. LIBRARY
”Michigan Stat!
.1 2; University

  

This is to certify that the
thesis entitled
Sex Role Portrayals 0n
Commercial Broadcast Television

presented by

Laura Lee Henderson

has been accepted towards fulfillment
of the requirements for

MJL degree in Lnnmunicati on

 

 

Majbr prowa

Date February 24, 1978

 

0-7639

 

 

—-

I ll“! Illzllljljllll an ll! ll 1m Will u m

(D

 

 

 

SEX ROLE PORTRAYALS ON COMMERCIAL
BROADCAST TELEVISION

By

Laura Lee Henderson

A THESIS

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

MASTER OF ARTS

Department of Communication

1978

ABSTRACT

SEX ROLE PORTRAYALS ON COMMERCIAL
BROADCAST TELEVISION

By

Laura Lee Henderson

Sex role portrayals on prime time and Saturday morning
television were content analyzed. Three categories of beha—
vior were defined and recorded from a social learning theory
perspective. The three categories were: Dominance/Deference:
Gives Orders; Nurturance/Succorance: Needs Support; and
Independence/Dependence: Makes Plans.

Orders were defined as directives to do,say, or think
something. Supports were defined as non-routine physical
and emotional needs of assistance. Plans were defined as
statements of a method for doing something.

The data were analyzed with t-tests and a difference of
proportions test. Results showed sex differences within
all categories. The data were also analyzed by program
type, broadcast time, and viewing preference.

The results were discussed in terms of the stereotyping
of male and female sex roles as well as the availability of

sex role models for imitation.

ACKNOWLEDGEMENTS

There is no way to begin even properly recognizing all
the folks who helped out along the way. My committee, headed
by Dr. Greenberg and filled out by Katrina Simmons and Tom
Muth, most certainly deserves special thanks. Each of my
committee members served me a different function for which I
was both delighted and grateful.

Beyond my committee it gets tough because I want to put
everyone first for they all contributed to this thesis and
its completion in their own special way. My parents gave me
their pride. Joyce gave me her ear and made me feel it was
worthwhile. Steve, Ed, and Nicky bolstered my confidence
(and my alcohol consumption). Kris, Bruce, and Kruce Arms
restored my sanity. Nancy, Bill, Paul, Ron, Mary, Kathy and
the entire second floor kept tabs on my progress and cheered
me on. Suzy answered my questions and helped me through the
maze of graduate school memos. Willa, my typist, was incred-
ibly patient with my unwillingness to proofread with any
speed.

Although this work has only one name on it, it was not
accomplished by just one person. All these people, plus many
others -- graduates, undergraduates, faculty and friends -—
contributed to the final product. As a part of Project
Castle this thesis was supported by Grant 90-0-635 to

ii

iii
Michigan State University, Office of Child Development,
Department of Health, Education and Welfare, and the data
was collected through the efforts of Karen Cottledge,
Melanie Williams, Chris Kesserling, Jim Day, Ellen Hadley,
Helen Short, Ken Zgraggen. and Mike Zgraggen. Special

thanks goes to M. Mark Miller for getting me started in

graduate school and making me feel like I had a future.

TABLE OF CONTENTS

CHAPTER ONE - THEORETICAL AND RESEARCH PERSPECTIVE
Dominance/Deference

Nurturance/Succorance
Independence/Dependence
CHAPTER TWO - METHODS
Overview of Procedures
Dominance/Deference: Gives Orders
Nurturance/Succorance: Needs Support
Independence/Dependence: Makes Plans
Training Procedures
The Sample
Reliabilities
Analyses
CHAPTER THREE - RESULTS
Dominance/Deference: Gives Orders
Post Hoc Analyses: Gives Orders
Program Types
Broadcast Time
Viewing Preference
Nurturance/Succorance: Needs Support
Post Hoc Analyses: Needs Support
Program Type
Broadcast Time
Viewing Preference
Independence/Dependence: Makes Plans
Post Hoc Analyses: Makes Plans
Program Type

Broadcast Time
Viewing Preference

iv

12
16
21
25
25
27
29
33
36
37
38
42
47
48
52
52
60
63
68
68

74
so

85
87

96
9L:

Summary
CHAPTER FOUR - DISCUSSION
APPENDICES

Appendix A- General Instructions for the Use
of Coding Forms

Appendix B - Viewer Training Packets
Appendix C - 1975-76 Sample of Television Programs
Appendix D - t-values and Standard Deviations

REFERENCES

97
102

113
126
133
136
162

TABLE
TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

10:

11:

12:

13:

17:

: Means,

: Means,

: Means,

LIST OF TABLES

Intercoder Reliability Estimates

Means, t-tests, and z-scores: Orders Category,
All Shows

Means, t-tests, z-scores: Order Types by
Program Type

Means, t-tests, z-scores: Order Receivers

and Followed by Program Type

Means, t-tests, z-scores: Order Types by
Broadcast Time
Means, t-tests, z-scores: Order Receivers and

Followed by Broadcast Times

Order

Means, t-tests, z-scores: Types by

Viewing Preference

Means, t-tests, z—scores: Order Receivers
and Followed by Viewing Preference

Means, t—tests, z-scores: Needs Support
Category, All Shows
Means, t—tests, z-scores: Physical Support
Types by Program Type
Means, t-tests z-scores: Emotional Support
Types by Program Type

Means, t-tests, z-scores: Support Asked and
Responded to by Program Type

Means, t-tests, z-scores: Support Given and
Respondents by Program Type
t-tests, z-scores: Physical Support
Types by Broadcast Time
t-tests, z-scores: Emotional Support
Types by Broadcast Time

t—tests, z-scores: Support Asked and
Responded to by Broadcast Time

Means, t-tests, z-scores: Support Given and
Respondents by Broadcast Time

vi

#0

51

54

55

58

59

61

62

67

69

7O

72

73

75

76

78

79

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TMABLE

18:

19:

20:

21:

22:

23:

24:

25:

26:

27:

28:

29:

D1:

D3:

D5:

D6:

vii

Means, t-tests, z-scores: Support Types by
Viewing Preference
Means, t-tests, z-scores: Support Asked,

Responded to, Given and Respondents by Viewing
Preference

Means, t-tests, z—scores: Plans Category, All
Shows
Means, t—tests, z-scores: Makes Plans and

For Others by Program Type

Means, t-tests, z-scores: Plan Executors
and Outcomes by Program Type

Means, t-tests, z-scores: Make Plans and
For Others by Broadcast Time

Means, t-tests, z—scores:
Outcomes by Broadcast Time

Plan Executors and
Means, t-tests, z-scores: Makes Plans and For
Others by Viewing Preferences

Means, t-tests, z-scores: Plan Executors and
Outcomes by Viewing Preference

Direction of Difference:
Orders, Plans, Support:

t-tests and z-scores:
All Shows

Direction of Difference: t-tests and z-scores:
Orders, Support, Plans, All Shows: Outcomes

Direction of Difference: t-tests and z-scores:
Orders, Support, Plans, All Shows and Outcomes
t-values and Standard Deviations: Orders
Category, All Shows

t-values and standard deviations: Order Types
by Program Type

t-values and standard deviations: Orders
Receivers and Followed by Program Type

. t-values and standard deviations: Order
Types by Broadcast Time
t-values and standard deviations: Orders

Receivers and Followed by Broadcast Time

t-values and standard deviations: Order Types
by Viewing Preference

82

83
86
88
89
92
93
95
96
105
106
107
136
137
138
139

140

141

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

TABLE

D7:

D8:

D9:

D10:

D11:

D12:

D13:

D14:

D15:

D16:

D17:

D18:

D19:

D20:

D21:

D22:

D23:

viii

t-values and standard deviations: Orders
Receivers and Followed by Viewing Preference

t-values and standard deviations: Support
Category, All Shows

t-values and standard deviations: Physical
Support Types by Program Type

t-values and standard deviations: Emotional
Support Types by Program Types

t-values and standard deviations: Support
Asked and Responded to by Program Type

t-values and standard deviations: Support
Given and Respondents by Program Type

t-values and standard deviations: Physical
Support Types by Broadcast Time

t-values and standard deviations: Emotional
Support Types by Broadcast Time

t-values and standard deviations: Support
Asked and Responded to by Broadcast Time

t-values and standard deviations: Support
Given and Responded to by Broadcast Time

t-values and standard deviations: Support
Types by Viewing Preference

t-values and standard deviations: Support
Asked, Responded to, Given and Respondents
by Viewing Preference

t-values and standard deviations: Plans
Category, All Shows

t-values and standard deviations: Makes Plans
and For Others by Program Type

t-values and standard deviations: Plan
Executors and Outcomes by Program Types

t-values and standard deviations: Makes
Plans and For Others by Broadcast Time

t-values and standard deviations: Plan
Executors and Outcomes by Broadcast Time

142
143
145
146
147
148
149
150
151
152

153

154
155
156
157
158

1539

ix

TABLE D24: t-values and standard deviations: Makes
Plans and For Others by Viewing Preference 160

TABLE D25: t-values and standard deviations: Plan
Executors and Outcomes by Viewing Preference 161

CHAPTER ONE

THEORETICAL AND RESEARCH PERSPECTIVE

With the advent of the women's movement, content
analysis has been used in a variety of media to demonstrate
the existence of narrowly defined categories of appropriate
behavior for men and women. The purpose of content analysis
is to specify in a systematic manner the characteristics
of a variable of interest. It is generally done to find
out "what's there" in a systematic fashion that enables
a researcher to make general statements about what constitutes
the content of a particular medium.

Children's literature, print and television advertising,
news treatment, and television programming have been subject
to extensive content analysis of sex role portrayals (see
Busby, 1975, for a comprehensive review of sex role content
analyses). These efforts have produced the general finding
that mass media create and perpetuate sex role stereotypes.
Although the emphasis has been on the "negative" or "tradi-
tional" images of women, researchers have also found a
"traditional" stereotype for men as well (e.g., Reeves &
Greenberg, 1977, document the limited dimensions on which
children differentiate television characters, male and female).

The process of stereotyping as articulated by Bowes (1976)

zuqd by Carter (1962) involves first the homogenization of an

2
image. Homogenization results when the characteristics of
a situation become increasingly associated with, and
predictable from, each other. The homily "A woman's place
is in the home" demonstrates such homogenization.

Homogenization may be followed by the pglarization
of characteristics or attributes when they are held in the
extreme. Polarization of sex roles occurs when one sex
can only be characterized by the polar opposite of the other
sex, as in "Men are strong: women are weak." Finally, if
the image persists over time it has fixedness. The image
becomes stable and resistant to change.

Although researchers have demonstrated that stereo-
typing is not necessarily a negative process (e.g., Gage,
1952), studies of mass media stereotypes have viewed the
stable and polarized images of the sexes as being less than
desirable. This concern with stereotyped sex role models
stems mainly from the work done by Bandura and others (e.g.,
Mischel, 1966; Bandura, 1965) demonstrating the effective-
ness of observational learning.

For many years, differences in the behavior of boys
and girls were attributed to innate biological functions.
IWore recently, however, the focus has shifted from physio-
logical sources of difference to socializing sources of
(iifference. Behaviorists have attributed differences in
‘behavior to differential reinforcement of male and female
“behavior. Bandura (1971) has departed from traditional

:reirfibrcement theory by highlighting the role of

3
observational learning as a source of socialized behavior
differences.

Bandura's early work with the imitation of aggressive
film models (e.g., Bandura, Ross & Ross, 1963) has helped
spark the current concern over, and research into, television
violence. While stressing the importance of direct exper—
ience and reinforcement in the acquisition and performance
of behavior repertoires, Bandura has theorized that much
of behavior is learned by seeing how others are reinforced
for their actions (Bandura, 1971).

According to Bandura (1971) learning may be achieved
in one of two ways: through direct experience or through
observation. Extending a traditional focus on direct
experience and reinforcement, social learning theory incor-
porates the ability to acquire response repertoires through
modeling. As with traditional learning theory, reinforce-
ment plays a major role in observational learning.

Reinforcement may also be direct or observed. Direct
reinforcement occurs when a personis rewarded or punished
for performing a behavior. Observed or vicarious reinforce-
ment occurs when a person observes a model being rewarded
or punished for the performance of a behavior. In general,
vicarious reinforcement promotes faster learning than direct
Ireinforcement, but not long term responsiveness. Direct and
observed reinforcement interact to provide a person with
zan.observed standard for judging the relative value of

ciirect reinforcement the person receives (Bandura, 1971, p. 25).

L:

The concern over television role models has arisen
due to the effectiveness of vicarious reinforcement in
acquiring and regulating behaviors. It is the most obvious
medium available for observational learning. Television
provides attractive and vivid models from which to learn.
The positive or negative reinforcement of a television
character's behavior may affect a viewer's tendency to
imitate the model's actions. The focus here is on the out-
come of the behavior available for imitation. The behavior
itself is relatively neutral. It is the reinforcement,
or outcome, contingent on the behavior, that increases or
decreases the likelihood of imitation by the observer.

Behaviors which produce positive outcomes for models
are more likely to be imitated than behaviors which produce
negative outcomes. The work with filmed and televised
violence has shown that violent acts which receive positive
reinforcement or have positive outcomes are more likely to
be imitated (e.g., Bandura, Grusec, & Menlove, 1966; Bandura,
1965).

The stereotyping of sex roles on television is more
subtle than the stereotyping of the positive outcomes of
violence. This may be partially attributed to the nature
of interactions between, and characteristics of, the sexes.
Violence is inherently more "action-packed," and there-
:fore more obvious than the portrayal of men and women on
'television. The subtlety may also be due to the nature of
tflie sex-typing process.

Mischel (1966) has articulated a social learning

5
interpretation of the acquisition and performance of sex-
typed behaviors. He theorizes that boys and girls learn
behavior repertoires for both sexes. During the learning
process, however, they also learn to label which behaviors
are appropriate or inappropriate for their respective sex
due to "the difference in outcome as a function of the
performer's sex" (p. 60).

Mischel also stresses that response consequences need
not be obvious or direct. They may also be inferred. Just
as Mischel states "a man does not have to be arrested for
wearing a dress in public to learn the consequences for
such behavior" (p. 61), a little girl receiving more atten-
tion when wearing a party dress can learn that females should
be interested in their appearance without receiving a direct
compliment for being dressed up.

Television, by providing relatively homogeneous role
models in many areas, facilitates the labeling process
Mischel describes. This homogeneity of portrayal has been
demonstrated in terms of occupational portrayals and status
(Simmons, Greenberg, Atkin, & Heeter, 1977; Miller & Reeves,
1976), minority role portrayals (Simmons, gt al., 1977),
pro- and anti-social behaviors (Gerbner & Gross, 1976;
Feshback, 1972), demographic characteristics (Simmons, gt
§§L., 1977; Tedesco, 1974; Katzman, 1972), and sex role
gportrayals (Busby, 1974; Streicher, 1974; Downing, 1974).

Using the definition of stereotype provided earlier,
'television is a major medium for stereotyping. The homo—

ganeity has been demonstrated. Polarization is also evident

6

in television portrayals. Given time constraints which
force plot lines and characters into simplistic and easily
contrasted roles, polarization is necessary. The good guys
wear white hats; the bad, black. Men are strong; women,
weak (Tedesco, 1974). The continuing production of such
roles provides television with the fixedness sufficient

to perpetuate stereotypes.

Content analyses of sex roles have not been conducted
long enough, however, to document their fixedness. It can
only be assumed to have existed in the past. The fixedness
of violence has more support. And, with the continuing
assessment of violence on television (Greenberg, Atkin,
Edison, Korzenny, Heald, & Wakshlag, 1977; Gerbner & Gross,
1976), the medium itself is being stereotyped as antisocial.

In the context of social learning theory, the persis-
tence of such homogeneous role models is expected to have
an effect on viewers. The outcomes of male behavior are
different from the outcomes of female behavior. The effect
of this vicarious reinforcement of stereotyped role models
is to provide viewers with a homogenous (stereotyped)
behavior repertoire, especially in performance areas where
the observer has no previous direct experience with the
role model or the behavior.

This effect has been most clearly demonstrated in the
area of occupational roles. Miller & Reeves (1976) have
shown that for children, viewing traditional television

occupational roles (i.e., polarized by sex), decreases the

7
perceived appropriateness of cross-sex occupational role
portrayals (e.g., a female police officer). On the other
side of the coin, children who view counterstereotyped role
portrayals see the cross-sex occupational status as being
more appropriate. It should be noted that television pro-
vides many occupational role models with which children have
little or no direct experience (Simmons, et al., 1977).

Content analysis of television programming for sex
role stereotypes is a relatively recent phenomenon. The
majority of studies on television sex roles have been pub—
lished since 1974 (e.g., the Journal 9: Communication, Spring,
1974, published a symposium on women). It is interesting
to note that content analytic studies of other media tend
to precede those done on television (e.g., Chase, 1972;
Courtney & Lockeretz, 1971; Nilsen, 1971). However, since
researchers turned their attention to television, they
have left no area of programming untouched. Educational
television, news, prime time, day time, cartoons, and adver—
tising have all been subject to the scrutiny of vigilant
coders. Game shows appear to be the only program type left
to be analyzed.

Sex role variables tend to fall into five general
categories in television content analysis. Most studies
look into many of these categories at the same time to
:provide a general picture, or profile, of the program type(s)
'under study. These categories are: (1) head counts, (2)
demographic characteristics, (3) occupation, (4) physical

characteristics,(5) personality traits.

8

(1) Head Counts. It is almost prerequisite to tally
the number of male and female characters appearing on tele-
vision in one form or another. Busby (1974) found a male to
female ratio of 2.5 to 1 in cartoons. Miller & Reeves (1976)
report their results in percentages for prime time program-
ming: 72% male; 28% female (a ratio of 2.57 to 1). Other
studies (e.g., Turow, 1974; Sternglanz & Serbin, 1974)
report similar findings for prime time and cartoon programming.
Men and women appear with almost equal frequency in daytime
serials and as product representatives in television adver-
tising (Katzman, 1972; Courtney & Whipple, 1974).

Also included in this area of analysis is the number of
men and women appearing in major and minor, or regular and
supporting roles (Miller & Reeves, 1976; Busby, 1974; Courtney
& Whipple, 1974; Long & Simon, 1974). The percentages in
this area generally correspond to overall male/female
percentages.

(2) Demographic Characteristics. Common to many studies
is the recording of such variables as age, race, marital
status, and parental status. Long & Simon (1974) report
that more than half of the women on children's and family
(programs are married and more than two-thirds of the married
‘women have children. Tedesco (1974) reports similar findings
.in.prime time programs, with women also being reported as
younger than men .

Other studies reporting women being younger than men
irnzlude Dominick & Rauch (1972, advertising), and Downing

(197Wq daytime serials). The resultant profile of a

9

television female is a woman who is most likely youthful,

married, and a mother. Race is not a significant variable,
due mainly to the small percentage of non-white television
characters, particularly non-white females (Simmons, gt gt.,
1977; Downing, 1974).

(3) Occupation. Although occupation can be considered
a demographic variable, it merits a separate category due
to the large amount of attention it has received. Almost
every study of television sex roles includes some form of
occupational analysis. The basic finding is that few women
have jobs on television. Tedesco (1974, prime time) reports
40% female employment; 64% male employment. Long & Simon
(1974) report 18% female employment in children and family
programs. Downing (1974,day time serials) reports 80%
female employment; 89% male employment. Her figures, however,
include the category "housewife" which accounts for almost a
third of female employment in daytime serials.

When they are portrayed as being employed, the women
are most likely to be single and confined to a much narrower
range of job types than men. Men are more likely to have
professional or managerial positions, while women are most
likely to have in-home or clerical positions (Miller & Reeves,
1976; Long & Simon, 1974; Katzman, 1972).

(4) Physical Characteristics. The variables coded as
Iflrysical characteristics have included physical attractive-
rness, clothing, body traits (e.g., weight, height), physical
atrtivities, and size and color (in cartoons). Long & Simon

(197W) found that the majority of women they coded were

10
tall, thin, attractive, and well-dressed. Busby (1974) found
females to be smaller than males in cartoons. Tedesco (1974)
reports women to be more attractive and less physically active
than men.

(5) Personality Traits. Tedesco (1974) and Busby
(1974) have done the most extensive coding of personality
traits. Tedesco had coders rate each character on 15 semantic
differential scales. Busby used 40 semantic differential
scales. Other studies have coded characters on a smaller
range of traits (e.g., Miller & Reeves, 1976; Long & Simon,
1974; Hennessee & Nicholson, 1972).

Traits receiving the most research attention have been
independent/dependent, dominant/submissive, strong/weak,
and intelligent/unintelligent. Men are rated more positively
on these variables than women. Other polar opposites on which
men are rated positively and women negatively are: violent/
peaceful, rational/emotional, and active/passive. (The
first trait listed is considered the "positive" end of the
variable.)

By way of summary, these studies have delineated the
characteristics of male and female television role models.
While Bandura, Mischel, and others place emphasis on the
Inodel's characteristics in the facilitation of response
imitation, the outcomes (reinforcement contingencies) of
the model's behavior play an important, if not more important,
:role in social learning.

A relatively small number of studies have looked into

tflie amount and kind of reinforcement, direct or inferred,

11
that television sex role models receive for their actions.

Turow (1974) observed advising and ordering interactions
between the sexes on daytime and prime time television.
Turow used "episodes" of advising and ordering (treated as
an aggregate) to study "the relationships between knowledge,
activity, and sex of characters" (p. 138). He found that
television men and women conform to traditional stereotypes
in their advising and ordering behavior.

This stereotyping of knowledge points to the develop-
ment of a sex role standard, by which viewers may judge the
relative value of the direct reinforcement received for their
own advising and ordering. Women, by the television standard,
espouse typically "feminine" knowledge, e.g., in categories
such as love, the family, and the home. Men exhibit knowledge
in "masculine'areas, such as business, law, and government.
Turow has demonstrated Mischel's and Bandura's concepts of
inferred or indirect reinforcement of televised models by
showing the relationship between a character's sex and his/her
espousal of a certain kind of knowledge.

. Sternglanz & Serbin (1974) recorded a wide range of
behaviors found in television cartoons. They went a step
further in their coding by recording the consequences of the
behavior (consequences were scored as positive, neutral, or
negative). The outcome of male behavior was more often
positive than negative or neutral. The outcome of female
“behavior was more often negative or neutral. Sternglanz
.& Serbin interpret these neutral outcomes to mean "that in

,general their behavior had no environmental consequences"

(1% 713)-

12

The content analysis undertaken for this thesis attempted
to follow interactions between television characters (both
same sex and between sex) from the initial portrayal of
the behavior through to its outcome. Based on previous re-
search (especially, Sternglanz & Serbin, 1974), three cate-
gories of behavior were chosen as representative of behaviors
most often stereotyped by sex, culturally and in their tele—
vision portrayal. These categories are: Dominance/deference,
Nurturance/succorance, and Independence/dependence.
DominanceZDeference

Dominance is a broad concept which has been applied to
instances of bullying, influence attempts, toughness, direct-
ive behavior, machiavellianism, and leadership (see Maccoby
& Jacklin, 1974, for a comprehensive review). In general,
there are no sex differences in dominance, although male
children tend to be slightly more dominant in terms of
influence attempts than female children (e.g., Sutton-Smith
& Savasta, 1972).

