarity it is important to
take into account the individual diﬁ’erence variables of the audience and to ascertain
when and why people emulate the behaviors of mediated models rather than simply pay
more attention to them.
Future Research
Hopefully, future research will continue to address the long-term inﬂuence of

male media images and the eﬂ‘ect of individual difference variables on this process.
More speciﬁcally, it is hoped that ﬁrture research will address some of the questions
raised in the present study. For instance, it would be interesting to explore whether more
traditional men do have a narrower latitude of acceptance for male gender role behaviors
and interests than do less traditional men. Also, does the difference in their latitudes of
acceptance cause more traditional men to have more stable gender role attitudes and
cause less traditional men to have rather unstable, context sensitive gender role attitudes?
In addition, it is hoped that future research will more fully explore the duration of the
impact a single exposure to male media portrayals has on male gender role attitudes. It
would also be of interest to more fully explore the tie between more traditional gender
role attitudes and greater ageist tendencies, rigidity, and adherence to social norms.

In order to more ﬁrlly understand the inﬂuence of mass media portrayals on male

gender role attitudes, research needs to be carried out across different types of

61

advertisement formats (e.g., lifestyle and celebrity type advertisements), different types
of mass media (e. g., television and radio), and also across a wide variety of program
styles (e.g., sitcoms and made for television movies). In addition, examination of the
impact of mass media portrayals in more realistic settings (i.e., with more incidental
exposure) will provide a more authentic view of the actual transmission of male gender
role attitudes via mass media sources.

Unfortunately, the present study was unable to test whether male viewers of
magazine advertisements will reject male media portrayals that possess highly feminine
but not masculine characteristics. Future research needs to address how viewers of
advertisements would respond to feminine/nonmasculine male portrayals. Lastly, due to
problems of regression to the mean of more traditional and less traditional participants’
gender role attitudes ﬁom pretest to posttest, the extent to which participants’ attitudes
became more or less traditional due to exposure to the magazine advertisements was not
available. Future research can address this shortcoming through use of a control
condition.

Conclusions

Despite this and other research, it still remains an relatively unanswered question
whether by virtue of constant repetition, mass media images can create long—term, stable
attitude change in its audience. This research only begins to address the question of
whether the stereotypical male gender role interests and behaviors that men see in the
vast quantity of mass media they consume have a stable impact on their male gender role
attitudes and by what process this occurs. The current research clearly shows that mass

media male images do aﬂ‘ect men’s gender role attitudes. What eﬁ’ect these images had,

62

however, is just as clearly dependent on individual diﬂ‘erence variables of the receivers.
The present research adds to the domain of knowledge regarding male gender role
attitudes by demonstrating that men’s original attitudes clearly moderate how they
respond to male media depictions that vary on their level of femininity and also by their
demographic similarity to the male receivers. In addition, this research also reinforces
that the principles of social learning theory and of the theory behind the eﬂ‘ects of
similarity on persuasion, are not straight-forward linear processes. If nothing else, the
present research reinforces the conclusion that how people perceive the world around

them, dramatically inﬂuences how they are inﬂuenced by the world.

APPENDICES

APPENDIX A

Appendix A

S'IM'HI' I

63

64

FEMININE PORTRAYAL; YOUNG MODEL: SCENARIO l

9%W'j gourmet coffee I I

N A ME :
70hn Hadon
HOME:
Chicago. :7/l/nois

AGE:
20

PROFESSION:
College student

 

 

 

6

IOIIIES.
fn/ogs going to cancers
OUTLOOK ON RELATIONSHIPS:

yohn values the companionship aspects of a steadg dating relationship
and the opportunig to act naralrant. He especial/g likes it when his ‘ _

date takes the initiative in the relationship since being in charge makes
him ancom/brtahle. 70hn takes special care with his appearance and

wans to be admired and wedatedhg his date.

PHILOSOPHY REGARDING MARRIAGE:
70hn [Eels that it is aecepWe/br a husband to sacriﬁce his own carreer
advancementin order to Wand enmwage his wires carreer ad-
vancement. He aso fees that iris the man: responsibility to share
maria/lg hi all [hm/lg responsibilities such as child care. cooking, and
cleaning. yohn/eels that a lather should spend signiﬁcant amoans of
quality time with his children. 9deallg. Yohn sees marriage as an emo-
tional anion of two people.