With adults, it appears that only a general statement
can be made about dominance. Maccoby & Jacklin (1974)
point out that men generally have higher formal status as
well as generally higher status (p. 262). Due to this
generally higher status, men tend to take dominant roles
more often.

Although dominance is not a sex-typed trait in the
“real world," content analysis of television programming

sfluows that dominance exists as a sex-typed trait for male

13
and female characters.

Turow (1974) reports that 70% of the advising and
ordering episodes recorded during prime time television were
initiated by men. That is, men gave 70% and women gave 30%
of the advice and orders directed to a member of the opposite
sex. Turow found that daytime programming is not as blatantly
sex-typed on this variable, although men initiated 56% of the
episodes recorded.

Sternglanz & Serbin (1974) recorded both dominance and
deference in children's cartoons. Males were not shown to
be significantly more dominant than females, but females
were much more deferent than males. The definitions formu-
lated by Sternglanz & Serbin served as a starting point in
operationally defining dominance.

Dominance. To influence or control others, to per-

suade, prohibit, dictate. To lead and direct. To

restrain. To organize the behavior of a group.

Deference. To follow directions or example (imitate)

of a leader (except to ridicule) - either willingl

or under threat. To admire or compliment. (p. 712

These definitions were found to be too broad, including
forms of behavior that may not necessarily be dominant or de-
ferent. In doing the content analysis for this thesis,
verbal dominance, in the form of directive, or ordering,
behavior was recorded. Given the simplistic nature of role
portrayals on television, occupational status confers greater
authority on men than on women (Miller & Reeves, 1976;
fPurow, 1974: Katzman, 1972). With this in mind, two general
‘types of directives were identified according to the role

fromlwhich the directive was given: authority orders and

14

simple orders.

Authority orders are directives given by a person
in authority to a subordinate. This authority may be con-
ferred on the person by his/her occupational status, e.g.,
a doctor ordering a nurse; by the nature of his/her position
as a social agent, e.g., a police officer ordering a citizen;
or by parental status, as when a parent orders a child.

Simple orders are directed toward another peer. Peers

 

are defined as persons interacting with equal status along
any dimension, e.g., marital, as husband/wife; social, as
friends; occupational, as co—workers.

It was found after initial testing of the scheme that
threats can sound very much like orders. Therefore, a

threat order category was added to the analysis. Threat

 

orders are directives with statements concerning bodily

harm attached. Tedesco (1974) and Simmons, gt gt. (1977)
found that men are more often cast as criminals than women.
Therefore, men would be more likely to use threats as a form
of dominance.

Due to the deference demonstrated by women in previous
content analyses, it was felt that women would be more
likely to explain or justify their directives. Such ex-
planations serve to modify the directive,tone it down in
essence, to improve its chances of being followed or carried
out. Consistent with the non—authoritative female stereo-
type, women television characters would be expected to
explain or justify more orders than men.

General hypotheses concerning the giving of orders

15
were formulated based on these conceptions. Due to the un-
equal proportion of male to female television characters,
hypotheses, in all categories, are stated with the assump-
tion that "more" or "less" is used with respect to the rela-
tive frequency of male and female characters on television.
That is, chance occurrence is based on doing a behavior more
or less than the expected proportion of occurrence.

H1: Male characters will give proportionately more
authority orders than female characters.

H : Male and female will give proportionately equal
numbers of simple orders.

H3: Male characters will use proportionately more
threat orders than female characters.

H4: Female characters will explain proportionately
more of their orders, authority or Simple, than
male characters.

In keeping with social learning theory, the outcomes of
the directive behavior were recorded. In this analysis, the
outcome of an order is the degree to which the order was
followed. A character whose orders are followed is positive-
ly reinforced for being dominant. A character whose orders
are not followed is negatively reinforced for being dominant.
In line with the sterotyping of television sex roles, it is
expected that more male orders will be followed than female
orders, regardless of the type of order given:

H5: Orders given by male characters will be followed

proportionately more often than orders given by
female characters.
Of interest in these dominance interactions is also the

sex of the character being ordered. Turow (1974) restricted

16

his analysis to cross-sex orders. Sternglanz & Serbin (1974)
recorded dominance and deference as separate behaviors. In
this analysis the sex of the character receiving the order
was recorded regardless of the sex of the order giver.
Viewing directive behavior as an interaction between a
possibly dominant and possibly deferent character ties the
two forms of behavior together. With this perspective it
is possible to assess whether female characters are more
often cast in roles deferent only to male characters, or
whether they are cast in roles deferent to dominance in
general. It is hypothesized that females will be more often
cast as deferent to both sexes:
H6: Proportionately, male characters will order
other male characters more often than female
characters will order male characters.
7. Female characters will be the receivers of
orders proportionately more than males Will be
the receivers of orders, regardless of the sex
of the order giver.
Nurturance/Succorance
Murray (1938) used the terms nurturance and succorance
to describe giving aid to others and receiving aid from
others. Nurturance was considered a response to succorance,
and the succorant person was considered weaker, younger, and
dependent. Nurturance is generally considered to be care-
taking behavior. It has been almost exclusively labeled
a female role behavior, and is often used synonymously
\Nith "maternal" behavior.
Maccoby & Jacklin (1974) review psychological research

studying nurturant behavior by men and women. Included in

the review is a discussion of altruism, a concept similar to

17
nurturance. Maccoby & Jacklin describe the body of research
on altruism as dealing with "helpful, supportive behavior
that a person may direct toward a variety of other persons"
(p. 221). The emphasis in this research is most often on
voluntary offers of aid or voluntary assistance to another.
Sears, Rau, & Alpert (1965) defined nurturant behavior as
"voluntarily guiding or assisting another with the intent
of being helpful or performing a service" (in Maccoby &
Jacklin, 1974, p. 223).

The definition used in this content analysis combines
aspects of the "maternal" and the "altruistic" characteristics
of nurturant behavior. The person in distress need not
necessarily be weak or young, nor is the nurturant behavior
given only on a voluntary basis.

Deaux (1976) details a number of studies concerning
male and female personality traits. A consistent finding
emerging from her review has been that, culturally, women
are considered to be dependent and emotional, while men are
considered independent and active. Deaux also comments that
television is a force in creating and perpetuating stereo-
types. Maccoby & Jacklin (1974), however, could find no
sex differences in their review of studies on nurturant
behavior.

The stereotyping of nurturant/succorant behavior on
television has been recorded systematically in only one
content analysis. Sternglanz & Serbin (1974) recorded

rnxrturance and succorance as separate behaviors:

18

Nurturance. To nourish, aid or protect a defenseless

other. To express sympathy. To "mother" a child.

To give information to help carry out a project (see

Succorance).

Succorance. To seek aid, protection, sympathy, infor-

mation to help carry out a project. To cry for help.

To plead for mercy, to adhere to another person for

security. To be dependent. (p. 712)

They found male characters to be significantly more
succorant than females, but there was no sex difference for
nurturance. As with dominance/deference, these definitions
were found to be too broad due to the inclusion of information
seeking behavior, and were later defined more explicitly
in their operational definition.

Other studies touch on nurturance/succorance only peri-
pherally. Long & Simon (1974) report that a traditional
view of women is presented in television commercials in that
women are dependent and have socio-emotional family roles.
Busby (1974) reports that women are more emotional, fragile,
and dependent on others, while men are more adventuresome,
sturdy and self-reliant. Turow (1974) reports that women
in daytime and prime time programming are most knowledgeable
in "feminine" areas such as personal problems, the family,
and the home. Men display more knowledge in areas like
business, crime, and coping with danger.

The resultant research profile of nurturance/succorance
is consistent with the stereotypes held in this culture.
(Deaux, 1976; Donelson, 1975; Broverman, Vogel, Broverman,

Clarkson, & Rosenkrantz, 1972, all discuss a cultural stereo-

type of "warm" females and "competent" males). If television

19
stays true to the cultural stereotypes of active, adventurous
men and dependent, emotional women, there will be differences
in the ways men and women are portrayed in terms of succorance.
Male television characters will be more likely to find them-
selves in physical danger, while female characters will be
more subject to emotional distress.

The findingsof Sternglanz & Serbin (1974) appear to
conflict with this line of reasoning (i.e., males, in general
will be more succorant). They, however, made no distinction
between physical and emotional succorance. They also restricted
their analysis to seeking aid, which eliminates the previ—
ously defined succorant "condition" wherein aid might not
be requested. A succorant person, for the purposes of this
analysis, need only evidence a need for supportive behavior,
i.e., be in trouble, have a problem.

Nurturance is the response to that succorant need. It
may be given voluntarily or it may be asked for. A nurturant
response is a response with which another person attempts
to relieve the danger or distress. A nurturant respondent
need not be successful to evidence nurturance. Very simply,
a person in a succorant condition has a problem and a person
who responds to that problem is being nurturant. A person
znust be in a succorant condition in order for nurturance to
occur.

Although it would be expected that nurturance would be
21 female activity, the overabundance of male characters on

'television may negate any sex differences in the response

20‘

to succorant characters. Males may also be more likely to

respond to typically male problems, i.e., coping with physi—

cal danger. It is, however, hypothesized that female charac-
ters will do more than their share of nurturing others,
especially those who find themselves in emotional trouble.

H8: Male characters will be portrayed in physically

succorant conditions proportionately more than

female characters.

9. Female characters will be portrayed in emotionally
succorant conditions proportionately more than male
characters.

H : Female characters will respond to succorance with

10 .
nurturance proportionately more than male charac-
ters.

As with dominance/deference interaction, nurturance/
succorance is also considered in terms of its outcomes. In
this instance, the recorded behavior is the succorant condi—
tion and the outcome is the nurturant response. A nurturant
response giving aid to the succorant character is positive
reinforcement for needing aid or expressing a need for aid.
A response not giving support to succorance is negative re-
inforcement.

If female characters are nurtured only when in emotional
distress, and male characters only when in physical danger,

the characters are being differentially reinforced for be-

havior according to the sex of the performer.

H11. Male characters will be nurtured (receive positive
' reinforcement) for physical succorance more than
female characters.
H12: Female characters will be nurtured (receive posi-

tive reinforcement) for emotional succorance
proportionately more than male characters.

21
Independence/Dependence

The purpose in this study for recording instances of
independent behavior stems from previous content analyses
in the area. Male and female television characters have
been rated on their independence as a personality trait
(Busby, 1974; Tedesco, 1974; Long & Simon, 1974), but not
as a behavior.

Sternglanz & Serbin (1974) defined two categories of
behavior similar to the concept of independence: Achieve-
ment-construction and Autonomy.

Achievement—construction. Planning and carrying out

one's own plans, building, stating desire to overcome

obstacle, to surpass self or others.

Autonomy. To resist influence or coercion. To defy

an authority or seek freedom in a new place (emigrate).

To strive for independence. (p. 712)

They found male characters to be more constructive than
female characters. There was no sex difference for autonomy.
Their definitions are again too broad and also include an
element of dominance. The intent here was to define a di-
mension of behavior which showed independence that need
not be closely associated with dominance.

Defining independence poses a difficult task. The
problem with independence is that it is rarely defined.
Most often its polar opposite, dependence, is defined and
studied (e.g., Maccoby & Jacklin, 1974; Mischel, 1966).
.About the most that can be said about independence is that

it is not being dependent.

Although previous research (e.g., Busby, 1974; Tedesco,

:22
1974) has shown male characters to be more independent than
female characters, the characteristics of such behavior have
not been defined. In searching for a way to assess the
many shades of meaning independence can take on, a number
of concepts were considered that appear to be indicative
of independence. This culminated in the development of
a category labeled "makes plans."

Making plans was originally conceived as a decision
making category. Decision making is a part of being able
to plan actions. In order to plan one must decide on a course
of action to be implemented and also decide how to imple-
ment that action.

Making plans also incorporates the notion of leadership
in that not all plans may be carried out alone. A planner
may have to solicit the aid of others in order to implement
that plan. The concept of leadership is most often associated
with small group behavior. It is not necessary, however, to
lead a group to show independence of thought, autonomy, or
constructive behavior.

Creativity and originality of thought are also elements
of plan making. Making a plan and attempting to carry it
out show a measure of being able to think on one's own
without seeking the advice or opinions of others. Plan
Inaking also shows independence with the realization by the
lean maker that something needs to be done.

For this analysis, then, independence is defined as

'the ability to recognize the need for, and decide upon,

23
a plan of action when it is called for. "Makes plans"
attempts to tap this category of behavior by defining a
plan as a statement of a method for doing something.

The typical assumption has been that girls show more
dependency than boys, and research has shown that girls are
indeed a bit more dependent (Mischel, 1966). This assump-
tion infers the reverse as well, that boys will be more
independent or less dependent than girls. Given the stereo-
typic portrayals on television, it is hypothesized that male
characters will exhibit the independence of thought and
action defined as inherent in plan making more than females.

13 Male characters will make plans proportionately
more than female characters.

H

For this category, the behavior outcome is the success
or failure of the plan. Positive reinforcement for indepen-
dent behavior is seeing a plan through to its completion.
Negative reinforcement is having a plan fail. If female
characters are portrayed in traditional roles, they will
not be positively reinforced for independent behavior and,
therefore, more of their plans will fail.

H14: Male characters will formulate proportionately

more ultimately successful plans than female
characters.

In summary, the basic focus of this study is on the
behaviors and outcomes that may show sex role stereotyping
on television. The behaviors recorded in this analysis were
chosen on the basis of previous contest analysis studies,

as well as research into cultural attitudes towards sex

roles.

24
This study attempts to go beyond previous research into
television sex roles by incorporating social learning theory
into the content scheme. Social learning theory emphasizes
not only the behaviors available for imitation, but also the
outcomes of these behaviors. If differential reinforcement
of the sexes on television is present as hypothesized, viewers

are being presented with stereotyped sex role models.

CHAPTER TWO
METHODS
Overview of Procedures

The three major categories for the sex role content
analysis presented here were developed, tested, revised,
and implemented at the same time. Conceptually, however,
they were treated as, and still remain, distinct categories
requiring separate consideration. The development of each
category scheme will be presented as a separate procedural
task. This overview is presented here to provide a basic
outline for the development and execution of the sex role
content analysis scheme as a whole.

From the conceptual definitions of the major categories,
operational definitions were developed. For each conceptual
category, broad operational categories were defined and types
of behavior within each category were delineated: Dominance/
deference was operationalized as giving orders; Nurturance/
succorance as needing support; Independence/dependence as
making plans.

Following the broad category definitions, behavior
sequences for each category were outlined. These sequences
zattempted to logically structure the expected occurrence of
'behaviors conceptually following the occurrence of the main

cnategory behaviors. The behavior sequence for giving orders

25

26

was outlined: who gave what kind of order; to whom was it
given; and, was the order followed? The behavior sequence
for needing support was outlined: who needed what kind of
support; if someone responded, who was it; and was support
given? The behavior sequence for making plans was: who made
what kind of plan; was it for him/herself and/or others; who
carried out the plan; was it a success or failure? The con—
sequences of each behavior sequence were also considered a
part of the logical structure.

With a rough form drawn up for each category scheme,
another graduate student,M. Mark Miller (now a faculty mem-
ber of the School of Journalism, University of Wisconsin,
Madison), and I tested the scheme by viewing portions of
videotaped television programs. Changes and additions were
made in areas where we found difficulty in making decisions
regarding the recording of behaviors and in the use of the
forms.

The scheme was tested a second time for five to six weeks
by two undergraduate students. After consulting with them
about the scheme as well as their training, the coding forms
and variables were revised a final time in order to incorpor-
ate improvements suggested by their work with the scheme.
Revisions of the scheme will be noted in the following dis-
cussion of the variables coded and their operational defini-
tions.

A new group of coders was then trained in the final
*version of each category scheme. This group did the formal

Iwecording of sex role behavior sequences. Training procedures

2'2

will be outlined following the operational definition of
each category.
DOMINANCE/DEFERENCE: GIVES ORDERS

As previously stated, the Dominance/deference behavior
category became the orders category in the content analysis
scheme. Giving an order was defined as the beginning of a
Dominance/deference behavior sequence. The television char-
acter is the unit of analysis; therefore, all definitions
are given in terms of what the character does.

Characters were defined as persons, animals, or things
with speaking parts in a television program. The sex, male
or female, for each character was recorded.

In this category, giving an order is defined in this
fashion:

GIVES ORDERS: The character gives a directive for others

to do, say, or think something. Orders are coded accord-

ing to how the directive is or is not modified by the
ordering character. The order types are mutually exclu-
sive and exhaustive.

Originally, three types of orders were defined. During
the initial testing of the scheme we had difficulty distin-
guishing some kinds of threats from orders. Before pretesting
the scheme we added a category Threat Orders to the scheme.

Order Types:

Threat: An order given with a statement that physical

harm will result for the character being ordered if it

is not complied with. Example: Get in the car or you
won't know what hit you.

Authority: An order to be complied with because of

occupational position (e.g., boss), social agent (e.g.,

police officer, nurse), or parent. If a character

has been explicity made a delegate of any of the

above he/she is capable of giving an authority order.
Example: Get back to work.

28

Simple: An order given among equals or peers;
husband/wife, brother/sister, friends, co-workers,

etc. An order given by someone in an authority posi-
tion may be considered a simple order if the characters
are interacting as peers, e.g., in a special setting.
An order is simple unless clearly given as a threat

or an authority order. Example: Hurry up!

Explained: Any of the above orders may be further
modified by the inclusion of a justification for
why an order should be followed. This justification
must be made immediately prior to, or following the
giving of an order. Examples:

Threat Explained: I get nervous when I have to
hold a gun. So if you make a sudden move you might
get hurt.

Authority Explained: Be quiet, Jane. You're
disturbing the other students.

Simple Explained: Come back here. I want to
tell you why I said that.

Other variables in the behavior sequence:
Receiver: The receiver in the orders behavior
sequence is the character(s) to whom the order was
given.
Followed: An order was considered to have been
followed if the receiver carried out the order as it
was given. If the order was not carried out by the
receiver as given, the order was not followed.
Consequences: Consequences were defined as any action
or verbal statement that happened as a result of the
order being followed or not followed. They were coded
as present or absent, with space for comments as coders
found appropriate.
Coding Forms
An example of an Orders coding form and directions
for its use may be found in Appendix A. In general, letter
codes (e.g., "A" for authority orders) were used for record-
ing order types. "Check columns," columns which required a
check mark or an "X", were used for recording the sex of

'the character and whether the order was followed. All order

29,

types, followed and not followed orders, and order conse-
quences were briefly described by a quote, paraphrase, or
action following any letter code or check column. These
descriptions were provided as a check on coder accuracy.

A final column named "Notes" was included in the form during
its final revision. Coders could write in comments about
each behavior sequence if they encountered trouble in re-
cording that behavior sequence.

Behavior sequences were viewed and recorded in "time
segments." Initially, a time segment was two minutes long.
The first time segment began with the start of the program.
Time segments were numbered consecutively, used to organize
coding, and to record when the order was followed. Coders
found it too difficult to operate a time clock along with
their coding duties. In place of a two-minute time segment,
they used the counter on the videotape machine and viewed
programs in time segments of 50 as determined by the counter.
A videotape count of 50 on the machines used roughly approx-
imates two minutes.

NURTURANCE/SUCCORANCE: NEEDS SUPPORT

Needing Support, the second general category, was
defined as the beginning of a Nurturance/Succorance behavior
sequence. As with Giving Orders, the character is the unit of
analysis. It was defined as:

NEEDS SUPPORT: A character is in danger or distress.

This category does not include routine requests for

Assistance or social courtesies. It does include non-

routine requests or needs which are relevant to program

plots, subplots, or character development. The support

types are not mutually exclusive and may, therefore, oc-
cur in any combination during character interaction or

30

scene. They are exhaustive in terms of the range of
behaviors to be coded.

Originally, four types of need for support were defined:
Physical Internal, Physical External, Humiliation, and Inabi-
lity to Cope. During initial testing, Physical Confinement
was added to the physical needs for support. Physical Con-
finement was added because involuntary confinement fell into
neither of the two physical support types. Psychological
support and Concern for Others were added to the emotional
needs for support during the second revision of the scheme
when our coders told us they had difficulty fitting these
needs into the existing emotional needs for support.

The decision to add these support types was made on
the basis of viewing the segments of television programs
in which it was apparent that a character was in a succorant
condition for which we had no category.

The distinction between routine and non-routine requests
for assistance was made during the final revision of the
scheme. The coders had become overenthusiastic in their work
and were including statements such as "Can you tell me the
time?" as needs for support. The routine/non-routine distinc-
tion focused the coding on our original variables of interest.

Support Types:

Physical External: A character is in danger of being

killed, injured, or beaten. The threat of physical

harm comes from outside the character. Example: A

character is in danger of being caught in a cave-in,

landslide or other natural dangers.

Physical Internal: A character is suffering from a
disease, illness or internal malady. The threat of

31

physical harm comes from within the character.
Example: A person has cancer.

Physical Confinement: Characters are jailed, trapped,
or held against their will. One character's movements
have been restricted by another. The confinement is
involuntary. Example: Robbers lock people in a

bank vault.

Inability to Cope: A character states that he/she
has a problem that he/she cannot solve; is in need of
ego support. The source of emotional distress or
self-inadequacy for the character comes from within
the character. Example: A person can't get along
with boss, parents, spouse, etc.

Humiliation: A character states a fear that he/she
will be disliked or held in low esteem by others.
The character is afraid of being "put down."
Example: A person fears that someone will reveal
that they are homosexual, have a criminal record, or
other socially unacceptable characteristic.

Concern for Others: A character discusses help for
a friend, relative, or associate with a third person
(person needs support because someone else is in
trouble). Note that at least three people are in-
volved: The person expressing concern, the person
to whom concern is expressed, and the person in
trouble. Example: Person notes that someone is
late and expresses worry that he/she is lost.

Psycho Support: A character has a problem because

of the actions of others but does not express inabil-
ity to cope, fear of humiliation, or concern for
others. Example: Person's spouse has left them.

Another set of support categories was added during the

final revision at the request of the coders. The following

categories are occurrences that are not essential to plot

development; support is easily given; needs are momentary

and easily resolved within a scene or interaction. These

categories were developed to help coders make a conceptual

distinction between pertinent interactions and behaviors,

and less relevant interactions.

Compliments: A character receives praise, compliment,
or encouragement when not requested or particularly
needed. Example: You did that very well.

32

Approval: A character seeks confirmation of ideas,
opinions, or actions. Example: Do you like my new
dress?

Disappointment: A character is visibly upset, de-
pressed, or blue because efforts are unsuccessful.
The emotional setback is temporary, almost fleeting.
Example: What do you mean you can't meet me at the
airport?

The above variables do not appear in later analyses
and discussions due to their use as a coder aid and their
essentially "non-problem" (not involving distress or danger)
status.

The other variables in the behavior sequence are:

Asks Support: A character in need of support may ask
or not ask for aid in relieving the need. Therefore,
coder identification of a support need is not depend-
ent on the character asking for help.