 

wmri

CEN‘I' ACHIEVEMENT:
Received a service award from the social club he belongs to.

HIS BRAND OF COFFEE:

9M9

 
 

 

65

FEMININE PORTRAYAL; YOUNG MODEL: SCENARIO 2

EM x;

NAME:
Steve Wilson

HOME:
Pheonilr. Arizona

AGE:
18

PROFESSION:
C’ollege student

 

 

Steve has a reputation forspending large amounts of time on his
appearance when he gets ready to go out in order to please his
dates. He enjoys the company o/intelligent women who take the
lead in Meir relationship and whom he can look up to Steve is er-
tremely considerate to those around him by being a good. sympa-
thetic listener who is atremely sensitive to others 'moods

 

OUTLOOK ON RELATIONSHIPS: ;
i
i

PHILOSOPHY REGARDING MARRIAGE:

9n an Ideal marriage. Steve thinlrs a man should be ﬂee to stay at
home and care [or the children while his spouse wonts outside of the

home. Steve feels that men should ﬁeauenﬂy and openly wuss

aﬂection toward their children with hugs and kisses. Steve also

believes that men should be especially sensitive to and responsive to
their wives c’hanges in moods and be willing to compromise in order
to ensure harmony in their relationships.

 
  

RECENT ACHIEVEMENT:
Won a trivia contest

HIS PERSONAL COMPUTER:

EPS

66

FEMININE PORTRAYAL; OLD MODEL: SCENARIO l
_ i
QM ‘J gourmet coffee (:4 /

N A M E :
Yohn Hadon

HOME:
Chicago. .‘7/lIno/s

AGE:
66

PROFESSION:
Retired

 

 

 

l
75

HOII :
Enjoys going to concerts

OUTLOOK ON RELATIONSHIPS:
7ohn values the companionship aspects o/a steady dating relationship
and the opportunity to act nurUIrant. He 6pecial/y likes it when his
date takes the initiative in the relationship since being in charge makes
him uncomfortable. yohn takes special care with his appearance and
wants to be admired and appreciated by his date.

."

PHILOSOPHY REGARDING MARRIAGE:
9ohn feels that itis accepmbAe/or a hisband to sacriﬁce his own carreer
advancement in order to sulport and encourage his wife: carreer ad-
vancement. He also ﬁels that it '5 the man s responsibility to share
equally in all family responSIbilib'es such as child care. cooking. and
cleaning. Yohn [eels that a father should spend signiﬁcant amounts of
quality time with his children. gdeally. 70hn sees marriage as an emo-
tional union of two people.

 

CENT ACHIEVEMENT:
Received a service award ﬁom the social club he belongs to.

HIS BRAND OF COFFEE:

2W9

 
 

67

FEMININE PORTRAYAL; OLD MODEL: SCENARIO 2

BPS-

‘b
KL:

NAME:
Steve Wilson

HOME:
Pheonix. Hrizona

AGE:
65

PROFESSION:
Retired

 

 

sums/mm.

OUTLOOK ON RELATIONSHIPS:

Steve has a rqoutation forspending large amounts of time on his
appearance when he gets ready to go out in order to please his
dates. He enjoys the company of intelligent women who take the
lead in their relationship and whom he can look up to. Steve is ex-
tremely considerate to those around him by being a good, sympa-
thetic listener who is extremely sensitive to others ’moods.

 

PHILOSOPHY REGARDING MARRIAGE:

. 9n an ideal marriage. Steve thinks a man shou/dbe free to stay at
"‘ home and care [or the children while his spouse works ouside of the
home. Steve feels that men should frequently and openly wress
aﬂection toward their children with hugs and kisses. Stew also
believes that men should be especially sensitive to and responsive to
their wives’changes in moods and be willing to compromise in order

to ensure harmony in their relationships.

 

 
  

RECENT ACHIEVEMENT:
Won a triwa contest

HIS PERSONAL COMPUTER:

EPS

68

ANDROGYNOUS PORTRAYAL; YOUNG MODEL: SCENARIO l

QMMMIJ gourmet coffee :/

I

l
l
I

    

NAME:
Yohn Hadon
HOME:
chicago, glli'nois
A GE:

20

PROFESSION:
College student

 

Enjoys going to cancers
OUTLOOK 0N RELATIONSHIPS:
9ohn values the companionship aspects of a steady dating relation-
ship and the opportunity to act nurturant. He always is the one to
take the ﬁrst step to start a relationship but Ilka it when his date
takes the initiative in the relationship since always being in charge
makes him uncom/ortable. 9n his leisure time. 70hn enjoys shooting
pool and restoring automobiles with his buddia.

PHILOSOPHY REGARDING MARRIAGE.

9ohn/ee5 oratit5sometr’mes necssavy/oreitherthehusbandor

m tosacriﬁcehis orhercarreeradvancementin ordertosupport
and encourage their mates cameer adVancement. He also [eels that

husbands should be able to take an occnional night out with their

friends. 7ohn thinks that it is the mans responsibility to share in
jam/7y responsibilities such as child care and washing the car. 70hn

feels that a father should spend quality time with his children.

 

i
“M’W5

RECE NT A CHIEVEMENT :
Received a service award from the social club he belongs to.

  

HIS BRAND OF COFFEE:

9%eoaé’r'o'o

69

ANDROGYNOUS PORTRAYAL; YOUNG MODEL: SCENARIO 2
EPS-
_.....“

NAME:
Steve Wilson
HOME:
Rheonik. air/Iona

AGE:
16'

   

‘ PROFESSION:
\ Collegestudent

 

Err/bashed»;-

OUTLOOK ON RELATIONSHIPS:

Steve much prefers to take the ﬁrst stqo to start relationships and
therefore. is usually the one who plans the evenings activities. He
enjoys the company of intelligent women who take the lead in their
relationship and to whom he can look up to. Steve is ectreme/y
considerate to those around him by being a good. sympathetic

listener who is extremely sensitive to others’

a
E
i

 

PHILOSOPHY REGARDING MARRIAGE:
Steve feels that a married man should be able to have a drink and
relax and unwind aftera tough day at work. He also assumes that
men should interact with their children by encouraging their interest
in spars activities and teaching them responsibility with money.
Steve also believes that men should be especially sensitive to and
responsive to their wives ’ changes in moods and be Willing to compro-
mise in order to enSure harmony in their re/an‘onships.

 
 
  

RECENT ACHIEVEMENT:
Won a trivia contest

HIS PERSONAL COMPUTER:

EPS

70

ANDROGYNOUS PORTRAYAL; OLD MODEL: SCENARIO 1

 

QM? gourmet coffee

NAME:
70hn Hadon

HOME:
Chicago. 9l/inoi's

AGE:
66‘

PROFESSION:
Retired

 

 

HOBBIES:
Enjoys going to concern:

OUTLOOK ON RELATIONSHIPS:
”n values the companionship aspects ofa steady dating relation-
ship and the moortunity to act nurturant. He always is the one to
take the first step to start a relationship but 07(6 it when his date
takes the inin'ative in the relationship since always being in charge
makes him uncomfortable. 9n his leisure time. 9ohn enjoys shooting
. pool and ‘ ' , ‘ " with his buddia.
PHILOSOPHY REGARDING MARRIAGE:
9ohn feels that it '5 mar-res necssaryfbr either the husband or
wife to sacriﬁce his orher carreer advancement in order to sumort
and encourage their mates carreer advancement He also feels that
husbands should be able to take an occasional night out with their
friends. 7ohn drinls that it is the man: responsibilig to share in
hmily responsibilities such as child care and washing the car. 7ohn
[Eels that a firther should spend quality time with his children.

RECENT ACHIEVEMENT:
Received a service award from the social club he belongs to.