Respondent: A respondent is defined as a character
who recognizes that another character is in need of
support. The responding character shows in some
way, through physical and/or verbal action, that he/
she knows that another character has a problem. A
respondent need not provide support in order to be
identified as a respondent.

Support Given/Not Given: Support is given when the
respondent attempts to provide aid to relieve that
particular need for support. Support is not given
when the respondent does not or cannot provide the
aid necessary to relieve the character's need.

Consequences: Any behavior that results from the
attempt to give aid or from the absence of aid is
defined as a consequence. They were coded as pre-
sent or absent, with space for comments as coders
found appropriate.
Coding Forms
An example of a Support coding form and directions
.for'its use may be found in Appendix A. As with the orders
cnategory, letter codes (e.g., "PI" for Physical Internal)

xvere used for recording support types. Check columns and

33
describe columns were used in a parallel fashion as well.
Coding of support behavior sequences was also organized by
time segment.
INDEPENDENCE/DEPENDENCE: MAKES PLANS
The final general category of behavior is the Plans
category. Formulation of a plan was defined as the begin-
ning of an Independence/dependence behavior sequence.
Again, the character is the unit of analysis.
MAKES PLANS: A description of a method for achieving
a goal. Making a plan is not merely a statement of
intent to do something (e.g., I'm going to make an
appointment). It is the statement of a method for
doing something (e.g., I'm going to make a doctor's
appointment to see if I can do something about my
sinus headache). The plan must be explicit in state-
ment or action. The categories are considered mutually

exclusive and exhaustive. The "other" category is
to be used only if the plan cannot be classified in

any other way.

Originally, a very small number of plan types were
defined for the analysis. After the final revision, however,
14 different plan types had been defined within the scheme.
This wide variety of plan types reflects the diverse nature
of the television programs coded. The plans behavior sequence
was the most difficult for the coders to record and the large
number of plan types both helped and hindered their efforts.
They helped by making it easier for a coder to recognize a
plan. They hindered by making the scheme more cumbersome
than the Orders and the Support schemes. Examples for
euach plan type are not given because the type of plan is
gin example in itself. A sample plans coding for and directions

.for'its use may be found in Appendix A.

34
Plan Types:

Makes Appointments: A character states that he/
she will set up a time to meet with someone for
a purpose.

Housework or Maintenance: A character states that
he/she will clean, repair, or fix-up, etc., for a
specified reason.

Social Affairs/Family: A character states he/she
will arrange an outing, party, etc., for family
members or relatives.

Social Affairs/Non-family: A character states
he/she will arrange an outing, party, etc., for
others outside of the family, or for the family
and out-siders (non-relatives), e.g., friends,
co-workers.

Business Deals: A character states that he/she
will prepare a contract, or arrange a meeting
in order to conduct business.

Strategies: A character states how he/she will
achieve a goal through movement and/or placement
of other people.

Criminal Activities: A character, identified as
a criminal, states how he/she intends to carry
out an illegal action.

Criminal Apprehension: A character states how
he/she intends to catch a criminal, or stop an
illegal action.

Construction: A character states how he/she
will build or put together something.

Acquisition: A character states how he/she intends
to get something legally that he/she doesn't
presently have.

Rescue: A character states how he/she will attempt'
to find, release, or rescue someone who is trapped,
lost, captured, etc.

Behavior Intent/Personal: A character states how
he/she will do something for him/herself, e.g.,
I'm going to take a walk so I can clear my head
and get some exercise.

The

Executor (carries out):

35

Behavior Intent/Interaction: A character states
how he/she will do something with another person,
e.g., I'm going right out there and straighten
him out. A future interaction is explicit in
the statement of method.

Other: If a statement of method cannot be class—
ified by any of the above categories, it is coded
as an "other" then described.

other variables in the behavior sequence are:

For Whom: A plan may be formulated for execution
by, and the benefit of, the character making the
plan, for other characters, or for both the plan—
making character and others.

Complexity: The complexity of the plan was rated
on a scale of 1 to 5 as defined below. Complexity
was added during the final revision in order to
free coders from repeatedly recording a very com-
plex plan.

1 = Very Simple, immediately executed; instruc-
tions implicit in the plan, e.g., I'll go
get the mop to clean up that mess.

2: Simple, executed within the same time seg-
ment; very simple set of instructions, e.g.
Officer Smith, you cover the back door;
Fred and I will break in through the front
and catch the burglar.

3 = Moderate, instruction necessary for execu-
tion which takes place within one or two
time segments. It advances the story line
but is not a major contribution to the plot,
e.g., plans for a surprise party which re-
quire moving and hiding people, coordinating
lights, etc.

4 = Complex, large set of instructions; carries
across more than two time segments. Some
of the plot will revolve around the plan and
execution, e.g., a bank robbery plan that
requires time, people, blue prints, etc.

5 = Very Complex, requires a major portion of
the show to plan and execute. Most of the
plot will revolve around the plan and its
execution, e.g., a Mission Impossible plot.

The executor is the character

who carries out the plan. The executor may be the
character who originally made the plan, or another
character acting as executor.

Execution:

Our coders felt the need to record whether

the execution of the plan was seen or not seen. They

36

found that many plans are formulated but that
their execution and outcome are only implied.

If the execution of the plan is portrayed during
the program, it is seen. If implied within the
plot or assumed by the characters, but not por-
trayed, the execution is not seen.

Consequences: The consequences of formulating and
executing a plan are its success or failure. Some
plans both succeeded and failed, sometimes more than
once. Each success and/or failure of a particular
plan were recorded.

TRAINING PROCEDURES

Coder training consisted of a series of discussion and
practice sessions. Before viewing any programs, the coders
discussed the variables with the investigators in order to
become familiar with them. The coding procedure may be sum-
marized as follows:
Discussion of conceptual definitions for each category.
Discussion of operational definition (category types
and other variables) for each category.
Instruction in the use of coding forms.
. Practice with the schemes.

Discussion of problems and questions concerning
the schemes.

\n-P‘Ki) NH

This sequence was repeated a number of times over a
two week period of training. Coders worked in pairs, initially
to discuss what they were recording. Following the train-
ing period they continued to work in pairs but were coding
independently for reliability purposes. Coder training
packets were developed during the training period to act
as aids in the recording of behavior sequences. These may
be found in Appendix B.

The criterion for the coders to meet was to identify
and record behavior sequences as defined. This meant that
the group of coders as well as their trainers had to agree

upon the identification and recording of sample behavior

37

sequences. Agreement between the two investigators was
reached during the development and initial testing of
the schemes. It was necessary then to bring the coders into
agreement with us, with each other, and with the definitions
given in their training packets. The reliability discussed
pin a later section reflect how well the criterion was met.

The formal coding was done by a group of four under-
graduate, female coders. They were all work study students.
It would have been most desirable to have a mixed-sex group
of coders. We attempted, but were not able, to hire an equal
number of male and female coders. The final group of coders
was all female. One male coder was involved in initial
training sessions but left the group before formal coding
began. The formal data collection was conducted during the
months of January to June, 1976.
THE SAMPLE

Data were collected on one videotaped episode each of
all prime time (8-11 p.m.) and Saturday morning (8 a.m. -
12 p.m.) regular network programming. Selected midseason
replacement shows (e.g., "One Day at a Time") were also
coded. Replacement shows were selected mainly on the basis
of having female lead characters. The original sample of
shows was aired and videotaped during a four-week period in
the fall of 1975. Midseason replacements were videotaped
in early 1976.

Not included in the sample of shows coded were variety

38
shows,documentaries or news programs, movies, and special
programs. The total number of shows coded, 79, represents
59.5 television hours.

Characters included in the final analysis were all
characters with speaking parts. A demographic analysis
(Simmons, Greenberg, Atkin, & Heeter, 1977) of this sample
of television programs identified 1212 speaking characters;
885 were males (73%) and 327 were females (27%).

It was possible for the coders to record nonspeaking
characters (e.g., a character receiving an order who doesn't
speak). In such cases only the sex was recorded and used in
analysis. Characters with nonspeaking parts who exhibited
a main category behavior (e.g., needing support) were not
included in the final analysis.

It was also possible for groups of people to be recorded
(e.g., a group of police officers receiving an order). If
all the characters in the group were of one sex it was re-
tained in the analysis and treated as one character. Mixed
groups were not retained in the analysis. Groups were not
capable of exhibiting a main category behavior by definition
and so were not recorded.

RELIABILITIES

Reliabilities were calculated on all behavior sequence
variables coded and retained in the analysis. About half
of the sample programs were double coded, i.e., two coders
viewed the same show independently. Intercoder reliabili-

ties were computed as follows:

39

1. All variables were indexed for each character
identified by a coder as having exhibited a main
category behavior.

2. The indices were created by summing across each _
variable type recorded by a coder for a particular
character. For instance, the number of authority
orders recorded by a coder for one character be-:
came the authority orders index for that character.
All dichotomous variables (e.g., orders followed/
not followed) were broken out and treated as two
variables (e.g., the number of orders followed and
the number of orders not followed).

3. Coders were randomly designated as Coder A or Coder
B. The variable index for Coder A was then correl-
ated with the variable index for Coder B. The cor-
relation coefficient for each variable was used as
the estimate of intercoder reliability. Coder A was
then used for the formal analysis.

Reliability estimates may be found in Table 1. Behaviors
recorded as consequences, with the exception of a plan's
success or failure, were dropped from the analysis before
reliabilities were calculated due to their extremely low
frequency of occurrence.

The reliability estimates show that the Orders category
was the category within which there was the most agreement
between coders on all variables. The Support category shows
the second highest level of intercoder reliability. The
Plans category is the least reliable of the three. This is
not too surprising. The Orders category is the simplest
category conceptually and operationally. It had the fewest
number of main category variables to be coded as well as
the fewest number of behaviors in its sequence.

The Plans category was the most complex conceptually
and operationally. Initial reliabilities showed less than
marginal reliability for the identification of a plan type.

Also, the total number of plans recorded was rather small

40

Table 1

Intercoder Reliability Estimates

Category: Orders N =

Variable

authority orders
simple orders
authority explained
simple explained
female receivers
male receivers
orders not followed
orders followed

all orders

Category: Support N =

Variable

physical internal
physical external
physical confinement
ego support

concern for others
psycho support
support not asked for
support asked for
support not responded to
support responded to
female respondents
male respondents
support not given
support given

all physical support
all emotional support
all support

Plans N =
Variable

Category:

plans

plans made for self
plans made for females
plans made for males
simple complexity
moderate complexity
high complexity
female executor
male executor

plan success

plan failure

277 characters

294 characters

I‘

.69
.67
-75
.50
.74
.48
.54
.63
.67
.68
.79
.63
.66
.66
.81
.60
.68

133 characters
r

.74
.14
.43
.65
.27
.26
.34
.31
.61
~59

41
in comparison to the two other main categories. Because of
this, plan types were collapsed into one variable, Plans.
The reliability estimate (.74) for this variable shows that
while coders could not agree on the type of plan a character
was formulating, they did agree that the character was
making a plan.

All variables coded in the Orders and Support categories
were retained in the analysis. Threat and Threat Explained
Orders had been dropped before reliability estimates were
computed due to their extremely low frequency.

Although the Support variables Ego Support and Psycho-
logical Support show, at best, marginal reliability (.50
and .48, respectively), they have been retained due to their
importance to theanalysis. Ego Support is a composite vari—
able. Humiliation and Self-inadequacy were collapsed into
one variable because of their low individual frequencies
and their conceptual similarity.

Four variables were dropped from the Plans analysis
due to extremely low reliability; Plans made for self and
the three complexity variables. Although complexity was
originally a five-point scale, initial frequencies showed
that coders were using basically a three-point scale. Very
Simple and Simple plans (1 and 2), were combined into Simple
Complexity. Complex and Very Complex (4 and 5) were combined
into the High Complexity variable. Moderate Complexity (3)

remained the same.

Two Plans variables were retained in the analysis in

4.2

spite of their low reliability estimates: Plans made for
other females (.43) and female executors (.31). These vari-
ables will be retained because the corresponding male vari-
ables reach an acceptable level of reliability (.65 and .61,
respectively), and because of their conceptual interest for
this study.
ANALYSES

As with most content analyses, the data collected in
this analysis are nominal. This level of measurement does
not allow for comparison of results across categories. The
creation of additive indices (frequencies) within categories
changes the structure of the data, however, from a nominal
level to an interval level. As long as inferential statis-
tics assuming an interval level of measure are computed only
for the indices within categories, the results are meaningful.

The hypotheses generated in Chapter One suggest that
there are two populations of interest in this content anal-
ysis. First, there is the group of characters who exhibit
at least one main category behavior. An example of an hypo-
thesis concerning this population is:

Male characters will need physical support more
than female characters.

In general, this type of hypothesis asks whether there
is any difference between the sexes in the number of behaviors
each sex exhibits. Assuming a normal distribution for this
population, the t-test for difference of means will answer
the questions these hypotheses pose. The results of these

t-tests will show whether there is a difference in the per

43
character rate of behavior exhibition. Or, more simply, a
significant t-test of these hypotheses will allow the infer-
ence that the rate (or average number per character) of a
behavior by one sex is very different from the rate of behav-
ior by the other sex. The means in each category of behavior
will show which sex has a higher rate.

The second population of interest is the entire popula-
tion of television characters with speaking parts. The
characters who exhibit main category behaviors are a subset
of this larger population.

This larger population is not normally distributed. The
number of characters found in each main category (691,
Orders; 613, Support; 212, Plans) is at best half of the
total sample under consideration. The distribution of all
characters, then, is skewed toward zero. An inferential
statistic assuming a normal distribution would be mislead-
ing in this instance.

The hypotheses generated about this population also ask
a different question than those of the subset of characters
who exhibit at least one main category behavior. These
hypotheses were formulated with the unequal representation
of the sexes on television in mind (males, 73%, females 27%).
An example of this type of hypothesis is:

In general,physical support needs will be proportion-

ately overrepresented as a male behavior and proportion-

ately underrepresented as a female behavior.

In general,this type of hypothesis asks whether the

sexes are performing behaviors in proportion to their rep-

resentation in the total population. The expected frequency

 

44

of behaviors performed by each sex is proportional to the
expected frequency of males and females in the population.

The statistic used to test these hypotheses is a z-
statistic. This z is a normal approximation of the binomial
distribution and similar to a Chi-square with 1 degree of
freedom (see Hays, 1963, p. 585, for a full discussion of
this statistic.) No population distribution assumptions
are made.

The formula for this 2 is:

- f

fol e1

V<f91><fe2>/N

01 observed frequency of female acts

 

 

where:

f

fe1 = expected frequency of female acts (27% of total)
fe2 = expected frequency of male acts (73% of total)
N = total number of acts

This z not only provides a test of difference of propor-
tions, it also gives an indication as to whether one group
is overrepresented or underrepresented in the population.

A negative z-score for this test will indicate that female
behaviors are proportionately underrepresented. A positive
z-score shows that female behaviors are proportionately
overrepresented. The reverse is true for male behaviors.
That is, a negative 2 means that male behaviors are over-
represented; a positive z that male behaviors are under—

represented.

45

In summary, then,two statistics have been computed
for these data: the t-test, which tests the difference in
mean rate of behavior performance by the sexes, and the
z—test, which tests the difference in proportion of behavior
performance by the sexes.

A number of exploratory post hoc tests were also per-
formed on the data. The main hypotheses concern television
programming in general and do not take into account the
different types of programming or time periods available.
(A list of television programs coded may be found in Appen-
dix C.) These post hoc tests were done in three breakdowns
of the data.

The data were broken down by: type of programming,
broadcast time, and viewing preferences. As defined in the
previous demographic analysis of this sample (Simmons, gt
gt., 1977) three program types were analyzed: situation
comedies, action adventure shows, and other show types,
which include family and medical shows. Saturday morning
shows could have been split into cartoons and noncartoons
for this category. This split was not made because the
number of shows, and therefore, characters, falling into
these categories became too small for meaningful analysis.

The broadcast time of the shows was split into three
time periods: Saturday morning, 8—9 p.m. (popularly known
as "the family hour"), and 9-11 p.m.

Viewing preferences were split into two categories:
preferred shows (Top 40 viewing) and non—preferred shows

(Low 60 viewing). This split was made on the basis of a

46
survey conducted in the Lansing area in the fall of 1975
with children in the fourth, sixth, and eighth grades.
Shows viewed weekly by 40% or more of the sample were desig—
nated preferred shows. The remaining shows were labled non-
preferred shows.

The results of this data analysis will be presented as
follows: Each main category (Orders, Support, Plans) will
be presented as a separate analysis, with the results of the
general (all shows) t- and z-tests presented first. Post hoc
test results will be presented within each main category.

A summary table presenting general data from all three cate-

gories will be discussed at the end of the results chapter.

CHAPTER THREE
RESULTS

The analysis of the data, as discussed in the previous
chapter, provides two kinds of evidence for each hypothesis
to be tested. The test of difference of means (t-test) com-
pares the average number of behaviors performed by male and
female characters in each content category. The difference
of proportions test (z-test) compares the extent to which
behaviors performed by each sex are representative of the
extent to which each sex is present in the total sample of
TV characters. Negative z-scores indicate that male behav-
iors are overrepresented, while positive z-scores indicate
that female behaviors are overrepresented in the sample of
television characters.

Consideration of both types of evidence should not only
provide support or non-support for the hypotheses generated
in the first chapter; it should also give an indication of
the extent to which sex role stereotyping is present in the
sample of TV content.

This chapter is organized around the three content
categories: Dominance/deference, Nurturance/succorance,
and Independence/dependence. Within each category results

are presented for the hypotheses discussed in Chapter One,

47

MB;
followed by results of the post hoc analyses for program type,
broadcast time, and viewing preference.

DOMINANCE/DEFERENCE: GIVES ORDERS

The results of the "all shows" analysis of the Orders
category appear in Table 2. Table 2a presents the results
of the Order Types analysis; Table 2b, the Receivers analysis;
Table 2c, the Orders Followed analysis.

Table 2a shows that, across all shows, male characters
gave nearly three times as many orders per male character
as did the female characters ("all orders"). Thus for
each male, the rate of order giving is much higher than
for each female.

The large and negative 2 indicates that, furthermore,
the total number of orders given by females was much less
than would have been expected from the distribution of male
and female characters in the population. The distribution
of order—giving across all shows is 20% female and 80%
male. That is, 20% of the orders given were given by
females.

Both tests, then, support the hypothesis that, in
general male television characters will give more orders
than female television characters.

This result holds regardless of the type of order a
character gives. It was hypothesized that male and female
characters would be equal in their order giving behavior for
Simple and explained orders. These hypotheses were not
supported by the data, as can be seen by the significant

t-tests and large and negative z-scores in Table 2a.

49

Negative z-scores indicate that male behaviors are over-
represented in the sample of television characters. All
differences are statistically significant beyond the .05
level.

Support is provided in Table 2b for those hypotheses
concerning receivers of orders. In the orders behavior
sequence, the male-male sequence occurs three times as
often as a female-male sequence. That is, in ordering
other characters, males order other males at a higher rate
than females order males. Males also order other males in
a higher proportion than would be expected from the popula-
tion distribution. The male-male sequence occunsin 82%
of the orders behavior sequences.

When females are on the receiving end of an order, the
rate issflightly lower (closer to 2:1), but the proportion
of females being ordered is nearly that of what would be
expected in the population. The insignificant z-score shows
that females receive orders from both sexes in proportion
to their presence in the population. Although the average
number of female receivers is quite different for males
and females, this could be a function of the larger number
of orders given by males.

The effectiveness of an order, whether it is followed
or not followed, is demonstrated in Table 2c. An effective
order (one that is followed) occurred three times more for
a male order giver than for a female order giver. That is,

the rate of followed orders was much higher for males than

50
for females. The large and negative 2 shows that the total
number of orders followed was not distributed as would be
expected in the population, with males having more than their
share of orders followed (82% of the orders followed were
given by males).

In the case of an ineffective order (one that was not
followed), the male rate is again much higher than the female
rate. The proportion of ineffective orders given by males
and females conforms, however, to the proportion expected
from the population distribution of males and females.
Therefore, support is found for the hypothesis that orders
given by male characters will be followed more often than
orders given by female characters, but only in terms of
the rate of occurrence.

In summary, the findings for the Orders category are
these:

-Males have a higher rate of order giving than
females, regardless of the type of order given;

-The total number of orders given by males is
much greater than would have been expected from the
distribution of males and females in the population;

—On the average, males are the receivers of orders
given by males more than receivers of orders given by
females;

-Females receive more orders on the average from
males than from females, but in proportion to the
expected distribution of males and females in the
population;

-Orders given by males are followed at a higher
rate than those given by females.

-Male orders are also not followed at a higher rate
than female orders, but the proportion of orders not

 

 

 

Table 2

Means, t-tests, and z-scores:

female N
male N

294
395

Table 2a:

Orders Category, All Shows

Order Types

 

 

 

 

Significance .~
Females* Males of t 2
Authority Orders 0.27 1.33 (.0001 -7.68C
Simple Orders 0.92 2.12 <.0001 -1.97a
Authority Explained Orders 0.06 0.28 <.0001 -3.21b
Simple Explained Orders 0.20 0.57 <.0001 —2.42b
ALL ORDER TYPES 1.45 4.29 <.0001 —7.190
Table 2b: Receivers
Significance of
Females Males t z
Female Receivers 0.41 0.94 (.0001 -1.42
Male Receivers 0.91 2.99 <.0001 -7.35C
Table 20: Orders Followed

Yes (followed)

No (not followed)

.0

ap 5
bp .01

c .0001
P

Significance of

 

Females Males t z
0.94 3.17 <.0001 -7.84C
0.53 1.21 (.0001 -1.43

 

*For all tables, the column females designates the mean. number of
behaviors originated by female TV characters.
designates the mean number of behaviors for male TV characters.

The column male

52

followed for both sexes is not significantly different

from the proportion expected from the distribution of

males and females in the population.
POST HOC ANALYSES: GIVES ORDERS

Tables 3—8 present the results of the post hoc analyses
for program types (Tables 3-4), broadcast time (Tables 5-6),
and viewing preference (Tables 7-8). These post hoc tests
are a way of looking for the sources of the large sex
differences found in the "all shows" analysis of the Orders
category.

Program Types

Table 3 shows the results of tests done on the types
of orders given in Situation Comedies, Action Adventure
programs and medical and family dramas (Medfam dramas).
Saturday Morning programming is not included in this break-
down.

Overall, the differences in the average number of orders
given by males and females in any program type are small.
The male rate for Action Adventure programs for all orders
(Table 3b) is almost twice that of the female rate. This
may be attributed to the large difference in the average
number of authority orders given by males and by females.

The other order type which shows a difference in the
rate of male and female order giving is in Table 3c. The
male rate of simple explained order giving in Medfam Dramas
is higher than the female rate. This difference is not
large enough, however, to affect the overall rate of order

giving in Medfam Dramas.

53

The proportions of orders given by males and females in
each of these program types show some interesting contrasts.
Table 3a shows that in Situation Comedies, order giving is
overrepresented as a female behavior (34%). While the aver—
age number of orders given by males and females in Situation
Comedies is almost the same, female orders are present in a
greater proportion than would be expected in the TV character
population. This is particularly true for female simple
orders. The large and positive z-score for this variable
shows that, in Situation Comedies, simple order giving is a
female attribute.