   
 

HIS BRAND OF COFFEE:

ge/oooclr'o'o

"7/

6

WM "r

 

7l

ANDROGYNOUS PORTRAYAL; OLD MODEL: SCENARIO 2

NAME:
Steve Wilson

HOME: l
Pheonix. Arizona ‘

AGE:
65

PROFESSION:
Retired

 

 

_... .r
IOERI:
Ennis/brains-
OUTLOOK ON RELATIONSHIPS:
Steve much prefers to take the first step to start relationships and
therefore. is usually the one who plans the earnings activities. He
enjoys the company of intelligent women who take the lead in their
relationship and to whom he can look up to Stewis
considerate to those around him by being a good. sympathetic
listener who is extremely sensitive to others’moods

   

PHILOSOPHY REGARDING MARRIAGE:
Steve [be/s that a married man should be able to haw a drink and
" relax and unwind aftera tough day at work. Healsoassums that
men should interact with their children by encouraging their interest
in sports activities and teaching them responsibility with money.
Steve also believes that men should be especially sensitive to and
responsive to their wives’changes in moods and be willing to compro-
mise in order to ensure harmony in their relationships.

 
 
  

RECENT ACHIEVEMENT:
Won a trivia contest

HIS PERSONAL COMPUTER:

EPS

72

MASCULINE PORTRAYAL; YOUNG MODEL: SCENARIO 1

gm; gourmet coffee
/

5/

NAME:
Yohn Hadon
H O M E :
Chicago. 9/li'nois

AGE:
20

PROFESSION:
College student

 

 

 

W .
HOBBIES:
Enjoys going to concerts

00K 0N RELATIONSHIPS:
9ohn likes to take the initiative with women. He always 5 the one to
take the first step to start a relationship and decide what to do on a
date. Yohn also enjoys being seen in public with very attractive
women. 9n his leisure time. Yohn enjoys shooting pool and restoring
automobiles with his buddies. . ..

 

PHILOSOPHY REGARDING MARRIAGE:

Yohn feels that it is accqrtable for a wife to sacrifice her own carreer . 5 . ; ,
advancement in order to support and encourage her husbands ’
T . carreer advancement. 9ohn [be/s that husbands should have an I. ,6

‘ , occasional night out with their friends. He also thinks that it is the ‘ _ "
1 man s responsibility to share in family responsrbilities such as wash- 1' - 1‘

ing the car and taking out the garbage. 70hn feels that a father . '
should spend time with his children by playing catch or going to ball “ '
games.

 

RECENT ACHIEVEMENT:
Received a service award from the social club he belongs to.

   
 

HIS BRAND OF COFFEE:

9W9 .

73
MASCULINE PORTRAYAL; YOUNG MODEL: SCENARIO 2

EPS- 4;:

I!

NAME:
Steve Wilson

ME:
‘ Pheonix. Arizona
\ AGE:
16

PROFESSION:
. College student

 

 

6mm mm
OUTLOOK ON RELATIONSHIPS:

Steve loves to joke around with and play sports with his friends. He
has developed a reputation for making sure women have a good time
on their dates with him. Steve mud: prefers to take the ﬁrst stq to
start relationships and therefore. is usually the one who plans the
evening 5 activities.

PHILOSOPHY REGARDING MARRIAGE:
9n an ideal marriage. Steve thinks a woman shouldbeﬁee to stay at
home and care for the children while her spouse worhs ouside of the
home. Steve feels that a married man should be able to have a drink
and relax and unwind after a tough day at work. Stew also assumes

that men should interact with their children by encouraging their
interest in sports activities and teaching them responsibility with
money.

   

  

RECENT ACHIEVEMENT:
Won a triwa contest

HIS PERSONAL COMPUTER:

EPS

74

MASCULINE PORTRAYAL; OLD MODEL: SCENARIO 1

QWHIJ gourmet coffee 6/

NAME:
7ohn Hadon
HOME:
Chicago. Fill/hols

AGE:
66

PROFESSION:
Retired

 

 

 

i
E
@-

HOB]
Enjoys going to concerts

OUTLOOK ON RELATIONSHIPS:

Mn likes to take the initiative with women. He always is the one to
take the first step to start a relan‘onship and decide what to do on a
date. Yohn also enjoys being seen in public with wry attractiw
mmen. 9n his leisure time. 7ohn enjoys shooting pool and rstoring
automobiles with his buddies.

I

PHILOSOPHY REGARDING MARRIAGE:

Mn feels that it is accatable for a wife to sacriﬁce her own carreer
advancement in order to support and encourage her husband's
carreer advancement. 70hn feels that husbands should have an
occasional night out with their friends. He also thinks that it is the
man s responsibility to share in fumi‘ly responsibilities such as wash-
ing the car and taking out the garbage. 7ohn feels that a father
should spend time with his children by playing catch or going to ball
games.

 

"Wm?

RECENT ACHIEVEMENT:
Received a service award from the social club he belongs to.

    
 

HIS BRAND OF COFFEE:

Qoﬂoa‘ﬁ'o’o ,

75

MASCULINE PORTRAYAL; OLD MODEL: SCENARIO 2
EPS Technologies
[. 9—

NAME:
Steve Wilson

H0 ME :
Pheonix. Arizona

AGE:
65

PROFESSION:
Retired

 

 

 

Enjoys jogging.

OUTLOOK ON RELATIONSHIPS: .
Steve lows to joke around with and play sports with his friends. He I
has developed a reputation for making sure women have a good time
on their dates with him. Steve much prefers to take the first step to
start relationships and therefore. is usually the one who plans the
evening’s amirities.

PHILOSOPHY REGARDING MARRIAGE:

9n an ideal marriage. Stew thinks a woman should beﬁee to stay at
_ * home and care for the children while her spouse works ouside of the
’ ' home. Steve feels that a married man should be able to have a drink
and relax and unwind after a tough day at work. Stew also assumes
that men should interact with their children by encouraging their
interest in sports activities and teaching them responsibility with
money.

   

  

RECENT ACHIEVEMENT:
Won a trivia contest

HIS PERSONAL COMPUTER:

EPS

APPENDIX B

Appendix B

W

The composite stmcture that was used in the present research was based on the

screening sample and varied somewhat ﬁ’om the structure proposed by Ashmore et al.
(1992). The number provided for each item corresponds to the number used in the
screening questionnaire.

SCALE FOR THE GAS:

1 = Agree Strongly

2 = Agree Somewhat

3 = Agree Slightly

4 = Neither Agree nor Disagree
5 = Disagree Slightly

6 = Disagree Somewhat

7 = Disagree Strongly

TRADITIONAL STEREOTYPES (Cronbach’s a = 0.83):

l2
17
20
26
28
32
38
4O

42
46

Men are more competitive than women.

Men are more conﬁdent than women are.

Men are generally more adventurous than women are.
On average, men are more arrogant than women.
Men are more independent than women.

Women are more gentle than men.

Compared to men, women are more gullible.
Compared to men, women are more able to devote themselves completely to
others.

Men are generally more egotistical than women.
Compared to men, women tend to be weak.

WOMEN ARE SUPERIOR (Cronbach’s a = 0.74):

2
3
23
34
35
4]

Women are morally superior to men.

Compared to women, men tend to be cruel.

Compared to women, men have many undesirable qualities.

Women are a creative force, and men are a destructive force, in the world today.
Women are more helpful than men.

On average, women are better than men.

76

77

CHIVALRY (Cronbach’s a = 0.62):

4
21

27

44

Chivalrous gestures toward women on the part of men should be encouraged.
Men should continue to show courtesies to women such as holding Open the door
or helping them on with their coats.

Special attentions like standing up for a women who comes into a room or giving
her a seat on a crowded bus are outmoded and should be discontinued.

A man should protect a women from physical harm, even at his own expense.

WORK ROLE (Cronbach’s a = 0.71):

l4
19
29
39

All occupations should be equally accessible to both men and women.

It is appropriate to divide work into “men’s wor ” and “women’s work”.

In today’s world, the idea of “women’s work” and “men’s wo ” makes no sense.
A woman’s work and a man’s work should be fundamentally diﬁ’erent.

MEN’S ROLE (Cronbach’s a = 0.81):

l
13

The initiative in dating should come ﬁ'om the man.
As head of the household, the husband should have more responsibility for the
family’s ﬁnancial plans than his wife.