On the other hand, Table 3b provides the opposite results
for Action Adventure shows. Order giving is a male attribute
for all orders (88%), authority and authority explained or—
ders (93%) and simple orders (87%). Only in simple explain-
ed orders do the proportions of male and female orders equal
what would be expected from the population distribution of
males and females.

Results for the receivers of orders are similar to those
for order types. Table 4a shows no difference in the average
number of orders received by either sex in Situation Comedies.
But, females give orders to other females at a much higher
proportion than would be expected in the population (46%).

In Action Adventure programs (Table 4b), the results parallel
those found for all shows: males (88%) and females (89%) are
ordered by males at a higher rate and out of proportion to
what is expected in the population. There is no difference

in the receivers of orders in Medfam Dramas (Table 40).

Table 3:

Order Types by Program Type

Table 3a:

Means,

54

t’teSth

Situation Comedies

z—scores:

 

 

 

 

 

Females Males Significance
(N=32) (N=73) of t z score
Authority Orders 0.69 1.03 n.s. -0.92
Simple Orders 2.59 1.55 n.s. +4.82C
Authority Explained Orders 0.13 0.37 n.s. -1.64
Simple Explained Orders 0.69 0.62 n.s. +1.10
ALL ORDER TYPES 4.09 3.56 n.s. +2.84b
Table 3b: Action Adventures
Females Males Significance
(N=34) (Ns152) of t z-score
Authority Orders 0.47 1.59 (.01 -7.56C
Simple Orders 1.59 2.28 n.s. -6.08C
Authority EXplained Orders 0.21 0.32 n.s. —2.42b
Simple Explained Orders 0.50 0.43 n.s. -1.24
ALL ORDER TYPES 2.77 4.62 (.05 —9.66C
Table 30: Medfam Dramas

Females Males Significance

(N=28) LNF64) of t z-score
Authority Orders 1.39 1.72 n.s. -0.18
Simple Orders 2.04 1.88 n.s. +1.52
Authority Explained Orders 0.21 0.16 n.s. +1.16
Simple Explained Orders 0.18 0.50 <.05 -1.85a
ALL ORDER TYPES 3.82 4.25 n.s. +0.58

 

55

Table 4: Means, t-tests, z-scores:

Order Receivers and Followed by Program Type

Table 4a: Situation Comedies

Females Males Significance

 

(N=32) (N=Z3) of t z—score
Female Receivers 1.88 0.96 n.s. +4.94C
Male Receivers 1.97 2.23 n.s. +0.30
Yes (followed) 2.53 2.58 n.s. +1.20
No (not followed) 1.63 1.07 n.s. +3.36b

 

Table 4b: Action Adventures

Females Males Significance

 

(N=34) (N=152) of t z-score
Female Receivers 0.53 0.97 <.05 -4.72C
Male Receivers 2.00 3.44 v.05 -8.52C
Yes (followed) 1.91 3.43 (.05 -8.66°
No (not followed) 0.91 1.32 n.s. -4.61c

 

Table 4c: Medfam Dramas

 

Female Males Significance

(N=28) (N=64) of t z-score
Female Receivers 1.14 1.42 n.s. -0.20
Male Receivers 2.46 2.55 n.s. +0.89
Yes (followed) 2.38 3.11 n.s. -0.83

No (not followed) 1.54 1.18 n.s. +2.28a

 

56

The effectiveness of order giving also shows some inter-
esting results. While female order giving occurs at a high
proportion in Situation Comedies, so does the proportion
(40%) of female orders that are not followed (Table 4a).

This is also true for female orders not folowed (36%) in
Medfam Dramas (Table 4c). In Action Adventure programming,
however, male orders are followed at a higher rate and in a
higher proportion than would be expected (89%). Male orders
are also unsuccessful in a higher proportion (87%), but the
rates for unsuccessful orders are about the same for each
sex (Table 4b).

Broadcast Time

The results for broadcast time follow a similar pattern
(Tables 5-6). It should be noted that most Action Adventure
shows are aired during the 9—11 p.m. time period (Tables
5-6c); and many of the Situation Comedies appear between
8 and 9 p.m. (Table 5-6b).

Saturday Morning programming (Table 5a) shows no differ—
ence in the rate of all orders given by males and females,
but does show differences for authority and authority explain-
ed orders. The large and negative z-scores show, however,
that regardless of the type of order given, order giving is
overrepresented as a male behavior (83%).

Table 5b shows a proportional difference for all orders
given during 8-9 p.m. (77% male). This may be attributed to
the overrepresentation of authority order giving as a male

behavior. Otherwise, there is no difference in the average

57
number of the expected proportion of orders given by the
sexes. Table 50 shows average and proportional differences
in the male direction for order giving in the 9-11 p.m. time
period.

While Saturday Morning and the 8-9 p.m. periods show
no differences in the average number of order received by
males and females (Tables 6a & b), males proportionately
receive more orders from males in both time periods (83%,

80% respectively). The male-female orders behavior sequence
is proportionately overrepresented (88%) during Saturday
Morning programming, but not during the 8-9 p.m. period.

Table 6c shows somce curious results which may be due
to females outnumbering males in the 9-11 p.m. period. While
males order both sexes at a much higher rate than females,
only the male—male orders behavior sequence is overrepresent-
ed (82%) as indicated by the large and negative z-score.
There is no difference in the proportions of females receiv-
ing orders from males or females.

Table 60 shows that the 9-11 p.m. period is the only
time period during which the average number of orders follow-
ed and not followed is greater for males than for females.
While successful orders from males are proportionately over-
represented (83%) during the period, there is no proportional
difference for orders not followed.

Tables 6a & b show no differences in the rates of
effectiveness for orders given during Saturday Morning and

the 8-9 p.m. period. Male orders are followed in a greater

58

Table 5: Means, t-tests,

Order Types by Broadcast Time
Table 5a: Saturday Morning

z-scores:

 

 

 

 

 

Females Males Significance

(N=25) (N=106) of t z—score
Authority Orders 0.12 0.94 (.01 -5.53C
Simple Orders 3.00 2.43 n.s. -1.85a
Authority Explained Orders 0.04 0.23 «.05 -2.68b
Simple EXplained Orders 0.56 0.76 n.s. -2.57b
ALL ORDER TYPES 3.72 4.35 n.s. -5.45C

Table 5b: 8-9 p.m.

Females Males Significance

(N=46) (N=152) of t z—score
Authority orders 0.83 1.24 n.s. -3.69b
Simple Orders 2.39 1.92 n.s. -0.11
Authority Explained Orders 0.22 0.27 n.s. -1.25
Simple Explained Orders 0.50 0.51 n.s. -1.11
ALL ORDER TYPES 3.94 3.94 n.s -2.79b

Table 5c: 9 - 11 p.m.

Females Males Significance

(N=223) (N=132) of t z-score
Authority Orders 0.18 1.76 <.0001 -4.66°
Simple Orders 0.38 2.11 <.0001 -1.66a
Authority Explained Orders 0.03 0.32 (.0001 -1.94a
Simple Explained Orders 0.09 0.48 (.0001 -0.49
ALL ORDER TYPES 0.68 4.66 <.0001 -4.36C

 

59

Table 6: Means, t-test, z-scores:
Order Receivers and Followed by Broadcast Times

Table 6a: Saturday Morning

 

Females Males Significance

(N=25) (N=106) of t z—score
Female Receivers 0.36 0.61 n.s. -2.88b
Male Receivers 2.64 3.11 n.s. -4.64C
Yes (followed) 2.56 3.23 n.s. -5.13C
No (not followed) 1.16 1.18 n.s. -2.35b

 

Table 6b: 8 — 9 p.m.

 

Females Males Significance

(N=46) (N?157) of t z-score
Female Receivers 1.15 0.81 n.s. +0.67
Male Receivers 2.39 2.80 n.s. -3.65b
Yes (followed) 2.54 2.90 n.s. —3.49b
No (not followed) 1.44 1.13 n.s. 0.00

 

Table 6c: 9 - 11 p.m.

 

Females Males Significance

(N=223) (Ns132) of t z-score
Female Receivers 0.26 1.37 <.0001 -1.02
Male Receivers 0.40 3.11 <.0001 -4.53C
Yes (followed) 0.43 3.43 <.0001 -5.10C

No (not followed) 0.27 1.33 <.0001 —0.59

 

60
proportion, however, in both time periods (84%, 80%, res-
pectively). There is no difference in the proportion of
orders not followed during the 8-9 p.m. period, but a greater
proportion of male orders are not followed during Saturday
Morning programming (81%).
Viewing_Preference

The results of the breakdown of shows by viewing prefer-
ence can be found in Tables 7 - 8. It sould be noted that
the viewing preferences of our sample (those shows for which
40% or more of the respondents indicated they viewed regular-
ly), tend to conform to broadcast time patterns. That is,
many of the preferred shows were aired Saturday morning or
before 9:00 p.m.

Table 7a shows no difference in the rate of order giving
by the sexes, regardless of the order type for preferred
shows (Top 40). Table 7b shows a large difference in the
rate of order giving for non-preferred shows (Low 60).

There are proportional differences, though. Order
giving is overrepresented as a male behavior in both tables
(78%, 81%, respectively). Only simple orders in preferred
shows and simple explained orders in non-preferred shows are
present in proportions that would be expected from the popu-
lation distribution.

Table 8a shows no difference in the rate of orders
received by either sex in the preferred shows, but the male-
male orders behavior sequence is overrepresented (81%) as

indicated by the large and negative z—score. Table 8b

61

Table 7: Means, t-test, z-scores:

Order Types by Viewing Preference

 

 

 

Table 7a: Preferred Shows

Females Males Significance

(N=54) (N=185) of t z-score
Authority Orders 0.63 1.18 n.s. -4.82C
Simple Orders 2.56 1.90 n.s. +0.72
Authority Explained Orders 0.17 0.30 n.s. -2.49b
Simple Explained Orders 0.48 0.61 n.s. -2.29a
ALL ORDER TYPES 3.83 4.00 n.s. -3.52b

Table 7b: Non-Preferred Shows

Females Males Significance

(N=240) (N=210) of t z-score
Authority Orders 0.19 1.47 <.0001 -5.98C
Simple Orders 0.55 2.31 <.0001 -3.18b
Authority Explained Orders 0.04 0.25 <.0001 -2.28a
Simple Explained Orders 0.13 0.52 (.0001 -1.14
ALL ORDER TYPES 0.91 4.55 <.0001 -6.51C

 

62

Table 8: Means, t-tests, z-scores:
Order Receivers and Followed by Viewing Preference

Table 8a: Preferred Shows

 

Females Males Significance _
(N=54) (N=185) oft z-score
Female Receivers 1.26 0.84 n.s. +1.21
Male Receivers 2.13 2.68 n.s. -3.45b
Yes (followed) 2.52 2.94 n.s. -4.07C
No (not followed) 1.43 1.15 n.s. -0.13

 

Table 8b: Non—Preferred Shows

 

Females Males Significance _;

jN—240) (N-210) 0 ‘t z-score
Female Receivers 0.21 1.03 (.0001 -2.89b
Male Receivers 0.63 2.26 <.0001 -5.77C
Yes (followed) 0.58 3.37 (.0001 -6.88C

No (not followed) 0.33 1.26 <.0001 ‘ -1.82a

 

63

shows differences in both rates and proportions for orders
followed and not followed in non—preferred shows. Male
orders are bosh followed and not followed at a higher rate
than female orders and are overrepresented as a male behav-
ior (84%, 77% respectively).

In summary, post hoc findings for the Orders category
shows:

-Only during Situation Comedies is order giving
overrepresented as a female behavior, but that behav-
ior is not particularly effective in that female orders
are not followed out of proportion to what would be
expected in the population.

—Action Adventure programming and the 9-11 p.m.
time period show the highest rates of male order giving
behavior.

-In all time periods, male orders are proportion—
ately overrepresented as being effective (i.e., fol-
lowed).

-Saturday Morning programs show the only difference
in proportion of orders not followed, with ineffective
orders being overrepresented as a male behavior.

-Preferred Programs show no difference in the rate
of order giving by the sexes, but overrepresent order
giving as a male behavior.

-Non-preferred programs show a rate and a propor-
tional difference in order giving in the male direction.

-Preferred and non-preferred programs both overrep-
resent male orders as being more effective than what
would be expected in the population.

NURTURANCE/SUCCORANCE: NEEDS SUPPORT
The results of the "all shows" analysis of the second
content category, Needs Support, appear in Table 9.
Table 9 shows that, across all shows, males were in

need of support more often than females, but that needing

support occurred as a female behavior in a greater

64
proportion than what would be expected from the population
distribution of males and females. The reason for these
seemingly contradictory results can be found in Tables 9b
and 9c.

Table 9b shows that males need physical support, on the
average, three times as often as females. That is. the rate
of male need for physical support is much higher than that
of the female rate. The large and negative z-scores for all
types of physical support needs further show that males
needed support in a much larger proportion (82%) than would
have been expected in the population distribution of males
and females. Both tests, then, support the hypothesis that
male characters will be portrayed in physically succorant
conditions more than female characters.

Support is also shown in Table 9c for the hypothesis
that female characters will be portrayed in emotionally
succorant conditions more than male characters. While
there is no difference in average number of emotional support
needs of males and females, there is a large difference
proportionally. 40% of the emotional support needs are
female while female characters constitute only 27% of the
population. This difference is indicated by the large and
positive z-scores. These differences in need for support
hold regardless of the specific type of support needed, be
it a physical type or an emotional type.

Although not hypothesized, there are clear sex differences

in whether the character needing support asked for it

65

(Table 9d). Males did not ask for support at a much higher
rate than did females. There is no difference in the aver-
age number Of supports asked for by males and females. The
difference of proportions test shows, however, that not ask-
ing for support was overrepresented as a male behavior (78%
of supports not asked for) and that asking for support was
overrepresented as a female behavior (38% of supports asked
for).

Table 9e shows that males both were and were not respon-
ded to at a higher rate than females. Table 9f shows that
males both were given and were not given support at,a higher
rate than females. The difference of proportions test on
theses variables shows, however, that female support needs
were responded to in a proportion greater (38%) than that
expected from the population and support was given to females
in a proportion (35%) greater than that expected from the
population (27%) of television characters.

These tests give only indirect support for the hypo-
thesis that females will be nurtured for emotional succor-
ance more than males. This is due to the fact that the
data were not structured so that a direct test of this
hypothesis could be made. It can be inferred that indirect
support is shown, however, from the evidence that females
are typically portrayed as needing emotional support. Since
this is the case, it may be inferred that the responses and
the support given is directed mainly toward females in an

emotionally succorant condition.

6:6

The same is true for the hypothesis that males will be
nurtured for physical succorance more than females. In this
instance, however, there is indirect non-support for the
hypothesis. The difference of proportions test shows that
males were not responded to in greater proportion than would
have been expected in the population. Support was not given
in proportions equal to those expected from the population.
Drawing the same inference, there is no support for this
hypothesis.

Table 9g shows that the rates of females responding to
either sex are about the same. Males,however, average more
responses to males in need of support than to females. The
large and positive z-scores show that females were responded
to by either sex in a greater proportion than were males.
These results do not support the hypothesis that females
will respond to succorance with nurturance more than males.
They do Show, though, that females get a bigger response from
both sexes than do males when they need support.

In summary, the findings for the Support category show:

-Males have a higher rate and proportion of physical
need for support.

-Females have a higher proportion of emotional need
for support.

-Asking for support occurred as a female behavior
in a proportion greater than expected, while not asking
for support occurred as a male behavior in a proportion
greater than expected from the distribution of males
and females in the population.

-Response or non-response and giving or not giving
support occurs at a higher rate for males needing support.

67

Table 9: Means, t-tests, z—scores:
Needs Support Category, All Shows

 

 

 

 

 

 

 

 

 

 

 

Significance
Females Males of t z-score
Table 9a: ALL SUPPORT TYPES 2.16 3.47 <.0001 +3.20b
N = N = 357
Table 9b: Physical Support Types Significance
Females Males of t z-score
Physical Internal 0.21 0.62 <.0001 -2.86b
Physical External 0.25 0.85 <.0001 -4.35C
Physical Confinement 0.10 0.28 (.0001 -1.81a
ALL PHYSICAL SUPPORT TYPES 0.56 1.75 <.0001 -5.450
Table 9c: Emotional Support Types Significance
Females Males of t z-score
Ego Support 0.68 0.70 n.s. +6.25C
Concern for Others 0.34 0.27 n.s. +6.03C
Psycho Support 0.58 0.75 n.s. +3.65b
ALL EMOTIONAL SUPPORT TYPES 1.60 1.72 n.s. +8.98C
Table 9d: Support Asked Significance
Females Males of t z-score
Yes (support asked) 1.20 1.35 n.s. +7.23c
No (support not asked) 0.67 1.66 <.0001 -3.11b
Table 9e: Support Responded to Significance
Females Males of t z-score
Yes (response) 1.61 2.10 <.05 +6.16C_
No (no response) 0.26 0.91 <.0001 -4.57C
Table 9f: Support Given Significance
Females Males 0f t z-score
Yes (support given) 1.16 1.52 .<.05 +5.14C
No (support not given) 0.70 1.44 1.0001 -0.94
Table 9g: Respondents Significance
Females Males of t z-score
Female Respondents 0.39 0.44 n.s. +4.11C
Male Respondents 1.14 1.50 <.05 +4.95C

 

68
Females are responded to and given support in a higher
proportion than would be expected in the population,
while no response to males was overrepresented.

-Response to a female need for support occurs more
than would be expected from the population distribution
of males and females.

POST HOC ANALYSES: NEEDS SUPPORT

Tables 10-19 present the post hoc analyses of support
needs for program type, broadcast time, and viewing prefer-
ence.
Program Type

Consistent with the "all shows" results are the finding
for physical and emotional support types in Situation Come-
dies. Tables 10a and 11a show that there is no difference
by sex in the rates of needing emotional support, but there
is a large difference in needing physical support. These
results hold regardless of specific support type needed.
Tables 10a and 11a also show that needing physical support
is overrepresented as a male behavior (92%), while needing
emotional support is a female behavior (41%).

Tables 10b and 11b show a similar pattern of results
for Action Adventure programs. Females have a much higher
rate of needing emotional support. There is no sex difference
in the rate of needing physical support. Nonetheless, emotion-
al support is again overrepresented as a female behavior (40%)
and physical support as a male behavior (81%). Tables 100
and 11c show the same patterns of rate and proportional

difference for Medfam Dramas, but not to the extent seen in

Situation Comedies and Action Adventures.

69
Table 10: Means, t-tests, z—scores:
Physical Support Types by Program Type

Table 10a: Situation Comedies

 

Females Males Significance ..

(N=38) (N=70) 0f t zgggore
Physical Internal 0.03 0.26 (.05 -2.08a
Physical External 0.11 0.49 (.05 -2.21a
Physical Confinement 0.0 0.09 (.05 -1.74a
ALL PHYSICAL SUPPORT TYPES 0.13 0.83 (.01 -3.41b

 

Table 10b: Action Adventures

 

Females Males Significance or

(N=40) (N=125) of t z-Score
Physical Internal 0.60 0.54 n.s. —.23
Physical External 0.60 0.95 n.s. -2.81b
Physical Confinement 0.10 0.34 n.s. -2.68b
ALL PHYSICAL SUPPORT TYPES 1.30 1.82 n.s. -3.23b

 

Table 10c: Medfam Dramas

 

Females Males Significance A

(N: 40) (N=125) of t z-score
Physical Internal 0.26 0.67 n.s. -1.41
Physical External 0.37 0.47 n.s. +.39
Physical Confinement 0.15 1.37 n.s. -4.19C

ALL PHYSICAL SUPPORT TYPES 0.78 1.28 n.s. -.49

 

Table 11:

t-test,

z-scores:

Emotional Support Types by Program Type

Table 11a:

Ego Support
Concern for Others

Psycho Support

Situation Comedies

 

ALL EMOTIONAL SUPPORT TYPES 3.55

Table 11b:

Ego Support
Concern for Others
Psycho Support

Females Males Significance
(N=38) (N=70) of t z-score
1.47 1.21 n.s. +3.42b
0.74 0.34 n.s. +4.38C
1.34 1.27 n.s. +2.47b
2.83 n.s. +5.56c

 

Action Adventures

 

ALL EMOTIONAL SUPPORT TYPES 2.70

Table 11c:

Ego Support
Concern for Others
Psycho Support

 

 

ALL EMOTIONAL SUPPORT TYPES 4.15

Females Males Significance
(N=40) (N=125) of t z-score
1.25 0.38 (0.01 +5.49C
0.60 0.29 n.s. +2.34b
(3.85 0.60 n.s. +1.08
1.02? (0001 +L1’097d
Medfam Dramas
Females Males Significance
(N=27) (N=51) of t Z—score
1.59 0.86 n.s. +4.87C
0.93 0.49 n.s. +3.47C
1.63 0.94 n.s. +4.45C
2.29 <O.05 +7.44C

 

71

Table 12 shows little difference in the asking or not
asking for support. Only in Action Adventure programs do
females ask for support at a higher rate than males, other-
wise there is no sex difference in rate of asking. All three
program types also show that asking for support is overrepres-
ented as a female behavior. Not asking for support occurs
in proportion to what would be expected from the population
distribution of males and females.

Table 12a shows some interesting results for the res-
ponse variable in Situation Comedies. Females are not res-
ponded to at a higher rate than males and not responding to
females is proportionally overrepresented (63%)- Table 12b
shows, in contrast, that while there is no sex difference
in the rate of non-response, it is overrepresented as a male
behavior in Action Adventures (83%). There is no sex differ-
ence in non-response in Medfam Dramas (Table 12c). Being
responded to, however, shows no sex difference in rates in
Situation Comedies and Medfam Dramas. Females are responded
to more on the average than males in Action Adventures. Being
responded to is proportionally overrepresented as a female
behavior regardless of program type.

Table 13 shows that there is little difference of either
kind to the non-giving of support to either sex. There is a
sex difference, however, in support given. Action Adventure
programs and Medfam Dramas show a higher rate of support
.given to females. Situation Comedies and Medfam Dramas show

a greater proportion of giving support to females than would

72

Table 12: Means, t-tests, z-scores:
Support Asked and Responded to by Program Type

Table 12a: Situation Comedies

 

 

 

 

 

Females Males Significance 0‘
(N=38) (N=70) of t z-score
Yes (support asked) 2.42 1.84 n.s. +4.84C
NO (support not asked) 0.82 1.20 n.s. 0.00
Yes (response) 3.08 2.56 n.s. +4.84C
No (no response) 1.58 0.50 (0.01 +7.82C
Table 12b: Action Adventures
Females Males Significance ;:
(N=40) (N=125) of t Z—score
Yes (support asked) 1.85 1.16 (.05 +2.28a
No (support not asked) 1.65 1.59 n.s. -.83
Yes (response) 2.85 1.92 (0.05 +2.15a
No (no response) 0.65 0.83 n.s. -1.78a
Table 120: Medfam Dramas
Females Males Significance _:
(N=27) (N: 51) of t z—score
Yes (support asked) 3.04 1.82 n.s. +5.97C
No (support not asked) 1.04 1.02 n.s. +1.50
Yes (response) 3.82 2.45 n.s. +6.10C
NO (no response) 0.30 0.39 n.s. 0.0

 

Table 13:

Means,

73

t-tests, z-scores:

Support Given and Respondents by Program Type

Table 13a:

Yes (support given)

No (support not given)

Female respondents

Male respondents

Table 13b:

Yes (support given)

NO (support not given)

Female respondents

Male respondents

Table 130:

Yes (support given)

NO (support not given)

Female respondents

Male respondents

Situation Comedies

 

 

 

 

 

Females Males Significance cf
(N=38) (N=70) of t Z-snore
2.42 1.70 n.s. +5-Ll3C
0.82 1.34 n.s. -.60
1.21 0.47 < 0.01 +6.36C
1.79 1.97 n.s. +1.88a
Action Adventures

Females Males Significance .