16 I would not respect a man if he decided to stay at home and take care of his
children while his wife worked.

18 The husband should have primary responsibility for support of the family.

25 I approve of a wife entering the labor force and leaving her husband at home to
care for the children.

31 A woman should allow the man to take charge of their sexual relationship.

37 It’s all right for the woman to have a career and the man to stay home with the
children.

45 In marriage, the husband should take the lead in decision making.

CHILD CARE:

43 Care of children should be shared equally by both spouses.

APPENDIX C

Appendix C

W

MODEL PERCEPTION RATINGS

(Cronbach’s as = 0.92 for scenario 1; 0.90 for scenario 2)

1)

2)

3)

4)

5)

6)

How well liked do you think the person in the advertisement is by others?
NOT VERY VERY

WELL LIKED 1 2 3 4 5 WELL LIKED

How much do you think you would like the person in the advertisement?

NOT AT ALL 1 2 3 4 5 VERY MUCH

How similar do you perceive the person in the advertisement to be to you?

NOT VERY SIMILAR 1 2 3 4 5 VERY SIMILAR

Do you think your behaviors and expectations are similar to those of the person in
the advertisement?

NOT VERY SIMILAR 1 2 3 4 5 VERY SIMILAR

How many close ﬁiends do you think the person in the advertisement now has?
NOT MANY MANY

CLOSE FRIENDS 1 2 3 4 5 CLOSE FRIENDS

If you knew the person in the advertisement, how likely is it that this person
would be a good friend of yours?

NOT VERY LIKELY l 2 3 4 5 VERY LIKELY

78

79

7) Do you think the person in the advertisement would be a good role model for
children?
NOT VERY GOOD VERY GOOD
ROLE MODEL 1 2 3 4 5 ROLE MODEL

8) How much do you yourself identify with the person in the advertisement?
DO NOT IDENTIFY 1 2 3 4 5 IDENTIFY VERY MUCH

9) How much do you think you would enjoy working with the person in the
advertisement?
NOT AT ENJOY
ALL ENJOY 1 2 3 4 5 VERY MUCH

DIRECT RATINGS OF MODEL

In the type of advertisement you are viewing, a biography of a “typical” product

user is used to sell the product. To this end, we would like you to rate the person
portrayed in the advertisement on a number of dimensions. Remember, this is a study of
how advertisements impact their audience so it is important that you rate the person
portrayed frankly and honestly. Your responses will be kept conﬁdential. Please answer
all questions even if you are not sure. We are interested in your best judgment based on
the information available.

Rate the person in the advertisement on the following adjectives:

1) Not at all likeable 1 2 3 4 5 Very likeable

2) Not at all similar to me 1 2 3 4 5 Very similar to me
3) Not at all masculine 1 2 3 4 5 Very masculine
4) Not at all feminine 1 2 3 4 5 Very feminine

5) Not at all attractive 1 2 3 4 5 Very attractive

6) Not at all powerﬁil 1 2 3 4 5 Very powerful

7) Not at all competent
8) Not at all pleasant
9) Not at all normal

10) Not at all trustworthy

80

2 3 4 5 Very competent
2 3 4 5 Very pleasant
2 3 4 5 Very normal

2 3 4 5 Very trustworthy

LIST OF REFERENCES

List of References

Ashmore, R. D., Del Boca, F. K., & Bilder, S. M. (1992). The construction and
validation of the Gender Attitude Inventory: A structured inventory to assess the
multiple dimensions of gender attitudes. Unpublished manuscript, Rutgers-The State
University of New Jersey, New Brunswick.

Atkin, C, & Greenberg, B. S. (1980). Eﬂ‘ects of mass communication. In C. L.

300k (Ed) WNW (PP 194-224) St
Martin’s Press

Bandura, A. (1986). Somaljnundamnsoﬂheughtmgmnmmw
mm. Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A, & Barab, P. G. (1973). Processes governing disinhibitory effects
through Symbolic modeling. WWW; 1-9-

Berscheid, E. (1985). Interpersonal attraction. In G. Lindzey & E. Aronson,

(Eds) Muhammad Vol 2 PP 413-484) New York:
Random House

Bodenhausen, G. V. ,Schwarz, N., Bless, H., & Wanke, M. (1995). Eﬁ‘ects of
atypical exemplars on racial beliefs: Enlightened racism or generalized appraisals?