(N=40) (N=125) t z

1.98 1.37 <0.05 +1.56
1.53 1036 nos. ‘015
0.48 0.38 n.s. +.28
2.35 1.44 <.05 +2.72b

Medfam Dramas

Females Males Significance

(N=27), 31N=51) 0f t z-score
2.74 1.55 <.05 +6.02C
1.30 1.08 n.s. +2.62b
0.89 0.61 n.s. +2.73b
2.74 1.69 n.s. +5.53C

 

74
be expected (43%, 48%, respectively).

Table 13 shows that, in general, females are responded
to more than males regardless of program type. Females are
responded to by both sexes in a greater proportion than would
be expected in Situation Comedies and Medfam Dramas. In
Action Adventure programs, however, females are responded to
only by males in a greater proportion than would be expected
(34%). Males also respond more to females on the average in
Action Adventures. There is no difference in rate or propor-
tion of female respondents for Action Adventure programs.
Broadcast Time

The broadcast time breakdowns for support type show
fairly clear—cut results in Tables 14 and 15. When there
are significant differences by sex, either in rate or pro-
portion, they are in the male direction for physical support,
and in the female direction for emotional support.

There are no sex differences in the average number of
emotional needs for support, regardless of broadcast time.
The evening time periods (Tables 14b & 0) have males needing
all types of physical support more than females.

Although all physical support types are overrepresented
as male behaviors regardless of broadcast time, different,
specific support types within each broadcast time contribute
to the overall significant difference. Physical internal
(83%) and external supports (84%) are overrepresented as
male behaviors in Saturday Morning programming. Similar dif-

ferences are found for physical confinement in the 8-9 p.m.

75

Table 14: Means, t—tests, z-scores:

Physical Support Types by Broadcast Time

 

 

 

 

 

Table 14a: Saturday Morning

Females Males Significance

(N=28)tpgttN=111)7 of t z-score
Physical Internal 0.75 0.92 n.s. -2.44b
Physical External 0.86 1.15 n.s. -3.10b
Physical Confinement 0.61 0.41 n.s. 0.0
ALL PHYSICAL SUPPORT TYPES 2.21 2.48 n.s. -3.56b

Table 14b: 8 - 9 p.m.

Females Males Significance

ptNa51) (N=126) of t z-score
Physical Internal 0.26 0.40 n.s. -1.14
Physical External 0041 0.66 n.s. —1.55
Physical Confinement CL06 0.25 (CL05 -2.33b
ALL PHYSICAL SUPPORT TYPES 0.73 1.30 <0.05 —2.71b

Table 140: 9 - 11 p.m.

Females Males Significance

(N=172) (N=120) of t z-score
Physical Internal 0.11 0.58 <0.0001 -1.20
Physical External 0.10 0.78 < 0.0001 -2.78b
Physical Confinement 0.03 0.20 <0.01 -.47
ALL PHYSICAL SUPPORT TYPES 0.24 1.56 <0.0001 -3.13b

 

76

Table 15: Means, t-test, z—score

Emotional Support Types by Broadcast Time

 

 

 

 

 

 

 

Table 15a: Satuflay Morning
Females Males Significance
(N=28) (N=111) of t z-score
Ego Support 0.79 0.66 n.s. -1.41
Concern for others 0.29 0.09 n.s. +1.58
Psycho Support 0.57 0.51 n.s. -1.05
ALL EMOTIONAL SUPPORT TYPES 1.64 1.26 n.s. -.66
Table 15b: 8-9 p.m.
Ego Support 1.28 0.98 n.s. +2130a
Concern for others 0.63 0.38 n.s. +2.15b
Psycho Support 0.92 1.07 n.s. -.33
ALL EMOTIONAL SUPPORT TYPES 2.82 2.43 n.s. +2.33b
(N=51) (N=126)
Table 15c: 9-11 p.m.
Ego Support 0.49 0.45 n. +9.04C
Concern for others 0.26 0.31 +5.74C
Psycho Support 0.48 0.64 n. +6.96C
ALL EMOTIONAL SUPPORT TYPES 1.23 1.40 n.s. +12.63C
(N=172) (N=120)

77
period, and for physical external support in the 9-11 p.m.
period.

While there is no proportional difference in needing
emotional support during the Saturday Morning period, it
occurs during the evening time periods in a higher proportion
for females than would be expected (Table 15).

There are also clear sex differences in asking for sup-
port regardless of braodcast time. Saturday Morning (Table
16a) shows the only difference in rate of asking. Females
ask for support at a higher rate than males. Across all
broadcast times, however, asking for suppport is overrepres-
ented as a female behavior.

Not asking for support occurs at a higher rate for males
during Saturday Morning and the 9-11 p.m. intervals. Males
do not ask for support in a higher proportion than expected
only during Saturday Morning shows (86%).

In terms of being responded to and given support, Tables
16 and 17 show that males are responded to and given support
at a higher rate than females only during the 9-11 p.m. time
period. Saturday Morning overrepresents males being given
support (79%), while the evening time periods overrepresent
females being given support (34%, 8-9 p.m.; 46%, 9-11 p.m.).
The same is true for being responded to, except that Saturday
Morning shows no difference in proportion.

Males are not responded to and not given support at a
higher rate than females from 9-11 p.m., but not giving

support to females occurs at a higher proportion during

Table 16:

78

Means, t-tests, z~scores

Support Asked and Responded to by Broadcast Time

 

 

 

 

 

Table 16a: Saturday Morning
Females Males Significance
(N=28) (N=11t), of t z-score
Yes (Support asked) 1.93 1.04: (0.05 +1.38
No (Support not asked) 1.50 2.32 <0.05 -5.07°
Yes (Response) 2.54 1.85 n.s. -.54
NO (No response) 0.89 1.48 n.s -4.26C
Table 16b: 8-9 p.m.
Females Males Significance
(N=51) (N=126) of t z-score
Yes (Support asked) 2.02 1.72 n.s. +2.14a
No (Support not asked) 1.08 1.33 n.s. -.76
Yes (Response) 2.80 2.48 n.s. +2.11a
No (No response) 0.31 0.58 <0.05 -1.91a
Table 160: 9-11 p.m.
Females Males Significance
(N=172) (N=120) of t z-score
Yes (Support asked) 0.84 1.25 n.s. +8.51C
No (Support not asked) 0.41 1.39 <0.0001 +.88
Yes (Response 1.11 1.93 <0.01 +8.44C
No (No response) 0.14 0.72 <0.0001 -1.28

 

79

Table 17: Means, t-tests, z-scores:

Support Given and Respondents by Broadcast Time

Table 17a: Saturday Morning

 

 

 

 

 

Females Males Significance
(N=28) (N=111) of t z—score
Support Given 1.68 1.58 n.s. -1.96a
Support Not Given 1.75 1.77 n.s. -3.29b
Female Respondents 0.29 0.41 n.s. -1.87a
Male Respondents 1.75 1.20 n.s. 0.0
Table 17b: 8-9 p.m.
Females Males Significance
(Ns51) (N=126) of t z-score
Support Given 2.18 1.69 n.s. +3.01b
Support Not Given 0.90 1.27 n.s. -1.56
Female Respondents 0.78 0.46 n.s. +3.20b
Male Respondents 2.02 1.89 n.s. +1.34
Table 170: 9-11 p.m.
Females Males Significance
(N=172) (N=120) of t z-score
Support Given 0.78 1.30 <.05 +7.42C
Support Not Given 0.47 1.33 <.0001 +2.33b
Female Respondents 0.29 0.44 n.s. +4.66C
Male Respondents 0.77 1.38 (.01 +6.76C

 

80

this period (34%). Males are not responded to more than
females from 8-9 p.m., but there is no sex difference in not
being given support. Saturday Morning overrepresents not
responding to (87%) and not giving support (80%) to males.

Table 17 shows that only during the 9-11 p.m. interval
is there a difference in respondents. Males responding to
males needing support occurs at a higher rate than males re-
sponding to females. Proportionately, however, both sexes
respond more to females (40% female and 44% male respondents).
Table 17b shows that, proportionally, females respond more
to females (41%) and there is no difference in the respond-
ing of males during the 8-9 p.m. period.

Saturday Morning (Table 17a) shows a much different
result. Males respond to both sexes in their expected pro-
portion, but females respond to males (85%) at a greater
proportion than would be expected.

Viewing Preference

In terms of needing support types, the preferred pro-
grams show a distinct pattern of results (Tables 18a). Males
show a higher rate and proportion (87%) of needing physical
support. Females show a higher rate and proportion (42%) of
needing emotional support.

The results for the non—preferred programs are not as
clear cut (Table 18b). Males have a higher rate for needing
all types of physical support and emotional support in terms
of Psychological support. There is no sex difference in

rates for the other two emotional types of support.

81

The difference of proportions test for non—preferred
programs provides some interesting results. Needing physical
internal support is overrepresented (35%) as a female behavior,
while needing physical external support is overrepresented
(84%) as a male behavior. There is no sex difference in
physical confinement. Needing emotional support is over-
represented (36%) as a female behavior, with the exception
of psychological support, where there is no difference.

Asking for support in preferred shows (Table 19a) occurs
at a higher rate for females and is overrepresented (30%) as
a female behavior. While asking for support in non-pre—
ferred shows (Table 19b) is overrepresented as a female
behavior (38%) it occurs at a higher rate for males. In
both the Top 40 and Low 60 programs, males do not ask for
support more often than females (80%, 75% respectively)
with the exception of the non—preferred proportion, which
shows no difference.

Table 19 shows very similar results for being responded
to and given support. Preferred shows have higher female
rates of being responded to and given support. Non-prefer-
red shows have higher male rates on these variables. There
is no difference in not being given support in preferred
shows, but males have a higher rate of not being responded
to. Non-preferred programming has a higher rate on both
variables for males.

There is no proportional difference in not being given
support or not being responded to in non-preferred programs.

Preferred programs overrepresent these variables as a male

82

Table 18: Means, t-tests, z-scores:

Support Types by Viewing Preference

 

 

 

 

 

Table 18a: Preferred Shows
Females Males Significance

_LN=64) (N=180) of t z-score
Physical Internal 0.17 0.78 <0.0001 _5.480
Physical External 0.63 1.05 <0.05 -3.27b
Physical Confinement 0.13 0.31 <0.01 -2.55b
ALL PHYSICAL SUPPORT TYPES 0.92 2.14 <0.0001 -6.52C
Ego Support 1.58 0.87 < 0.05 +4.51C
Concern for Others 0.84 0.32 < 0.01 +5.13C
Psycho Support 1.23 0.61 <0.05 +4.59C
ALL EMOTIONAL SUPPORT TYPES 3.66 1.79 1<0.01 +8.02C

Table 18b: Non-Preferred Shows
Females Males Significance

_LN=187) (N=177) of t z-score
Physical Internal 0.23 0.45 1<0.05 +1.83a
Physical External 0.12 0.66 <0.0001 -2.88b
Physical Confinement 0.09 0.25 <0.01 +.29
ALL PHYSICAL SUPPORT TYPES 0.43 1.36 (0.0001 -.75
Ego Support 0.37 0.53 n.s +4.59C
Concern for Others 0.17 0.22 n.s. +3.23b
Psycho Support 0.35 0.90 <0.0001 +.75
ALL EMOTIONAL SUPPORT TYPES 0.89 1.65 <0.01 +4.52C

 

Table 19:

Support Asked, Responded To,

Means,

83

t-teStS ,

Given,

z-scores:

by Viewing Preference

and Respondents

 

 

 

Table 19a: Preferred Shows
Females Males Significance
(N=64) (N=180) of t z-score
Yes (support asked) 2.34 1.32 <0.01 +5.14C
No (support not asked) 1.38 2.01 (0.05 -3.51b
Yes (response) 3.25 2.19 (0.05 +4.13C
No (no response) 0.48 1.13 <0.0001 -4.71C
Yes (support given) 2.45 1.56 (0.05 +4.200
No (support not given) 1.25 1.67 n.s. -2.65b
Female respondent 0.81 0.46 <0.05 +3.11b
Male respondent 2.33 1.47 €0.05 +4.09C
Table 19b: Non-Preferred Shows
Females Males Significance
(N=187) (N=177) of t z-score
Yes (support asked) 0.81 1.38 <0.01 +5.09C
No (support not asked) 0.42 1.31 <0.0001 -.64
Yes (response) 1.05 2.01 <0.0001 +4.60C
No (no response) 0.18 0.68 <(L0001 -1.45
Yes (support given) 0.72 1.49 <CL0001 +3.17b
No (support not given) 0.51 1.21 "<0.0001 +1.53
Female respondent 0.24 0.41 £0.05 +2.69b
Male Respondents 0.73 1.54 .<0.0001 +2.90b

 

84

’4

behavior. Being given support and being responded to are
overrepresented as female behaviors in both the preferred and
non-preferred programs.

Females are responded to by both sexes at a higher
rate and proportion than males in preferred shows (Table
19a). Non-preferred programs Show, however, that males are
responded to at a higher rate by both sexes, while females
are responded to in a greater proportion by both sexes
(Table 19b).

In summary, the post hoc findings for the Support
Category show:

-Regardless of program type, the resultsshow a
general pattern of higher male rate and overrepresen-
tation of males needing physical support, while needing
emotional support is overrepresented as a female behavior.

-Regardless of program type, females needing
support are responded to and given support at a higher
rate and in a greater proportion than males.

—Females are responded to by both sexes more than
males in Situation Comedies and Medfam Dramas. Females
are responded to more by males in Action Adventures.

-Saturday Morning programs overrepresent needing
physical support as a male behavior. Otherwise, there
are no sex differences in the rate or proportion of

needing support, regardless of support type.

-Regardless of broadcast time, males need physi—
cal support more and females need emotional support more.

-Across all broadcast times, asking for support is
overrepresented as a female behavior.

—Females responding to males is overrepresented
during Saturday Morning programming.

:Preferred shows have higher rates and proportions
of males needing physical support, while females have
higher rates and proportions of needing emotional support.

85

-Being given support and being responded to occur
at a higher rate for females in preferred shows. They
occur at a higher rate for males in non-preferred
shows. They are overrepresented as a female behavior
in both preferred and non-preferred shows.

—Response to a female's need for support by both
sexes occurs in a proportion greater than would be
expected in the population in both preferred and non-
preferred programs.

INDEPENDENCE/DEPENDENCE: MAKES PLANS

The results of the "all shows" analysis of the Plans
category appear in Table 20. Support for the hypothesis
that males will make more plans than females can be found
in Table 20a. Although there is no difference in the
average number of plans made by males or females, the large
and negative 2 indicates that making plans is overrepresented
as a male behavior. So, while there is no difference in
the rate of making plans, males make plans in a proportion
greater (82%) than what would be expected in the population
distribution of males and females (73% and 27%, respectively).

Although there is no hypothesis stated concerning who
these plans are made for, the results in Table 20b follow
those in 20a. There is no difference in the rates of making
plans for others, but making plans for others is overrepres-
ented as a male behavior regardless of the sex of the charac-
ter the plan is made for. (Males make 86% of the plans made
for females, and 82% of the plans made for males). That is,
males make plans for others in a greater proportion than
females make plans for others.

There is also no hypothesis concerning who executes the

plan. The results in Table 20c show, hOWEVEr,that in terms

Table 20:

Means,

86

t-tests,

z-scores:

Plans Category All Shows

Female N=44
Male N=167

Table 20a: Makes Plans

Table 20b: Plans for
Others

For Females

For Males

Table 200: Executors
Female executors

Male executors

Table 20d: Outcomes
Success

Failure

 

 

 

 

 

 

 

Females Males Significance z-score
of t

1.41 1.64 n.s. -3-45b
0.14 0.20 n.s. -1.77a
0.66 0.80 n.s. -2.65b
0.50 0.04 <0.0001 +5.81%
0.32 1.08 <0.0001 —6.28°
0073 0089 n08! ‘208Li’b
0.59 0.65 n-S- 'I'95a

 

87
of both rate and proportion, females execute plans made by
females, and males execute plans made by males.

There is support for the hypothesis that male plans will
succeed more than female plans (Table 20d). Although there
is no difference in the success rate of plans, a success-
ful plan is overrepresented as a male outcome. While sup-
port for this hypothesis would imply that more female
plans would end in failure, the data do not support such
an implied hypothesis. There is no difference in the fail-
ure rate, but unsuccessful plans are overrepresented as a
male behavior.

In summary, the findings for the Plans category are:

-Making plans occurs as a male behavior in a pro-
portion greater than would be expected in the popula-

tion of males and females.

-Making plans for others (male or female) occurs
as a male behavior in a proportion greater than expected.

-Females carry out plans made by females, and males
execute plans made by males. Both occur at a higher
rate and proportion than would be expected.
-The success and failure of a plan is represented
as a male behavior in a greater proportion than would
be expected in the population.
POST HOC ANALYSIS: MAKES PLANS
Tables 21-26 present the results of the post hoc anal-
ysis for program type, broadcast time, and viewing preferences.
Program Type
Table 21a shows that plan making by males occurs at
a higher rate than plan making by males in Situation Comedies.

Table 21b shows that plan making is overrepresented (90%) as

a male behavior in Action Adventure programs. There is no

Makes
For

For

Makes
For

For

Makes
For

For

Table 21:

Means,

88

t-tests, z-scores:

Makes Plans and For Others by Program Type

 

 

 

 

 

Table 21a: Situation Comedies
Females Males Significance
(N=17) (N=39) of t z-score
Plans 1.06 1.33 (0.05 -.27
Females 0.12 0.21 n.s. -.69
Males 0.35 0.56 n.s. -3.30b
Table 21b: Action Adventures
Females Males Significance
(N=7) (N=53) of t z-score
Plans 1.43 1.76 n.s. -3.98C
Females 0.43 0.25 n.s. -.58
Males 0.86 1.09 n.s. -3.12C
Table 210: Medfam Dramas
Females Males Significance
_LN=10) (N=22) of t z-score
Plans 1.50 1.41 n.s. +1.01
Females 0.10 0.32 n.s. -.82
Males 0.80 0.46 n.s. +1.58

 

Table 22:

89

Means, t-tests,

Z-SCOI‘GS:

Plan Executors and Outcomes by Program Type

Female Executors

Male Executors

Success

Failure

Female Executors

Male Executors

Success

Failure

Female Executors

Male Executors

Success

Failure

 

 

 

 

 

Table 22a: Situation Comedies
Females Males Significance
_(N=17) (N339) of t z—score
0.53 0.03 <0.01 +4.14C
0012 0085 <O0O()Ol -2070b
0059 0074 11080 "036
002““ 0039 n03 -052
Table 22b: Action Adventures
Females Males Significance
(N=7) (N:53) of t z-score
0.57 0.06 n.s. +1.67a
0.29 1.26 <0.05 --4.58C
0057 1000 nos “3031b
0.86 0.70 n.s. -2.O4a
Table 220: Medfam Dramas
Females Males Significance
(N=1Q), (N=22) of t z-score
0.50 0.09 n.s. +2.50a
0030 O0 91 (0005 "10Ll’3
0.80 1.00 n.s. 0.0
0.60 0.59 n.s. + .52

 

90
difference in plan making in Medfam Dramas (Table 210).

Table 21 shows that, as in the "all shows" analysis,
males are proportionately overrepresented as making plans
for males in Situation Comedies (79%) and Action Adventures
(91%). There is no difference in Medfam Dramas. This gener-
al finding of no difference in Medfam Dramas may be a
function of the small number of plans made in these shows,
as well as the equal rate and proportional representation
of plan making in these shows.

The results of the analysis of plan executors (Table
22) show the same pattern as the "all shows" findings.

Males carry out male plans and females carry out female

plans. Table 22 also shows no difference in the success

and failure rates across all program types. Proportionally,
however, successful (93%) and unsuccessful (80%) plans are
overrepresented as male behaviors in Action Adventure programs
but not in the other program types.

Broadcast Time

Tables 23-24 present the results of the broadcast time
post hoc breakdown for the Plans category. The results are
fairly consistent with the "all shows" and program type
analyses.

Table 23 shows no difference in the average number of
plans made by either sex. Regardless of broadcast time,
however, plan making is represented as a male behavior in
a proportion greater than would be expected in the popula-
tion.

When making plans for others, the male behavior sequence

91
once again wins out in general. Table 23a shows no differ-
ence in the rate or proportion of making plans for males
during Saturday morning shows. It does show that males
make plans for others at a higher rate and proportion than
females make plans for others. It is interesting to note
that there are no female plans made for females during this
time period. Table 23b shows no sex difference in the
average number of plans made for others or in the propor—
tion of their occurrence during the 8-9 p.m. interval. Dur-
ing the 9-11 p.m. period, males make plans for other males
at a higher rate and proportion, 87% (Table 23c). There is
no difference in the rate or proportion of plans either sex
makes for females.

In terms of execution there is no sex difference in the
average number of plans carried out by either sex on Saturday
morning (Table 24a). During the 8-9 p.m. period, females
show a higher rate of executing female plans, while males
show a higher rate of executing male plans (Table 24b).
Females execute male and female plans at the same rate from
9-11 p.m. (Table 240), while males execute male plans at
a higher rate than female plans. Proportionally, females
execute female plans in a greater proportion (55 to 92%)
and males execute male plans in a greater proportion than
expected (90 to 94%). This result holds regardless of
broadcast time.

Table 24 shows no difference in rate or proportion of

the success or failure of plans except during the 9-11 p.m.

92

Table 23: Means, t-tests, z—scores:

Makes Plans and for Others by Broadcast Time

Table 23a: Saturday Morning

 

Females Males Significance
(N=10) (N=53) of t z-score
Makes Plans 1.90 1.83 n.s. —2.52b
For Females 0.0 0.11 <0.05 -1.74a
For Males 0.90 0.81 n.s. -1.56

 

Table 23b: 8-9 p.m.

 

Females Males Significance
(N=18) (N=62) of t z-score
Makes Plans 1.33 1.50 n.s. -1.66a
For Females 0.22 0.27 n.s. -.97
For Males 0.67 0.58 n.s. -.32

 

Table 230: 9-11 p.m.