W 48-63

Busby, L J. (1975). Sex-role research on the mass media. lgumalgf
WM“) 107-131

Carter, D. B. ,& McCloskey, L. A. (1983-1984). Peers and the maintenance of
sex-typed behavior: The development of children’s conceptions of cross-gender behavior

in their peers W 294-314.

Deaux, K. (1985). Sex and gender. W 49-81.

Dillehay, R C. (1978). Authoritarianism. In H. London & J. E. Exner, Jr.
(EdS) Dimminnaoﬁpersonalitywp 35- 127) New York John Wiley& Sons

Durkin, K. (1985a). Television and sex-role acquisition: 1. Content. British

81

82

WA, 101-113.

Durkin, K. (1985b). Television and sex-role acquisition: 11. Effects. British
WWWA. 191-210

Durkin, K. (1985c). Television and sex-role acquisition: III. Counter-
stereotyping NW 211-222

Eisenstock, B. (1984). Sex-role differences in children’s identiﬁcation with
counter-stereotypical televised portrayals. W 417430.

Fejes, F. (1989). Images of men in media research. WW
communicating 215-221

Flerx, V. C., Fidler, D. S., & Rogers, R. W. (1976). Sex role stereotypes:
Developmental aspects and early intervention. W 998-1007.

Frueh, T., & McGhee, P. E. (1975). Traditional sex role development and
amount of time spent watching television. WM], 109.

Geis, F. L., Brown, V., Jennings (Walstedt), J., & Corrado-Taylor, D. (1984a).
Sex vs. status in sex-associated stereotypes. SexRQleLLL 771-785.

Geis, F. L., Brown, V., Jennings (Walstedt), J. J., & Porter, N. (1984b). TV
commercials as achievement scripts for women. W 513-525.

Greenberg, B. S. (1990). Mass communication. In G. L. Dahnke, C. F emandez-

Collado & G W Clatterbuck (EdS) Hummmmunicationthmandmamh (PP
183-207). Belmont, CA. Wadsworth Publishing Company.

Harris, R. J. (1989). Amenitmwhelwfmmmmmimim Hillsdale,

NJ: Lawrence Erlbaum.

Hass, R. G. (1981). Effects of source characteristics on cognitive responses and

persuasion. InR. E. Petty, T. M. Ostrom,&T. C. Brock (Eds), Win
W (pp. 141- 172). Hillsdale, NJ: Lawrence Erlbaum.

Hicks, D. J. (1965). Imitation and retention of ﬁlm-mediated aggressive peer and
adult models WW2, 97-100

Hort, B. E., Fagot, B. I., & Leinbach, M. D. (1990). Are people’s notions of

maleness more stereotypically framed than their notions of femaleness? W
197-212.

Hovland, C. 1., Harvey, O. J., & Sherif, M. (1957). Assimilation and contrast

83

effects in reactions to communication and attitude change. WW
Smiallmhelmji. 244-252

Jackson, L. A. (1983). Gender, physical attractiveness, and sex role in
occupational treatment discrimination: The inﬂuence of trait and role assumptions.

WW1. 443-458

Jeﬁ‘ery, L., & Durkin, K. (1989). Children’s reactions to televised counter-
stereotyped male sex role behaviour as a ﬁmction of age, sex and perceived power

SodalBshardeuLA. 285-310

Jennings (Walstedt), J., Geis, 1., Florence, L, & Brown, V. (1980). Inﬂuence of
television commercials on women’s self-conﬁdence and independent judgment. Journal

W 203-210

Jensen, R. J. (1989). The effects of male role images in the media on sex role
attitudes (Doctoral dissertation, University of Florida, 1988). Wm
WM 5568-

Johnstonl &Ettema,J S (1982) MW
shildrenlatelerdsioir Beverly Hills 3388

Jones D A, & _MlllerA G- (1993 April) Iheintluencenﬁexpestatimand
presented at the annual meeting of the Midwest Psychological Association, Chicago.

 

Kanner, B. (1990, May 21). Big boys don’t cry: Returning the macho message.
W PP- 2041-

Katz, D. (1960). The ﬁrnctional approach to the study of attitude. Public
WM 163-204.

Kazdin, A. E. (1974). Covert modeling, model similarity, and reduction of
avoidance behavior. WM 325-340.

Kimball, M. M (1986). Television and sex-role attitudes. In T. M. Williams

 

301). Orlando: Academic Press.

Komhaber, R. C., & Schroeder, H. E. (1975) Importance of model similarity on

extinction of avoidance behavior rn children. W
W 601-607

Lehrer, J. (1989). The new man: That’s entertainment. W 8-
11.

84

Lovdal, L. T. (1989). Sex role messages in television commercials: An update.
W 715-724.

McGuire, W. J. (1985). Attitudes and attitude change. In G. Lindzey & E.

Aronson. (Eds.,) Ward ed Vol 2 PP 233-346) New
York: Random House.

Orlofsky, J. L. ,Ramsden, M. W., & Cohen, R S. (1982). Development of the
revised sex-role behavior scale l1mnm.1_Qf_Penman.li13LA.sserssrnem36, 632-638.

Pearson, J. C., Turner, L. H., & Todd-Mancillas, W. (1991). Gendeuml
mmmunicatignwp. 250-265). Dubuque, IA: Wm. C. Brown.

Piliavin, J. A. (1987). Age, race, and sex similarity to candidates and voting
preference WW 351 H368

Qian, Z., & Preston, S. H. (1993). Changes m American marriage, 1972 to 1987:
Availability and forces of attraction by age and education. W
M 482-495

Reeves, B. & Greenberg, B. S. (1977). Children’s perceptions of television
characters WW 113- 127

Remafedi, G. (1990). Study group report on the impact of television portrayals
of gender roles on youth. InumaloiAdeleseenthlthEmJL 59-61.

Robinson, J. B., & Insko, C. A. (1969). Attributed belief sirnilarity-dissimilarity
versus race as determinants of prejudice: A further test of Rokeach’s theory. lgumalgf

Emmenmmhinhmnaﬁmﬁ. 72-77

RogerS.E.M. (1973). W New
York: Free Press.

Shaw, M. E. &Costanzo, P. R. (1982). W. New
York: McGraw-Hill, Inc.

Sigelman, L., & Sigelman, C. K. (1982). Sexism, racism, and ageism in voting
behavior: An experimental analysis. WNW 263-269.

Signorielli, N. (1990). Children, television, and gender roles. W
AdolescemﬂealthEmJL 50-58.

Signorielli, N., & Lears, M. (1992). Children, television, and conceptions about
chores: Attitudes and behaviors. W 157-170.

85

Simons, H. W., Berkowitz, N N., & Meyer R. J. (1970). Similarity, credibility,
and attitude change: A review and a theory. W 1- 16.

Skelly, G. U. ,& Lundstrom, W. J. (1981). Male sex roles rn magazine
advertising, 1959-1979.16uma16f£6mmum§atm_31(4),52-57.

Spence, J. T. ,Deaux, K, & Helrnreich, R. L. (1985). Sex roles in contemporary

American society In G Lindzeyec E Aronson (Eds) Handbooknﬂsncialnmhnm
(3rd ed., Vol. 2, pp. 149-178). New York: Random House.

Tan, A. S. (1985). MasaCemmuninatinnJIrenrieunrlReseamh New York:
John Wiley & Sons.

Triandis, H. C. (1961). A note on Rokeach’s theory of prejudice. .ImrmaLQf
AbnonnalaniSnsialehelouﬁz. 184-186

Vaughan, J. E., & Fisher, V. L. (1981). The eﬂ’ect of traditional and cross-sex
modeling on children’s sex-role attitudes and behaviors. W
191, 253-260.

Vidmar, N. ,& Rokeach, M. (1974). Archie Bunker’ s bigotry: A study in
selective perception and exposure W24, 36-47.

Wagman, M. (1955). Attitude change and authoritarian personality. W
W 3-24-

Wilson, T. D, & Hodges, S. D. (1992). Attitudes as temporary constructions. In

L L Martin&A Tesser(EdS) WWW? 37-65)
Hillsdale, NJ: Lawrence Erlbaum.

nrcurcau STATE UNIV. LIaRnRrEs
lllllllllllmlllllllllllllllllllllllWWIWIWll
31293014091353