 

Females Males Significance
(N=16) (N352) of t z-score
Makes Plans 1.19 1.60 n.s. -2.00a
For Females 0.13 0.21 n.s. -1.20

For Males 0.50 1.04 <0.05 -2.56b

 

93

Table 24: Means, t-tests, z—scores:

Plan Executors and Outcomes by Broadcast Time

Table 24a: Saturday Morning

 

 

 

 

 

Females Males Significance
(N=10) (N=53) of t z-score
Female Executors 0.40 0.02 n.s. +3.37b
Male Executors 0.70 1.15 n.s. -3.02b
Success 1.00 0.85 n.s -1.52
Failure 1.00 0.83 n.s. -1.52
Table 24b: 8-9 p.m.
Females Males Significance
(N=18) (N=62) of t z-score
Female Executors 0.67 0.02 <0.01 +4.82C
Male Executors 0.17 0.98 <0.0001 -3.97C
Success 0.78 0.81 n.s -.85
Failure 0.56 0:53 n.s. -.68
Table 2400 9'11 p0m0
Females Males Significance
(N=16) 5(N%52) of t z-score
Female Executors 0.38 0.10 n.s. +2.03a
Male Executors 0.25 1.14 <0.0001 -3.69b
Success 0.50 1.04 (0.05 -2.56b
Failure 0.38 0.62 n.s. -1.48

 

94
period. Table 240 shows that more male plans are success-
ful on the average, and that successful male plans are
overrepresented (89%).
Viewing Preference

The breakdown by viewing preference (Tables 25-26)
shows the same results as the "all shows" analysis. Re-
gardless of viewing preference, there is no difference in
the average number of plans made by either sex, but plan
making is overrepresented as a male behavior (Table 25).

There is also no difference in the average number of
plans made for others (Table 25). With the exception of
plans made for females in non-preferred shows, males making
plans for others occurs in a proportion greater than would
be expected (82 to 92%).

Execution of plans follows the pattern stated earlier,
regardless of viewing preference (Table 26). Females exe-
cute plans made by females at a higher rate and proportion
(90%, 68%) than plans made by males. Males execute plans
made by males at a higher rate and proportion (94%, 91%)
than plans made by females.

Table 26 shows no difference in the success or failure
rate of plans made by either sex. Preferred programs repre-
sent the success (84%) or failure (81%) of a plan as a male
behavior in a proportion greater than expected. There is no
proportional difference in the success or failure of plans
in non-preferred shows.

In summary, the post hoc findings of the Plans category

Table 25:

Means, t-tests,

95

Z-SCOI‘ES:

Makes Plans and for Others by Viewing Preference

 

 

 

Table 25a: Preferred Shows
Females Males Significance
(N=21) ,(N=93) of t z-score
For Females 0.10 0.24 n.s. -1.89a
Table 25b: Non-Preferred Shows
Females Males Significance
(N=23) (N=74) of t z-score
Makes Plans 1.30 1.55 n.s. -1.69a
For Females 0.17 0.16 n.s. 0.0
For Males 0.61 0.87 n.s. -1.79a

 

96

Table 26: Means, t-tests, z-scores:

Plan Executors and Outcomes by Viewing Preference

Female Executors

Male Executors

Success

Failure

Table 26b:

Female Executors

Male Executors

Success

Failure

 

 

 

Females Males Significance
(N=21) (N=93) of t z-score
0.43 0.01 <.01 +4.14C
0029 1005 <.0001 '4087C
0.76 0.91 n.S. -2.47b
0.71 0.71 n.s. -1-75a
Nonpreferred Shows

Females Males Significance

(N=23) (N=74) of t z—score
0.57 0.08 <.01 +4.170
0.35 1.12 <.0001 -3.99C
0070 0087 n0S0 “1050

0048 0058 nos0 ‘1022

 

97
show:

-Females make more plans on the average than males
during Situation Comedies. Making plans is a male
behavior in Action Adventure programs.

-Making plans for others is proportionately over-
represented as a male behavior in Situation Comedies
and Action Adventure programs.

-All post hoc breakdowns show that females execute
female plans and males execute male plans.

-Plan making is overrepresented as a male behavior
regardless of broadcast time.

-Only during the 8-9 p.m. time period is there no
sex difference in making plans for others.

-Only during the 9-11 p.m. time period are male
plans successful more than female plans.

-Results of the viewing preference analysis follow
those of the "all shows" analysis.

SUMMARY
By way of summary, each hypothesis generated in Chapter
One will be restated and discussed in terms of the data's
support or non—support of it. The implications and general
findings of the results of this analysis will be discussed
further in the next chapter.

H1: Male characters will give proportionately more
authority orders than female characters.

This hypothesis is supported (Table 2a). The significant
t-test indicates that males give authority orders at a higher
rate than females, while the z-score indicates that males
give proportionately more authority orders than females.

In fact, 87% of the authority orders given in this sample
of television programs were given by male characters.

H . Male and female characters of the same status

2 will give proportionately equal numbers of simple
orders.

98-

This hypothesis is not supported (Table 2a). Once
again, males give simple orders at a higher rate and pro-
portion than females. The proportional difference for
simple orders is not as great, however, as the proportional
difference for authority orders. Males gave 76% of the
authority orders; females gave 24%.

H3: Male characters will use proportionately more
threat orders than female characters.

This hypothesis was not tested with the data. The
number of threat orders given in this sample of television
programs (87 total threat orders) was too small to justify
further analysis. This small number means that just over
one threat order was identified per television program.

H4: Female characters will explain proportionately
more of their orders, authority or Simple, than
male characters.

This hypothesis was not supported (Table 2a). In

terms of both rate and proportion, males gave more explained
orders. Males gave 86% of the authority explained orders,
and 79% of the simple explained orders.

H5: Orders given by male characters will be followed
proportionately more often than orders given by
female characters.

This hypothesis was supported (Table 20). Male orders
were followed both at a higher rate and in a greater propor-
tion than female orders. Although male orders also were not
followed at a higher rate and proportion, a greater propor-
tion of female orders were followed than were not followed.

'18% of all followed orders were given by females, while 25%

of all not followed orders were given by females.

'99

H6: Male characters will order other male characters
proportionately more often than female characters
will order male characters.

This hypothesis is supported (Table 2b). The male-
male orders behavior sequence occurs at a higher rate and
proportion than the female-male behavior sequence. 82% of
the orders given to other males were given by males.

H7: Female characters will be the receivers of orders
proportionately more than males will be the receiv—
ers of orders, regardless of the sex of the order
giver.

This hypothesis is not supported (Table 2b). Females
receive orders from males and from females in a proportion
almost equal to that expected from the population. The
female-female behavior sequence constitutes 25% of the
orders given to females.

H8: Male characters will be portrayed in physically
succorant conditions proportionately more than
female characters.

This hypothesis is supported (Table 9b). Males need
physical support at a higher rate and in a greater propor-
tion than female characters. 82% of the physical support
needs were portrayed by male characters.

H9: Female characters will be portrayed in emotional-
ly succorant conditions proportionately more than
male characters.

This hypothesis was supported (Table 90). Although

there is no difference in the rate of needing emotional
support, females need proportionately more emotional support.

40% of the emotional support needs were portrayed by female

characters while females constitute only 27% of the

100
population of characters.

H Female characters will respond to succorance
with nurturance proportionately more than male

characters.

10‘

This hypothesis was not supported (Table 9g). The
rates of females responding to either sex are about the
same, while males respond to males at a higher rate than to
females. Proportionally, females are responded to more by
both sexes: 38% of the female response to a need for sup—
port is to females; 35% of the male response to a need for
support is to females.

H11‘ ”313.: iiiiiifiiiiefitilrbi “it”??? (“0913.51 ”Si”

proportionally more tRanpfggaIe cizgggiergT

This hypothesis (Tables 9c and 9f) was not supported
but with indirect evidence. A direct test of the hypothesis
was not possible due to the structure (indexing) of the'
data. Males are typically portrayed as needing physical
support and from this it can be inferred that males were
not responded to for physical succorance more than females.

H Female characters will be nurtured (receive

positive reinforcement) for emotional succorance
proportionately more than male characters.

12‘

This hypothesis was supported (Tables 9e and 9f) but
with indirect evidence. As with hypothesis 11, a direct
test was not possible. Support is found through the infer-
ence that females are typically portrayed as needing emotion-
al support and are responded to in a proportion greater than
expected from the population.

H : Male characters will make plans proportionately
13
more than female characters.

103

This hypothesis was supported (Table 20a). Males made
a greater proportion of plans than that expected from the
population. 82% of the plans identified in this sample of
television programs were made by males, while 18% were made
by females.

H14: Male characters will formulate proportionately
more ultimately successful plans than female
characters.

This hypothesis was supported (Table 20d). Successful
plans were made by males more often than by females. 82%
of the successful plans were formulated by males. This
hypothesis does not receive unqualified support, however,
because 81% of unsuccessful plans were also formulated by
males.

While not all hypotheses received support from the data
the instances of non-support lend support to the notion that
character portrayals are stereotyped on television. This

general result will be discussed at greater length in the

next chapter.

CHAPTER FOUR
DISCUSSION

This final chapter will discuss and interpret the re—
sults of this content analysis in terms of the stereotyp-
ing and reinforcement of sex role behavior on television.

To make interpretation of the general findings easier, a
series of summary tables have been constructed. These tables
provide a visual presentation and distillation of the large
amount of data presented in the Results section.

Tables 27-29 present a summary of the "all shows"
findings for Orders, Support, and Plans. These tables are
constructed so that it is possible to see clearly which
behaviors may be considered male or female behaviors.

Table 27a, made up entirely of "male" entries shows
that order giving behavior is clearly a male act, regard-
less of the statistical test used. Table 27b shows the same
thing for needing physical support. Needing emotional
support is only a female behavior in terms of proportional
representation. Table 27c provides similar information for
making plans: It is a male behavior only in terms of propor-
tional representation.

There are some apparently incongruous findings in these
tables, as well as the tables discussed earlier. For example

102

103
the results of the all support types analysis in Table 27b
shows that in terms of rate, males dominate, but in terms
of proportion, females dominate. How should these findings,
as well as those which show no difference for one test, and
a significant difference for the other test, be interpreted?

The difference in rates of needing support means that
among all characters who needed at least one kind of sup-
port, males needed support more often. And so, "males"
appears in the t—test column. The difference in proportions
of needing support means that, among all speaking characters,
more than 27% of needs support behaviors would be portrayed
by females. And so, "female" appears in the z-score column
for that variable.

A simpler, although not as precise, way of stating the
difference in interpretation is: "male" or "female" in the
t-test column means that one sex or the other averaged more
portrayals of that behavior. A "male" or "female" in the
z-score column means that the behavior was portrayed in a
greater than expected percentage by one sex or the other.

Inserting the example from Table 27b, the above would
read: "male" in the t-test column means that males average
more portrayals of needing support (one male would need
support more often than one female). "Female" in the z-score
column means that needing support was portrayed by females
significantly more than 27% of the time (needing support oc-
curs more often than expected in all female portrayals than

it does in all male portrayals.)

104
This interpretation can be extended to explain why,

in Table 27b, there is no difference in rate of all emotional
types of support needed, yet proportionally it is a female
behavior. Here the interpretation would read: the "n.d."

in the t-test column means that neither males nor females
averaged more portrayals of needing emotional support.
"Female" in the z-score column means that needing emotional
support was portrayed by females more than 27% of the time.

Moving on to Table 28, the direction of difference in
receivers/respondents is again clear. Table 28a shows that
males order everybody, regardless of sex. Table 28b shows
that males average more responses to males needing support,
while responding to females needing support occurs regard-
less of the sex of the respondent. Table 28c shows that
female plans are executed by females more on the average and
in percentage portrayal. Male plans are executed more by
males on the average and in percentage portrayal.

Table 29 shows a preponderance of outcomes in the male
direction, regardless of category. The difficulty here is
interpreting how positive and negative outcomes can occur
for males. The logical result would be that of matching
up one outcome with each sex, yet the findings are that male
orders are followed and not followed (Table 29a); that males
are and are not given support more( Table 29b); and that
male plans succeed and fail more (Table 29c).

These findings should be viewed in conjunction with the

findings in Tables 27-29. In terms of orders, males give

105

Table 27:
Direction of Difference: t-tests and z-scores:

Orders, Plans, Support: All Shows

Table 27a: Order Types

Female N=294

Male N = 395

 

 

 

 

 

t z
Authority Male Male
Simple Male Male
Authority Explained Male Male
Simple Explained Male Male
ALL ORDER TYPES Male Male
Table 27b: Support Types
Female N=251
Male N=357
t 2
Physical Support
Physical Internal Male Male
Physical External Male Male
Physical Confinement Male Male
ALL PHYSICAL TYPES Male Male
Emotional Support
Ego Support n.d. Female
Concern for Others n.d. Female
Psycho Support n.d. Female
ALL EMOTIONAL TYPES n.d. Female
ALL SUPPORT TYPES Male Female
Table 270: Plans
Female N=44
Male N=167 t 2
Makes Plans n.d. Male

 

106

Table 28

Direction of Difference: t-tests and z-scores:

Respondent/Receiver: Orders, Support, Plans: All Shows

Table 28a: Order Receivers
Female N = 294
Male N = 395

 

t 2
Female Receivers Male Male
Male Male

Male Receivers

 

Table 28b: Support Respondents

 

Female N = 251
Male N = 357
t 2
Female Respondents n.d. Female
Male Female

Male Respondents

 

Table 28c: Plan Executors

 

Female N = 44
Male N = 167
t 2
Female Executors Female Female
Male Male

Male Executors

 

' 107

Table 29
Direction of Difference: t-tests and z-scores:
Orders, Support, Plans, All Shows and Outcomes
Table 29a: Orders Followed

Female N = 294
Male N = 395

 

t z
Followed Male Male
Not Followed Male n.d.

 

Table 29b: Support Given

 

Female N = 251
Male N = 357
t 2
Given Male Female
Not Given Male n.d.

 

Table 29c: Plan Outcome
Female N = 44
Male N = 167

 

Success n.d. Male

Failure n.d. Male

 

108
more orders to everybody so the outcomes, positive or
negative, should reflect male order-giving in general.
For needing support, males average more support needs and
more male respondents, so the outcome rate should reflect
male support needs in general. This is also true for
females in terms of support portrayals (z-scores). The
outcomes should reflect female support needs. In the
Plans category, males make more plans, which are executed
by males, so the outcomes should reflect male plan-making
behavior.

To find out which outcome occurs most often, the best
indication of this is the raw frequency of occurrence.
Treating the outcome of a behavior as two separate variables
obscures the relationship between them. Therefore, across
all shows, more male orders were followed (1252 followed
orders, 478 not followed orders); more males were given
support (542 acts of giving support, 514 acts of not giving
support to males); and more plans succeeded (148 successful
plans, 108 non-successful plans).

This comparison of raw frequencies points up one of
the problems in this analysis- comparision across variable
splits were not possible due to the structure of the data.
Each content category was indexed for reliability estimation
and analysis. While this allowed the use of the t-test

within categories, the data are still essentially nominal.

109
Comparison across the indexed categories is not justified
given the nominal structuring of the data.

Another weak point in this study is reliability estima-
tion. Although individual behaviors could be identified
by the time segment within which it was coded, it was still
not possible to match up individual acts by coder. There-
fore, the reliability estimate used in this study was not
the most rigorous test of reliability because only the total
sum of each variable was correlated for the coders. This
estimate gives an indication only of how many acts coders
have identified the same, not whether individual acts were
identified the same. A preferred method would be percentage
agreement.

Percentage agreement is simply counting the acts that
coders agree on, then dividing by the total acts they saw.
It is more tedious, but more accurate. This provides two
estimates: an act by act estimate, are the coders seeing
the same thing at the same time; and an overall estimate,
are they seeing the same frequency of a specific variable?
This provides more information than the reliability estimates
used in this study.

A problem that may never be resolved and is inherent
to all content analysis studies, is, even when the coders
are reliable in the recording of acts, is that what is
really going on? Two coders may agree on a simple order,
but is it really a simple order? Would an untrained observ-

er identify the order as such? It is actually a validity

110
question and can only be answered through replication of
the content analysis and the study of television viewers
in regard to their ability to identify the orders, supports,
and plans analyzed for this study.

While the results presented in this thesis are clear
cut in many instances, they serve to generate more questions
about the content of television programming. Before entering
into a discussion of the models and outcomes present in this
sample of television shows, these questions should be consid-
ered.

While 1212 speaking characters were identified in this
sample, not all of these characters played major roles in
these programs, nor are they all seen regularly across a
television season. A major question that should be asked
of these data is the extent to which these behaviors are
typical of major, regular characters. The answer to this
question was not available from this data set. A secondary
analysis combining the demographic characteristics of tele-
vision characters with the sex role behaviors as defined in
this study would provide the answer (the demographic analysis
of Simmons, gt gt, 1977, provided the information that would
allow this analysis).

The question must be considered due to the nature of
social learning theory. Models which are seen on a regular
basis and receive a lot of air time in television programs,
are potentially stronger models for television viewers.

They are more familiar and receive more attention. The

111

effect of any stereotyping of the analyzed behaviors would
be diminished if they are being portrayed by "fringe" (non-
regular, minor) characters. If however, a large portion
of regular, major characters portray the behaviors analyzed
in this content analysis, conclusions and generalizations
about the potential effect of these portrayals on viewers
would have more credence theoretically.

Another analysis which would have provided more informa—
tion regarding stereotyping is an analysis of the same—sex
or cross-sex interactions among characters. With the aggre-
gation and indexing of this data set, individual behavior
sequences were obscured. While the difference of proportions
test used in this analysis gives an indication of which
sequences predominate, it is not possible to say, for in-
stance, that the male response to female need for support
occurred most often in cases of physical need for support.
Such information is not retrievable from this data set.
Presently it is only possible to make inferences regarding
the actual sequencing of behaviors. In the revision of the
coding scheme for the second year of analysis1 this considera-
tion was included. Each behavior sequence recorded will also
be characterized by the same-sex or cross-sex interaction it

engendered.

 

1This sex role content analysis is the first of three ana-
lyses to be completed for Project Castle, done over three
years of television. As part of a three-year project, it
is possible to learn from initial mistakes and attempt to
correct them. In the discussion of needed revisions, steps
have been taken to incorporate the changes suggested here
into succeeding analyses.

112

This study and analysis, then is not without its
faults. Care has been taken to correct the recognized
faults in the second year of analysis. With these caveats
in mind, then, a discussion of the theoretical and research
implications of this study follows.

The analysis of these data provides fairly good evi-
dence for two of the three factors of stereotyping as de-
fined by Bowes (1976) and Carter (1962). Within each
category of behavior, the images resulting from the findings
show a homogeneity and pglarization of behavior.

These behaviors are homogeneous for the sexes because
it is possible to predict from one characteristic to the
other and back. Knowing that an order has been given, one
would be most likely to predict that a male had given it.
Knowing that a male is a speaking character, one would be
most likely to predict that he would be the one to give
an order. The prediction would not be perfect, to be sure,
nor could one predict the making of plans, due to the lower
frequency and reliability of the plan making category.
Nonetheless, there is evidence for the homogenization of
sex role images among these variables.

Polarization is evident in that the attributes of one
sex do not apply to the other sex. Males give orders, females
don't. Females need emotional support, males don't. Males
need physical support, females don't. Males make plans,
females don't. With this generalization, the following

behavior sequences might be considered "typical" of this

112a
sample of television shows:

Orders: A male character gives an order to another
male, who follows it.

Support: A female in need of emotional support will
be responded to and given support.

A male in need of physical support will be
responded to and given support.

Plans: A male makes a plan for another male to carry
out and it will be successful.

Across all of the post hoc analyses it becomes evident
that the breakdown by program type may be the most useful
for studying the stereotyping of the sexes. Situation Com—
edies and Action Adventure programs are found across all
broadcast times and viewing preferences. The results are
"muddier" for the broadcast time and viewing preference
breakdowns.

The program type results are fairly clear cut and appear
redundant with the broadcast time and viewing preference
breakdowns. Within the program type breakdown, the Situation
Comedies contrast with Action Adventure programs. Saturday
Morning shows are probably most properly included in this
breakdown rather than the broadcast time breakdown because
they do not overlap with other program types.

Taking a closer look at the program type breakdown,
each program type can be characterized by a particular pattern
of interaction. Plan making, however, is a male behavior
regardless of program type. Situation Comedies are the most

female oriented programs. Only in this program type do females

dominate males in their order giving behavior. Females give

112b
simple orders to other females, who follow them. More emotion-

al support is needed in Situation Comedies. Everyone (male
or female) responds to females in need of support, and
females are given support more often.

Action Adventure programs lean towards a male dominance.
Males give orders to other males, who follow them. More
physical support is needed in Action Adventures. When some-
one is in need of support, however, the female will be respond-
ed to more than the male, and this response will come from
a male.

Saturday Morning programming shows heavy male dominance
in all areas. Males give orders to other males, who follow
them. Males need physical support and are responded to and
given support by other males. And it is only during Saturday
Morning that males make plans for females as well as males.

Therefore, while the "all shows" analysis provides
support for stereotyping by sex, the program type breakdown
gives a clearer indication of what kind of stereotyping is
going on. It is almost possible to place the program types
on a continuum from female dominance to male dominance, with
Situation Comedies at one end, Saturday Morning shows at the
other, and Action Adventures falling somewhere in between,
but closer to Saturday Morning than Situation Comedies.

The only factor of the stereotyping process for which
there is no evidence is that of the fixedness of these images.

This study, however, is only the first in three seasons of

content analysis. When the analysis of all three years of

112C

network programming has been completed, the question of the
persistence of these attributes can be considered. It will
be especially interesting to see whether the program type
breakdown provides similar results across the three years
of content analysis.

In terms of social learning theory, there is a shift in
focus from the dominant attributes of stereotyped images to
the behavior sequences themselves. It was stated in the first
chapter that the outcome of the behavior (which is considered
relatively neutral) is of more interest. It is the outcome
of the behavior which increases or decreases the likelihood
that the behavior will be imitated. In this study positive
outcomes have been conceptualized as those that offer success
to the intial behavior. For order giving behavior (dominance)
this means having the order followed (deference). In needing
support (succorance), this means being responded to and given
support (nurturance). When making a plan (independence),
it means that the plan is carried out successfully (dependence).

Those behaviors which have positive outcomes (or are
positively reinforced) are the behaviors with an increased
likelihood of imitation. In terms of imitation of sex-typed
behaviors, however, the sex of the imitator (as well as other
factors to be discussed later) must be taken into account.
Mischel's (I966) interpretation of social learning theory
in the development of sex roles states that children will

come to value and model those behaviors which are performed

by their own sex.

112d

Given the results of this study, then, the theory would
predict that it is acceptable for boys to be dominant (give
orders because that dominance will be met with deference.
Dominance would then be a behavior that boys would value and
perform.

For the other dimensions of behavior the theory would
predict that it is acceptable for girls to be in emotionally
succorant conditions because that succorance would be met
with nurturance. Support was not found for the hypothesis
that males would be nurtured for physical succorance, so while
physical succorance is a male behavior, it is not positively
reinforced. Boys would find independence an acceptable male
behavior because it is met by dependence (others carry out
plans successfully).

From this perspective, then, there is support for the
notion that television provides models which are stereotyped
by sex and whose behaviors are differentially reinforced.

The above statement cannot be considered in isolation,
however. It assumes only that children develop a sex role
based on the modeling of same sex behaviors. As mentioned
earlier, other factors must be taken into consideration.

Mischel (1966) states that the "manner in which the
model's behavior is presented, with respect to the frequency,
rate, and clarity of presentation critically affects the
extent to which the modeled behaviors are acquired" (p. 59).

The evidence presented in this study, across all shows and

within program types, tends to support this statement,

112e

especially if clarity of presentation can be interpreted as
stereotyping.

A logical next step in the study of sex roles on tele-
vision, then would be to study how children make use of the
massive amount of information they can receive from television
about their sex role. The type of program the television show
is classified as has been implicated as a factor in the form
of presentation of sex-typed behaviors. Children who favor
Situation Comedies over Action Adventures find different
models available for imitation.

A closer look at the behavior of major and regular
characters is called for. Questions must be asked noteonly
of the individual characters, but also of the children who'
view them. Social learning theory (Bandura, 1971) dictates
questions such as: Are these characters attractive? Do
children perceive them as similar to themselves? Do children
value the activities of these characters? If these questions
can all be answered with a yes, then it would be expected that
heavy viewers of television would perform and value dominance,
succorance, and independence.

This study has made no distinction between highly valued
and lowly valued or status of the models. It could very weel
be, for example. that girls who find dominant female characters
attractive and of high status (e.g., Maude, Rhoda), will come

to value dominance, especially if they are rewarded for this
behavior when they attempt to perform it.

Studies already undertaken in this area show that

112f

children do find televison characters attractive, value their
activities, and identify with those similar to themselves
(e.g., Reeves & Greenberg, 1977). Knowing this, sex-typed
behaviors of television characters may be used to develop

a "portrait" of major characters. Children who identify with
attractive, valued, same-sex, and sex-typed characters may be
more likely to imitate the behaviors of those characters.

If it is found that most characters children identify with
evidence stereotypes, then the behavior repertoires of those
children have, in a sense, been restricted.

Concern is expressed in most content analytic studies
over stereotyped role models. This concern, however, has
most often been with the confinement of women to narrowly
defined roles (e.g., Beuf, 1974; Long & Simon, 1974). There
is evidence in this study, though, that males are stereotyped
on television as well as females. To make matters worse,
studies such as Miller (1976) and Miller & Reeves(l976) show
that girls are more likely to identify with cross-sex models
than boys.

If such is the case, then, concern with stereotyped male
images is long overdue. The argument can be made, however,
that girls have not had attractive same-sex models to identify
with and have, therefore, no choice but to make cross-sex
identifications. No attempt will be made to resolve the
argument here. The point, very simply, is that female
television roles are not the only ones subject to stereotyping;

male television roles also exhibit stereotyping of sex role

112g

behaviors.

The purpose of this study has been to analyze sequences
of behavior and to determine whether these Show evidence of
stereotyping and the ability to act as homogeneous sex role
models. The results of this content analysis provide clear
support that television provides homogeneous sex role models
which are positively reinforced. This homogeneity and
stereotyping will gain more perspective, however, upon analysis
of any trends in these behaviors over the three years of

analysis to be completed.

APPENDICES

113

APPENDIX A

GENERAL INSTRUCTIONS FOR THE USE OF CODING FORMS:

1) The sheets in this notebook have been reduced from

8%" x 14" paper. The larger sheets allow viewers the room
needed to write in character names, descriptions of behavior,
etc.

2) Across the top of each sheet is general identification
information: the tape number of the Show; the show name;

the viewer's name and date viewed (the viewer actually watches
the Show); the coder's name and date coded (coders transfer
the viewed material to computer coding forms for punching);
and how many pages make up the dimension set for that show
and viewer.

3) Column numbers in the following instructions refer to
columns on each dimension viewing sheet. These are only to
facilitate instructions and do not appear on the actual

viewing sheets on which the data are entered.

114

CATEGORY INSTRUCTIONS
Columns 1-4: All Categories

1 - TS: Time Segment. Each show is viewed in segments
Of two minutes. Viewers watch an entire segment
without stopping the tape. At the end of the seg-
ment they go back through, stopping the tape, to
write down any relevant behaviors in the three dimen-
sions. Viewers work on all three dimensions at once.
If nothing occurs during a time segment it is entered
anyway as a check.

2 - Character Name: Whenever an instance of a relevant
behavior occurs the viewer enters the name of the
character needing support, giving an order, or
making a plan.

3-4 — M-F: These are "check" columns. The viewer makes
a check mark in the appropriate column identifying

the gender (male or female) of the character.

SPECIFIC INSTRUCTIONS: ORDER CATEGORY
Columns 5-6: Give Orders
The viewers enter the following codes in Col. 5:

Code. A double code is entered if the order is

explained.

115
Authority = A

T

Threat
S

Simple
Explained = E
A short quote, paraphrase, or action is entered
in 001. 6: Describe, as a check on appropriate
coding in Col. 5 .

Columns 7 - 9: To Whom
These columns are used for entering the name of
the character being ordered; Col. 7, and the
appropriate gender: Col. 8: M; 001. 9: F.
Col. 8 & 9 are check columns.

Columns 10 - 13: Followed?
These columns are used to indicate whether the
order was carried out or obeyed. Col. 10: TS,
indicates in which time segment the order was
followed or not followed. The number of the time
segment in which this occurs is entered in C01. 10.
Col. 11: NF and Col. 12: F, are check columns for
indicating whether the order was followed.
Col. 13: Describe, is a viewer check on coding
and should contain a short quote, paraphrase, or
action indicating how the order was followed/not
followed.

(Solumns 14 - 15: Consequences
Actions or verbal behaviors that happen as a

result of the order being followed or not followed

116

are entered in these columns. Col. 14: TS, should
contain the number of the time segment of the con—
sequence. Col. 15: Describe, is for any action,
quote, or paraphrase which denotes that consequence.
Column 16: Notes

As in the support dimension, this column is for
viewer comments noting any questions or "grey area"
problems the viewer may have with a particular order

instance.

SPECIFIC INSTRUCTIONS: SUPPORT CATEGORY

Column 5: Needs Support
The following letter codes are entered in Col. 5.
As many codes as necessary are entered to describe
the instance of need for support.
Physical Internal = PI
Physical External = PE
Physical Confinement = PC
Inability to Cope = CC
Humiliation = H
Concern for Others = C0
Psycho Support = PS

Columns 6, 7, 8: Asks Support
Col. 6 & 7 are check columns - the viewer makes a
check mark under "N" for no: the person in need of
support doesn't or can't ask for it; or under "Y"

for yes: the person in need of support does ask

117
for it. 001. 8: Describe, is a short quote,
paraphrase, or action identifying how the person
asked for support if "Y" is checked. This is
useful for making spot checks on whether the
viewers are entering the appropriate codes in
Col. 5.
Columns 9 - 14: Response
Col. 9 & 10 are check columns for indicating whether
someone reacted to the indicated need for support.
If Col. 9: N is checked the remaining columns to
the right are to be left blank unless a response occurs
in a later time segment. If Col. 10: Y, is
checked then a character name is to be entered
in C01. 11: Who. Col. 12: M and Col. 13: F,
are check columns indicating the sex of the
respondent. Col. 14: TS (time segment), indi—
cates when the response was made. The number
of the response time segment is entered in Col. 14.
Columns 15 - 17: Support
001. 15: Given and Col. 16: Not given, indicate
whether the response was to assist, Col. 15, or
not assist, Col. 16, the character in need.
These are check columns. A quote, paraphrase,
or action is entered in 001. 17: Describe, to
indicate how the support was given or not given
The categories that have been identified as not

essential to plot development are entered in

118
Col. 17 by the following codes. They are entered
in this column because it is not known until the
support is given that a need existed or was per-
ceived by the character giving support to exist.
Compliment = 0
Approval = A
Disappointment = D
All columns prior to 001. 17 in these instances

are to be left blank.

Columns 18 - 19: Consequences

Column 20:

Any further action or verbal behavior that results
from the support behavior is a consequence. The
time segment of the consequence is entered in
C01. 18: TS. A short description of the conse-
quence is entered in Col. 19: Describe. In many
cases the consequence will be another need for
support, the giving of an order, or the making
of a plan. In these cases the viewers go to the
appropriate sheet and enter the consequence as
a separate instance of the appropriate dimension.
Notes
This column is for viewer comment on any inter-
action. If a viewer feels uneasy about a code
entered or feels the instance of behavior falls
in a "grey area," s/he can make notes about it
in this column. This is particularly useful

during training.

119

SPECIFIC INSTRUCTIONS: PLAN CATEGORY

Column 5:

Column 6:

Makes Plans
The following letter codes are entered in Col. 5
to indicate the type of plan made.
Makes Appointments = A
Housework or Maintenance = H
Social Affairs
Family = SAF
Non-family = SAN
Business Deals = BD
Strategies = ST
Criminal Activities = CAC
Criminal Apprehension = CAP
Construction = 0
Acquisition = ACQ
Rescue = R
Behavior Intent
Personal = BIP
Interaction = BII
Other = O
Other is used only When a plan doesn't fit any
of the other categories. Plans must be explicit,
not implied, by character actions which connote
that a plan was made in order to carry those
actions out.
For self
This check column indicates whether the plans

made apply to the character making them.

120

Column 7 - 9: For Others

Column 10:

If plans are made by a character for other charac-
ters these names are entered in Col. 7: Who.
The gender is identified in Col. 8: M and Col. 9:
F with a check mark. Groups may be entered in
both gender columns and a number entered in
Col. 7 to indicate how many people are in the
group.

Complexity
Col. 10: 1 to 5, indicates how involved a plan
is, and is scored according to the following
codes. This complexity measure was created at
viewer suggestion to allow them to reduce redun-
dant coding of very complex plans.

1

very simple, immediately executed; instructions

implicit in the plan; e.g., I'll get the mop

to clean up that mess.

2 = simple, executed within the same time seg-
ment; very simple set of instructions: e.g.,
Officer Smith, you cover the back door; Fred
and I will break in through the front and
catch the burglar.

3 = moderate, instruction necessary for execution

which takes place within one or two time

segments. It advances the story line but is

not a major contribution to the plot; e.g.,

Plans for a surprise party which require

moving and hiding people, coordinating lights,

etc.

121

4 = complex, large set of instructions; carries
across more than two time segments. Some of
the plot will revolve around the plan and
execution; e.g., a bank robbery plan that
requires time, people, blue prints, etc.

5 = very complex, requires a major portion of
the show to plan and execute. Most of the
plot will revolve around the plan and its
execution, e.g., a Mission Impossible plot.

Columns 11 - 14: Carries Out

The name(s) of the persons(s) who put the plan

into action is entered in Col. 11: Who. Gender

is identified in check Col. 12: M and Col. 13:

F. The time segment the plan is put into action

is entered in 001. 14: TS.

Columns 15 - 16: Execution

These columns were included at viewer suggestion

to handle those cases when it is apparent that

the plan was put into action but the execution
was not shown. Col. 15: Seen and Col. 16: Not

Seen, are check columns used to indicate whether

the execution was shown.

Columns 17 - 22: Consequences

Consequences in this dimension are the success

or failure of the plan. If the plan worked then

the time segment in which the successful outcome

occurred is entered in 001. 17: TS, and Col. 18:

S is checked. A quote, paraphrase, or action is

122

entered in Col. 19: Describe, to indicate how
or why the plan succeeded. Col. 20 - 22 are
handled in the same manner for a failure (Col.
21: F, is checked to indicate a failure).

In some shows one plan will succeed at first and
later fail, or vice versa. A plan may also
succeed or fail repeatedly during a show without
reformulation of the plan. These instances will
Show up in the consequences columns. As many
successes or failures associated with a single

plan as occur should be entered in these columns.

123

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

14'..- 1 11
Ill I,
on mu 3 5 m.— 5 3 1 5 S :0. m N. m N H
0056
05.5.8 3 05.8.8 as: 0003 .2; 0‘. u 2 05.800. r fin so" at
.30: 0003.60.80 9333 0380.. 9.3926 8.3
”and. «a "00-6 "90000 3 gr
ragga _ .cwom “as 32“ .

 

 

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S S S 2 we: a. o s a u N.
A .
2:88: . as 58.8 a 2. an a I... 290.8 88 8
685238 . 38:8 02.. on .088 2.8 0800.0
go "00.0; no .38 . 30000 is!» 3 an:
30’s “:8 30.8:

 

 

 

 

 

 

125

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

an i o, 2 S S 3 5 5 2S 2 S . a_ s. a n a
03.308 5 mp 05.08.. a fill ..
. MP 00. g ”h .— I g n .00 d 2H 1‘ :00 3‘ '.
ccockionsv gnu-«0.8" USO .35 balOdg ‘28 E E '3 i
«.3. 388 no .88 838 ”g 8 as:
L 00-0; A 38 30:0;

 

126

APPENDIX B
VIEWER TRAINING PACKETS

The following code and definition sheets can be put
together to form viewer packets for taining and actual data
collection. If used during training the packets become
handy references for the viewers during data collection.

It is best for each viewer to have his/her own packet in
order to make notes during training and data collection.
The packet's main function is to provide an anchor for the
viewers during training when the scheme is unfamiliar and

confusing.

127

ORDERS-CODES

Gives Orders:

Enter double code if explained

Authority

Threat

n
mar—33>

Simple

Explained

Followed?:

u
'11

Followed

NF

Not Followed

128

SUPPORT-CODES

Needs Support:

Enter as many codes as necessary

Physical Internal =

Physical External

Physical Confinement
(involuntary)

Inability to Cope
(self-inadequacy)

Humiliation =

Concern for Others

Psycho Support

Gives Support:

PI
PE

PC

CC

CO
PS

Enter in "describe" when applicable

Compliments =

Approval

Disappointment =

C
A

129

PLANS-CODES

Statement of Methods (makes plans):

Makes Appointments = A
Housework or Maintenance = H
Social Affairs
Family = SAF
Non-Family = SAN
Business Deals - BD
Strategies = ST
Criminal Activities = CAC
Criminal Apprehension = CAP
Construction = C
Acquisition = ACQ
Rescue = R
Behavior Intent
Personal = BIP
Interaction = BII
Other = 0
Complexity: Consequences:
Enter Number Success
Very Simple = 1 Failure
Simple = 2
Moderate = 3
Complex = 4
Very Complex = 5

130

Definitions for Sex Role Content Analysis

ORDERS: Directive for others to do, say or think something.

Types of Orders
Threat: an order given with a statement that physical

harm will be done if it is not complied with.

Authority: an order to be complied with because of
occupational position (boss), social agent (police-
man, nurse, doctor), or parent. In some cases by

explicit delegate of the above.

Simple: an order given among equals or peers: e.g.,
husband and wife, brother/sister, friends, co-

workers, etc., unless clearly one of the above.

Explanation: a justification for why an order should be

followed other than threat or authority.

131

Definitions for Sex Role Content Analysis

NEEDS SUPPORT: occurs when person is in danger or distress.
Does not cover routine request for assistance or social
courtesies.

Types of Need for Support
Physical External: person is in danger of being killed,
injured, or beaten

Examples:
person is about to be shot, knifed, etc.

person is in danger of being caught in cave-in, land-
slide, etc.

person is being chased by potential assailant

Physical Internal: person is suffering from disease,
illness or internal malady.

Examples:

person has cancer

person has hepatitis

android has malfunctioning circuits

Physical Confinement: person is jailed, trapped or
held against their will.

Examples:

robbers lock persons in bank vault

person is jailed

person is trapped in mine cave-in

Inability to Qgpg: person states that s/he has a

problem that s/he cannot solve; is in need of
ego support

Examples:

person can't get along with boss, parents, spouse,
etc.

person needs money

person has lost something and can't figure out how
to find it

132

NEEDS SUPPORT-2

Humiliation: person states fear that they will be dis-
liked or held in low esteem by others.

Examples:

person feels that others will think s/he is dumb, irres—
ponsible, or funny looking

person fears that someone will reveal that s/he is
homosexual, have a criminal record, etc.

person fears that others will make fun of s/he

Concern for Others: person discusses help for friend,
relative or associate with a third person. NOTE:
at least three people are involved: the person
expressing concern, the person to whom concern is
expressed, and the person in trouble.

person needs support because someone else is in trouble.
Examples:

person notes that someone is late and expresses worry
that s/he is lost

person asks ideas to help a friend who is depressed

person seeks assistance in rescuing someone who is
trapped or captured by others

Psycho Support: person has a problem because of the
actions of others but does not express inability
to cope, fear of humiliation, or concern for others

Examples:

person's spouse has left them

person's son/daughter has flunked out of school
person's dog is causing trouble in the neighborhood

The following categories are occurrences that are not essential
to plot development; support is easily given; needs are momen-
tary and easily resolved.

Compliments: person receives praise, compliment or encour-
agement when not requested or particularly needed.

Approval: person seeks confirmation of ideas, opinions, or
actions.

Disappointment: person is visibly upset, depressed, or blue
because efforts are unsuccessful: a temporary emotional
setback.

APPENDIX C

1975—76 SAMPLE OF TELEVISION PROGRAMS

Name of Show

All in the Family
Barbary Coast
Baretta

Barnaby Jones
Barney Miller
Beacon Hill
Bionic Woman*
Bob Newhart
Bronk

Bugs Bunny
Cannon

Chico and the Man
Doc

Doctors Hospital
Ellery Queen
Emergency
Emergency Plus 4
Family Holvak
Fat Albert

Fay

Good Times

Ghost Busters
Happy Days

Harry 0

Hawaii Five-O
Hong Kong Phooey
Isis

Invisible Man
Jeffersons*

Joe and Sons

Joe Forrester

Top 40
(x)

X

NNXXX

Type

Sitcom
ActAdv
ActAdv
ActAdv
Sitcom
Medfam
ActAdv
Sitcom
ActAdv

Cartoon

ActAdv
Sitcom
Sitcom
Medfam
ActAdv
Medfam

Cartoon

Medfam

Cartoon

Sitcom
Sitcom

Noncart

Sitcom
ActAdv
ActAdv

Cartoon
Noncart

ActAdv
Sitcom
Sitcom
ActAdv

Time

H

I I
H \o H
H

I
H
H

Ill
H\0H\o
H H

\O\OCI)\OCD\O\OCD\O
I I

H

5.1:

Sat.
9-11
8-9
8-9
9-11
8-9
8-9
Sat.
8-9
Sat.
8-9

Sat.
8-9
9-11
9-11
Sat.
Sat.
8-9
8-9
8-9
9-11

13#

Name of Show

Josie and the Pussy Cats
Kate McShane

Kojak

Land of the Lost

Laverne and Shirley
Little House on the Prairie
Lost Saucer

Marcus Welby

Mary Tyler Moore

M*A*S*H

Matt Helm

Maude

Medical Center

Medical Story

Mobile One

Movin' On

New Adventures of Gilligan
Oddball Couple

On the Rocks

One Day at a Time*
Pebbles and Bamm-Bamm
Phyllis

Pink Panther

Police Woman

Top 40

NNNN

N

N

Return to the Planet of the Apes

Rhoda

Rockford Files

Rookies

Run, Joe, Run

Sanford and Son

Scooby Doo, Where Are You
Secret Lives of Waldo Kitty
Shazam

Sigmund and the Sea Monsters

>4><><>4><><

Type

Cartoon
ActAdv
ActAdv
Noncart
Sitcom
Medfam
Noncart
Medfam
Sitcom
Sitcom
ActAdv
Sitcom
Medfam
Medfam
ActAdv
ActAdv
Cartoon
Cartoon
Sitcom
Sitcom
Cartoon
Sitcom
Cartoon
ActAdv
Cartoon
Sitcom
ActAdv
ActAdv
Noncart
Sitcom
Cartoon
Cartoon
Noncart
Noncart

Time

Sat.
9-11
9-11
Sat.
8-9

8-9

Sat.
9-11
9-11
8-9

9-11
9-11
9-11
9-11
8-9

8-9

Sat.
Sat.
8-9

9-11
Sat.
8-9

Sat.
9-11
Sat.
8-9

9-11
9-11
Sat.
8-9

Sat.
Sat.
Sat.
Sat.

Name of Show

Six Million Dollar Man
Space 1999

Speed Buggy

Starsky and Hutch
Streets of San Francisco
Swiss Family Robinson
Switch

That's My Mama

Three for the Road

Tom and Jerry/Grape Ape
Valley of the Dinosaurs
Waltons

Welcome Back Kotter
When Things Were Rotten

*Midseason Replacements

135

Top 40
(X)

Type

ActAdv
ActAdv
Cartoon
ActAdv
ActAdv
Medfam
ActAdv
Sitcom
Medfam
Cartoon
Cartoon
Medfam
Sitcom
Sitcom

Time

8-9
8-9
Sat.
9-11
9-11
8-9
9-11
8-9
8-9
Sat.
Sat.
8-9
8-9
8-9

Top #0 = Shows viewed regularly by 40% or more of a sample

of 4th, 6th, and 8th graders in Fall, 1975.
Sitcom = Situation Comedy

ActAdv
Medfam

Cartoon: Animated Cartoon

Noncart: Non-animated Cartoon
8’9 : 8-9 pom.
9-11 = 9-11 p.m.

Action Adventure

Medical/Family Drama

Sat. = Saturday morning (8 a.m.-12 p.m.)

136

APPENDIX D

t-values and standard deviations

Table D1
t-values and standard dev1ations: Orders Category.

All Shows (Table 2)

 

 

 

 

 

Order Type
t—value standard deviations
females males
AUthOFity '6078 1012 2083
Simple ”6043 2023 2067
Authority explained -4.79 0.34 0.80
Simple explained -6.73 0.50 0.91
ALL ORDER TYPES -10.06 2.93 4.48
Receivers
t—value standard deviations
females males
Female receivers —4.25 1.41 1.92
Male receivers -9.11 2.26 3.71
Orders Followed
t-value standard deviations
females males
FOllOWGd ”10025 2001 3063

Not followed -6.27 1.27 1.59

 

Program Type (Table 3)

Situation Comedies

Authority

Simple

Authority explained
Simple explained
ALL ORDER TYPES

Authority

Simple

Authority explained
Simple explained
ALL ORDER TYPES

Authority

Simple

Authority explained
Simple explained
ALL ORDER TYPES

137

Table D2
t-values and standard deviations: Order Types by

 

t-values standard deviations
females males

‘0088 1060 2027
+1.74 3.16 1.89
-1084 003,4 1002
+0.38 0.82 1.01
+0.65 3.86 3.87

 

Action Adventures

 

t-values standard deviations
females males

'3000 1054 3025

-1.30 2.75 2.94

-0.74 0.77 0.83

+0.57 0.56 0.76

-2.51 3.64 4.87

 

Medfam Dramas

 

t-values standard deviations
females males

“0055 1039 3016
+0.27 2.70 2.54
+0.54 0.50 0.41
-2.50 0.48 0.74
-0048 3.82 4025

 

138
Table D3
t-values and standard deviations: Orders Receivers

and Followed by Program Type (Table 4)

Situation Comedies

 

t-values standard deviations

females males

Female receivers +1.39 3.39 2.34
Male receivers —0.44 2.68 3.09
Not followed +1.39 2.08 1.36

 

Action Adventures

 

t-values standard deviations

females males
Female receivers -2.28 0.79 1.68
Male receivers -2.26 3.17 4.12
Followed —2.46 3.11 3.83
Not followed -1.50 1.36 1.68

 

Medfam Dramas

 

t-values standard deviations

females males

Female receivers -0.63 1.14 1.42
Male receivers -O.12 2.46 2.55
Followed -1.24 2.36 3.11
Not followed +0.82 1.54 1.17

 

139
Table D4
t-values and standard deviations: Order Types by Broadcast

Time (Table 5)

Saturday Morning

 

 

 

 

 

t-values standard deviations
females males
Authority -3.55 0.44 2.21
Simple +0.75 3.58 2.74
Authority explained -2.19 0.20 0.77
Simple explained -0.99 0.82 1.10
ALL ORDER TYPES -0.72 3.93 4.00
8’9 pom.
t-values standard deviations
females males
Authority -1.13 2.11 2.47
Simple +0.85 3.57 2.42
Authority explained -0.47 0.70 0.81
Simple explained -0.08 0.69 0.75
ALL ORDER TYPES -0.01 4.19 3.85
9-11 p.m.
t-values standard deviations
females males
Authority -5.06 0.82 3.54
Simple '6060 1018 2087
Authority explained -3.94 0.20 0.82
Simple explained -4.63 0.35 0.91
ALL ORDER TYPES -8.10 1.90 5.46

 

140

Table D5
t-values and standard deviations:

Followed by Broadcast Time (Table 6)

Female receivers
Male receivers

Followed
Not followed

Female receivers
Male receivers

Followed
Not followed

Female receivers
Male receivers

Followed
Not followed

Saturday Morning

Orders Receivers and

 

 

 

 

 

t—values standard deviations
females males
-1.27 0.64 1.58
-0-55 3-94 3-56
-o.97 3-03 3-34
-O.97 1-31 1-39
8—9 p.m.
t-values standard deviations
females males
+0.94 2.31 1-59
-o.69 3.60 3.22
-0.73 2.82 3.14
+0.92 2.13 1.42
9-11 p.m.

t-values standard deviations
females males

-4.98 1.17 2.41
-7.01 1.19 4.33
-7.75 1.28 4.34
-6.06 0.85 1.91

 

141
Table D6
t-values and standard deviations: Order Types by Viewing

Preference (Table 7)

Preferred Shows

 

t-values standard deviations
females males

Authority -1.82 1.61 2.90
Simple +1.39 3.24 2.37
Authority explained -1.56 0.42 0.89
Simple explained -1.03 0.77 0.95
ALL ORDER TYPES +0.26 3.82 4.34

 

Non-Preferred Shows

 

t-values standard deviations
females males

Authority -6.36 0.97 2.76
Simple -7.69 1.74 2.90
Authority explained -4.03 0.31 0.72
Simple explained -5.92 0.40 0.88

 

142
Table D7

t-values and standard deviations: Orders Receivers and

Followed by Viewing Preference (Table 8)

Preferred Shows

 

t—values standard deviations
females males
Females receivers +1.04 2.76 1.87
Male receivers -1.12 3.01 3.60
Not followed +1.01 1.84 1.48

 

Non—Preferred Shows

 

t-values standard deviations
females males
Female receivers -5.71 0.21 1.03
Male receivers -9.06 0.63 3.26
Followed -10.10 0.58 3.37
Not followed -7.06 0.33 1.26

 

143
Table D8
t-values and standard deviations: Support Category,

All Shows (Table 9)

ALL SUPPORT TYPES

 

 

 

 

 

t—values standard deviations
females males
ALL SUPPORT TYPES -4.64 3.31 3.64
Physical Support Types
t-values standard deviations
females males
Physical Internal —5.19 0.66 1.26
Physical External -6.98 0.72 1.40
Physical Confinement -4.12 0.40 0.68
ALL PHYSICAL TYPES -8.73 1.22 2.14
Emotional Support Types
t-values standard deviations
females males
Ego Support -O.15 1.43 1.64
Concern for Others +1.06 0.93 0.68
Psycho Support -1.50 1.38 1.50
ALL EMOTIONAL SUPPORT -O.54 2.82 2.64

 

Support Asked

 

t-values standard deviations
females males

Asked -O.92 2.01 1.81
NOt ASKGd -7006 1027 2018

 

Support Responded To

 

t-values standard deviations
females males
Response .203? 2047 2049

No Response -7.43 0.68 1.43

 

Given
Not Given

Female Respondents
Male Respondents

144

Table D8 (cont.)
Support Given

 

 

 

t—values standard deviations
females males

-2.26 1.94 1-93
-6006 1024 1078

Respondents

t-values standard deviations
females males

-0066 0091 0095
-2017 2002 2012

 

145

Table D9
t—values and standard deviations:

by Program Type (Table 10)

Situation Comedies

Physical Internal
Physical External
Physical Confinement
All Physical Types

Action Adventures

Physical Internal
Physical External
Physical Confinement
All Physical Types

Physical Internal
Physical External
Physical Confinement
All Physical Types

Physical Support Types

 

t-values standard deviations
females males

—2.03 0.16 0.93
—2056 0039 1013
-2018 0000 0033
-3044 0041 1060

 

 

t-values standard deviations
females males

+0.34 0.98 1.14

-1.96 0.93 1.15

-2.66 0.30 0.83

'1081 1038 2013

 

Medfam Dramas

 

t-values standard deviations
females males
-1089 0059 1031
-O-49 0-79 0-99
+0.10 0.46 0.40
”luau 1019 1086

 

146
Table D10
t-Values and standard deviations: Emotional Support Types

by Program Types (Table 11)

Situation Comedies

 

t-values standard deviations
females males

Ego Support +0.57 1.81 2.93
Concern for Others +1.81 1.22 0.74
Psycho Support +0.14 2.44 2.54
ALL EMOTIONAL TYPES +0.86 4.03 4.50

 

Action Adventures

 

t-values standard deviations

females males

Ego Support +2.69 1.98 0.82
Concern for Others +1.48 1.26 0.78
Psycho Support +1.20 1.17 1.07
ALL EMOTIONAL TYPES +3.00 2.86 1.69

 

Medfam Dramas

 

t-values standard deviations

females males

Ego Support +1.80 1.80 1.48
Concern for Others +1.47 1.41 0.83
Psycho Support +1.85 1.71 1.22
ALL EMTOIONAL TYPES +2.57 3.38 2.21

 

Responded to by Program Type (Table 12)

Asked
Not Asked

Response
No Response

Asked
Not Asked

Response
No Response

Asked
Not Asked

Response
No Response

147

Table D11

t—values and standard deviations:

Situation Comedies

Support Asked and

 

t-values standard deviations
females males
+1.20 2.42 2.32
-0.93 1.61 2.67
+0.79 3.04 3'67
"2075 0'37 0'91

 

Action Adventures

 

t-values standard deviations
females males

+2.25 1.73 1.55

+0.21 1.44 1.84

+2.27 2.28 2.17

-1.04 0.92 1.09

 

Medfam Dramas

 

t-values standard deviations
females males

+1.91 2.97 1-99

+0.05 1.48 1.45

+1.91 3.26 2.46

-O.62 0.54 0.83

 

148

Table D12
t-values and standard deviations: Support Given and Respondents

by Program Type (Table 13)

Situation Comedies

 

t-value standard deviations
females males
Given +1.31 2.79 2.66
Not Given -1.77 1.06 2.03
Female Respondents +3.03 1.21 0.47
Male Respondents -O.32 1.79 1.97

 

Action Adventures

 

t-values standard deviations
females males

Given +2.09 1.58 1.66
Not Given +0.56 1.44 1.66
Female Respondents +0.66 0.82 0.86
Male Respondents +2.32 2.28 1.71

 

Medfam Dramas

 

t-values standard deviations

females males

Given +2.11 2.52 2.06
Not Given +0.66 1.51 1.13
Female Respondents +0.94 1.40 0.94
Male Respondents +1.82 2.70 1.82

 

149
Table D13
t-values and standard deviations: Physical Support Types
by Broadcast Time (Table 14)

Saturday Morning

 

 

 

 

 

t-values standard deviations
females males
Physical Internal -O.62 1.24 1.47
Physical External -0.96 1.35 1.84
Physical Confinement +1.09 0.92 0.71
ALL PHYSICAL TYPES -O.6O 2.03 2.32
8-9 pom.
t-values standard deviations
females males
Physical Internal -1.12 0.72 0.88
Physical External —1.58 0.83 1.18
Physical Confinement -2.43 0.31 0.71
ALL PHYSICAL TYPES -2.45 1.22 1.82
9-11 p.m.
t-values standard deviations
females males
Physical Internal -3.67 0.44 1.34
Physical External -6.62 0.41 1.08
Physical Confinement -3.09 0.17 0.59
ALL PHYSICAL TYPES -6.54 0.73 2.13

 

150
Table D14
t—values and standard deviations: Emotional Support Types

by Broadcast Time (Table 15)

Saturday Morning

 

 

 

t-values standard deviations
females males
Ego Support +0.54 1.13 1.11
Concern for Others +1.42 0.71 0.29
Psycho Support +0.27 1.03 0.98
ALL EMOTIONAL TYPES +0.98 1.87 1.74
8—9 p.m.
t—values standard deviations
females males
Ego Support +0.98 1.56 2.34
Concern for Others +1.30 1.26 0.78
Psycho Support -O.57 1.35 2.01
ALL EMOTIONAL TYPES +0.76 2.92 3.64
9‘11 pom-
t-values standard deviations
females males
Ego Support +0.28 1.38 1.00
Concern for Others -O.49 0.82 0.79
Psycho Support -1.09 1.42 1.17
ALL EMOTIONAL TYPES +2.57 3.38 2.22

 

 

 

151
Table D15
t-values and standard deviations: Support Asked and Responded

to by Broadcast Time (Table 16)

Saturday Morning

 

 

 

 

 

t-values standard deviations
females males
Asked +2.34 1.86 1.53
Not Asked -2.32 1.45 2.32
Response -1091 1092 1081
No Response +1.72 1.29 1.96
8-9 pom.
t-values standard deviations
females males
Asked +0.79 2.40 1.92
Response +0.70 2.65 3.06
No Response -2.14 0.68 0.90
9-11 p.m.
t-values standard deviations
females males
Response -2.93 2.34 2.34
No Response -5.40 0.45 1.11

 

152

Table D16
t-values and standard deviations: Support Given and Responded

to by Broadcast Time (Table 17)

Saturday Morning

t-values standard deviations
females males
Given -0.05 1.63 1.60
Not Given +0.30 1.35 2.13
Female Respondents -O.74 0.66 1.07
Male Respondents +1.37 2.01 1.37
8—9 p.m.
t-values standard deviations
females males
Given +1.32 2.19 2.32
Not Given -1.64 1.24 1.61
Female Respondents +1.69 1.25 0.86
Male Respondents +0.32 2.26 2.82
9-11 pom.
t-values standard deviations
females males
Not Given -5.17 1.13 1.55
Female Respondents —1.52 2.79 0.92
Male Respondents -2.85 1.84 1.77

 

 

 

 

 

 

153
Table D17
t-values and standard deviations: Support Types by Viewing

Preference (Table 18)

Preferred Shows

 

 

t-values standard deviations
females males
Physical Internal -4.64 0.63 1.42
Physical External -2.32 1.05 1.72
Physical Confinement -2.59 0.38 0.73
ALL PHYSICAL TYPES -4.76 1.40 2.53
Ego Support +2.53 1.96 1.87
Concern for Others +2.74 1.48 0.71
Psycho Support +2.23 2.15 1.13
ALL EMOTIONAL TYPES +3.59 3.90 2.43

 

Non-Preferred Shows

 

 

t-values standard deviations

females males

Physical Internal -2.44 0.67 1.05
Physical External -6.74 0.51 0.94
Physical Confinement -2.85 0.41 0.62
ALL PHYSICAL TYPES -6.41 1.13 1.57
Ego Support -1.24 1.03 1.35
Concern for Others -O.78 0.54 0.65
Psycho Support -3.71 0.88 1.78
1.88 2.84

ALL EMOTIONAL TYPES -2.98

 

154

Table D18
t-values and standard deviations:

Support Asked, Responded

to, Given and Respondents by Viewing Preference

Asked
Not Asked

Response
No Response

Given
Not Given

Female Respondents
Male Respondents

Asked
Not Asked

Response
No Response

Given
Not Given

Female Respondents
Male Respondents

(Table 19)

Preferred Shows

 

 

 

 

t-values standard deviations
females males
+3.11 2.43 1.70
‘2018 1075 2'53
+2.51 3.03 2-50
-3.68 0.94 1-74
+2.48 2.61 2.01
-1.70 1.53 2.11
+2.04 1.22 1.06
+2.29 2.75 1.98

 

Non-Preferred Shows

 

t-values standard deviations
females males

-2.98 1.68 1.92
-6.08 0.96 1.69
-4.05 1.96 2.48
-5-95 0-55 0-97
-4.40 1.41 1.86
-5.54 1.06 1.31
-2.13 0.73 0.81
-4.00 1.50 2.26

 

t-values and standard deviations:

Makes Plans

For Females
For Males

Female Executors
Male Executors

Success
Failure

155

 

Table D19
Plans Category, All Shows
(Table 20)
t-values standard deviations
females males
-1.42 0.92 1.00

 

Plans for Others

 

 

 

 

 

t-values standard deviations
females males
-1.01 0.35 0.53
-1.03 0.75 0-93
Executors
t-values standard deviations
females males
+4.52 0.67 0.20
”6017 0068 0092
Outcomes
t-values standard deviations
females males
-1011 0087 0090
‘OoLl'O 0092 0086

 

156
Table D20
t-values and standard deviations:

by Program Type (Table 21)

Situation Comedies

Makes Plans and For Others

 

t-values standard deviations
females males

Makes Plans -2.38 0.24 0.62
For Females -O.72 0.33 0.57
For Males -1.42 0.49 0.55

 

Action Adventure

 

t—values standard deviations

females males

Makes Plans -0070 1013 1029
For Females +0.85 0.54 0.55

 

Medfam Dramas

 

t-values standard deviations
females males

Makes Plans +2.07 0.97 0.67
For Females -1.20 0.32 0.72
For Males +1.52 0.63 0.51

 

157
Table D21
t-values and standard deviations: Plan Executors and Outcomes

by Program Types (Table 22)

Situation Comedies

 

t-values standard deviations
females males

Female Executors +3.28 0.62 0.16
Male Executors -6.17 0.33 0.54
Failure -1.13 0.44 0.49

 

Action Adventures

 

t-values standard deviations
females males

Female Executors +1.72 0.79 0.23
Male Executors -3.04 0.76 1.08
Success -1.30 0.79 1.02
Failure +0.55 0.69 0.91

 

Medfam Dramas

 

t-values standard deviations
females males

Female Executors +1.76 0.71 0.29
Male Executors -2.75 0.48 0.75
Success -O.46 1.23 0.87
Failure +0.04 0.70 0.59

 

t-values and standard deviations:

158
Table D22

Others by Broadcast Time (Table 23)

Makes Plans

For Females
For Males

Makes Plans

For Females
For Males

Makes Plans

For Females
For Males

Saturday Morning

Makes Plans and For

 

 

 

 

 

t-values standard deviations
females males
+0.16 1.29 0.96
“2.20 0.00 0038
+0.29 0.88 1.02
8-9 p.m.
t-values standard deviations
females males
-0.82 0.77 0-74
-OOL1’O 0043 0063
+0.53 0.59 0.67
9'11 pom.
t-values standard deviations
females males
'1060 0075 1026
-0.76 0.34 0-54
-2.14 0082 1007

 

159

Table D23
t-values and standard deviations: Plan Executors and
Outcomes by Broadcast Time (Table 24)

Saturday Morning

 

 

 

 

 

t-values standard deviations

females males

Female Executors +1.72 0.70 0.14
Lale Executors —1.24 1.00 1.01
Success +0.47 0.94 0.84
Failure +0.33 1.56 1.05

8-9 panic

t-values standard deviations

females males

Female Executors +4.0' 0.69 0.13
Male Executors -6.39 0.38 0.72
Success -O.11 1.00 0.88
Failure +0.14 0.62 0.65

9-11 p.m.

t-values standard deviations

females males

Female Executors +1.74 0.62 0.30
Hale Executors —4.32 0.58 1.05
Success —2.59 0.63 0.97
Failure -1.24 0.02 0.84

 

t-values and

makes Plans

For Females
For Males

flakes Plans

Standard deviations:

by Viewing Preference

160
Table D24

Preferred Shows

Makes Plans and For Others

(Table 25)

 

t-values standard deviations'
females males
-0.67 1.08 1.10
-1.52 0.30 0.63
-O.14 0.78 1.03

 

Non-Preferred Shows

 

t-values standard deviations
females males
—1.33 0.77 0.86
+0.13 0.39 0.37
-1045 0072 0080

 

161
Table D25

t-values and standard deviations: Plan Executors and
Outcomes by Viewing Preference (Table 26)

Preferred Shows

 

 

t-values standard deviations
females males
Female Executors +2.82 0.68 0.10
Male Executors -4.14 0.72 0.90
Success —O.61 1.04 0.95
Failure +0.02 1.15 0.84

 

Non-Preferred Shows

 

t—values standard deviations

females males
Female Executors +3.42 0.66 0.28
Male Executors -4.58 0.65 0.88
Success -O.96 0.70 0.83
Failure -O.6O 0.67 0.88

 

REFERENCES

162
References

Bandura, A.
1965 Influence of Model's Reinforcement Contingen-
cies on the Acquisition of Imitative Responses.
Journal of Personality and Social Psychology.

1 9 589'595 '

Bandura, A.
1971 Social Learning Theory. Morristown, N.J.:
General Learning Press.

Bandura, A., J.E. Grusec, F.L. Menlove
1966 Observational Learning as a Function of
Symbolization and Incentive Set. Child
Development. 5, 449-455.

Bandura, A., D. Ross, S.A. Ross
1963 Imitation of Film-mediated Aggressive Models.
Journal of Abnormal and Social Psychology.

66, 3-11.

Beuf, A.

1974 Doctor, Lawyer, Household Drudge. Journal
of Communication. 24:2, 142-145.

Bowes, J.E.
1977 Stereotyping and Communication Accuracy.
Journalism Quarterly. 54:1, 70-76.

Broverman, J.K., S.R. Vogel, D.M. Broverman, F.E. Clark,
P.S. Rosenkrantz

1972 Sex-role Stereotypes: A Current Appraisal.
Journal of Social Issues. 28: 59-78.

Busby, L.J.
1974 Defining the Sex Role Standard in Commercial
Network Television Programs Directed Toward
Children. Journalism Quarterly. 51, 690—696.

Busby, L.J.

1975 Sex-role Research on the Mass Media. Journal
of Communication. 25: 4, 107-131.

Carter, R.F.

1962 Stereotyping as a Process. Public Opinion
Quarterly. 26, 77-91.

Chase, D.J.
1972 Sexism in Textbooks. Nation's Schools.
Deon, 31-35

163

Courtney, A.E., S. W. Lockeretz
1971 Woman's Place: An Analysis of the Roles
Portrayed by Women in Magazine Advertisements.
Journal of Marketing Research. 8:Feb., 92-95

Dominick, J.R., G.E. Rauch
1972 The Image of Women in Network TV Commercials.
Journal of Broadcasting. 16:3, 259-265.

Donelson, E.
1975 Sex Differences in Developmental Perspective.
Homewood Ill.: Learning Systems Company

Frueh, T., P.E. McGhee
1975 Traditional Sex Role Development and Amount
of Time Spent Watching Television. Develop-
mental Psychology. 11:1, 109.

Gage, N.L. '
1952 Judging Interests from Expressive Behavior.
Psychological Monographs. 66:18, Whole #320

Gerbner, G., L. Gross
1976 Living with Television: The Violence Profile.
Journal of Communication. 26:2, 173-199.

Greenberg, B.S., C.K. Atkin, N.G. Edison, F. Korzenny,
G.R. Heald, J.J. Wakshlag
1977 Pro-social and Anti—social Behaviors on Com-
mercial Television in 1975-76. Project CASTLE:
Report to the Office of Child Development,

I-IEWO #10
Hays, W.L.
1963 Statistics. New York: Holt, Rinehart, &
Winston.

Hennessee, J.A., J. Nicholson
1972 NOW Says: TV Commercials Insult Women. New
York Times Magazine. May 28, 12-13+.

Katzman, N.
1972 Television Soap Aperas: What's Been Going On
Anyway? Public Opinion Quarterly. 36:2,
200-212

Long, M., R.J. Simon
1974 The Roles and Statuses of Women on Children

and Family TV Programs. Journalism Quarterly.
51, 107-110 '

Maccoby, E.E., C.N. Jacklin
1974 The Psychology of Sex Differences. Stanford,
Calif.: Stanford University Press.

164

McNeil, J.C.
1975 Feminism, Femininity, and The Television Series:
A Content Analysis. Journal of Broadcasting.

19. 259-269-

Miller, M.M., B. Reeves
1976 Dramatic TV Content and Children's Sex-role
Stereotypes. Journal of Broadcasting. 20:1,

35-50-

Mischel, W.

1966 A Social Learning View of Sex Differences in
Behavior. in Maccoby, E.E. (ed.), The Develop-
ment of Sex Differences. Stanford, Calif.:
Stanford University Press. 56-81.

Murray, H.A.
1938 Explorations in Personality. New York:
Oxford University Press.

Nilsen, A.P.
1971 Women in Children's Literature. College
English. May,918-926.

Reeves, B., B.S. Greenberg
1977 Children's Perceptions of Television Characters.
Human Communication Research. 3:2, 113-127.

Sears, R.R., L. Rau, R. Alpert
1965 Identification and Child Rearing. Stanford,
Calif.: Stanford University Press

Seggar, J.F.
1975 Imagery of Women in Television Drama. Journal
of Broadcasting. 19, 273-281.

Simmons, K.W., B.S. Greenberg, C.K. Atkin, C.J. Heeter
1977 The Demography of Fictional Television Charac-
ters in 1975-76. Project CASTLE: Report to
the Office of Child Development. HEW. #2

Sternglanz, S.H., L.A. Serbin
1974 Sex Role Stereotyping in Children's Television
Programs. Developmental Psychology. 10:5,

710-715

Streicher, H.W.
1974 The Girls in the Cartoons. Journal of Com-
munication. 24:2, 125-129.

Sutton—Smith, B., and M. Savasta
1972 Sex Differences in Play & Power. Paper presented
at the Annual Meeting of the Eastern Psychological
Association, Boston.

165

Tedesco, N.

1974 Patterns in Prime Time. Journal of Commun-
ication. 24:2, 119-124.

Turow, J.
1974 Advising and Ordering: Daytime, Prime Time.
Journal of Communication. 24:2, 138-141.