Content-Type: text/html Prime Time Stereotyping on the New Television Networks B. Carol Eaton Syracuse University 743 Maryland Avenue Syracuse, New York 13210 [log in to unmask] (315) 423-2965 Abstract This content analysis examines gender portrayals in a sample of prime time promotional announcements broadcast on five television networks (ABC, NBC, CBS, Fox and UPN) during one week in 1995. Findings supported the hypothesis that stereotypical portrayals of women varied due to the television network's target audience. Specifically, television programs on networks that appeal to a younger male audience contained more stereotypical female characters than other network programming produced for a more general audience. Prime Time Stereotyping on the New Television Networks Television portrayals of women have been the focus of both popular and social scientific inquiry for decades. A great deal of social science research has been generated to examine the issue of female stereotyping in prime time network television programs and advertising. This research has typically described a male-dominated world with females delegated to stereotypical, minor roles. Although portrayals of women on the three traditional commercial broadcast networks (ABC, CBS, and NBC) may have become less stereotypical over the past decade, with shows like Murphy Brown and Designing Women, the launch of three new television networks in this decade has complicated the issue. These new networks must compete with the established networks for viewers and advertising dollars. In order to accomplish this formidable task, the Fox, United Paramount, and Warner Brothers networks developed a competitive start-up strategy to target younger viewers than the three veteran networks, and are specifically interested in a young male audience (Coe, 1995). This content analysis examines stereotypical portrayals of women in prime time television programs on each of these networks, testing the following hypotheses: 1. On all networks, female characters will not appear as frequently as male characters. 2. Television programs on networks that target a younger male audience will contain more stereotypical female characters than other network programming produced for a more general audience. Theory Uses and Gratifications Theory Commercial television networks sell advertising time based upon the size of their viewing audiences. The revenues or advertising dollars generated for each program, therefore, depend on the program's viewing audience size as reported by A. C. Nielsen ratings data. Networks try to maximize audience share in the desired target market in order to increase advertising revenues. Prior to this decade, the three traditional commercial television networks dominated the market and generally broadcast programs designed to appeal to a large, broad audience. When the Fox, Paramount, and Warner Brothers television networks launched during this decade, they faced the difficult task of competing with these firmly established, traditional networks. The fledgling networks adopted a marketing strategy to broadcast programs targeted at a narrow audience segment, specifically a much younger, male audience than the traditional networks. Uses and gratifications theory explains why network programmers target certain shows to specific audiences. Under this theoretical approach, researchers study the sources (both psychological and social) of audience members' needs which influence their expectations toward media (and alternative, competing sources), which result in various levels of exposure (or other activities) that cause the gratification of needs or some other (possibly unintended) consequence (Katz, Blumler, & Gurevitch, 1974, p. 20). In describing the uses and gratifications approach, Katz, Blumler, and Gurevitch specify the following five assumptions: 1. Audience members are described as active, demonstrating goal directed behavior toward the media; 2. Audience members link media choices with perceived gratifications; 3. Audience members choose between many sources, including the media, to satisfy their needs; 4. Audience members can explain their motives for using media sources at particular times; and 5. Value judgments pertaining to various audience uses of the media should be avoided (Katz, Blumler, & Gurevitch, 1974, p. 21-22). The uses and gratifications approach describes many examples of how media content can serve audience needs, including providing release from conflict or tension, supplying information or news, presenting entertainment or amusement, reinforcing personal identity, and assisting in integration into social group membership (Katz, Blumler, & Gurevitch, 1974, p. 27). Recent theoretical refinement of the approach has defined audiences as "variably active" and identified two types of audience attention to the media: "ritualized" or diversionary viewing which is a "less active or less goal-directed state" and "instrumental" or utilitarian viewing, which is "active and purposive" (Rubin, 1994, p. 427). Finn (1992) has defined audience activity levels as either proactive (serving mood management needs) or passive (providing social compensation). By applying this theoretical approach to television viewing behavior, networks attempt to predict viewers' selection of television programs based on the perceived needs and interests of the target audience. In order to cultivate programs for a specific audience, the Fox, Paramount, and Warner Brothers networks broadcast shows designed to appeal to the specific interests of the young, male viewer. One way to gain the attention of this particular audience is to portray attractive female characters in provocative attire. This type of portrayal exemplifies one pattern of female stereotyping. Stereotypes Walter Lippman is credited with the introduction of the concept of stereotyping to social scientific research in 1922 (Seiter, 1986). Over the past seventy years, vast quantities of research has been generated concerning this concept. Erving Goffman's (1974) frame analysis theory provides one theoretical basis for the stereotyping process. According to Goffman, individuals make sense of their experiences by framing each set of stimuli to corresponding sets of expectations that explain given social situations (1974, pp. 10 - 11). Universal or subtle social cues dictate which frame is appropriate in any given context. Frame analysis is based on the assumption that individuals view events as consistent experiences where different rules or frameworks can be employed as "a way of describing the event to which it is applied" (Goffman, 1974, p.24). Goffman (1979) illustrated his framing analysis theory with examples of gender portrayals in newspaper and magazine advertisements. Although portrayals in advertisements do not necessarily correspond to the actual behavior of men and women, they do represent how people perceive men and women to behave (Goffman, 1979, p. vii). Media cues depicting women as deferent, or sexual, therefore, could cause framing errors when applied to real life settings. A framing error describes a misinterpreted cue between two individuals (e.g., an epileptic seizure is mistaken for drunkenness) (Goffman, 1974, p. 311). Media cues generated by stereotypical portrayals of women in prime time network television programs could cause similar framing errors regarding gender social roles. Hansen and Hansen (1988) examined the effect of sex role stereotype portrayals in music videos on the accessibility of an individual's sex role schemas (or structured cognitive processing patterns). When multiple schemas exist to explain a given situation, the selected schema typically come from the most frequently activated schema in the past because it is easier to remember (or more accessible in recent memory); a schema can also be "primed" or enhanced by its recent activation from memory (Hansen & Hansen, 1988, p. 288). In an experiment, the researchers exposed individuals to either sexual stereotyped or non-stereotyped music videos and then assessed the subjects' evaluation of a man and woman interacting as sexually stereotyped or non-stereotyped behavior. Their findings supported that exposure to stereotyped music videos contributed to the "priming" of stereotyped schemas in subjects. Hansen and Hansen (1988) concluded that: "The impact of mass media fantasy depictions of sex role stereotypic persons and behaviors . . . can be extended to the domain of the real by their capacity to prime biased appraisal of subsequently encountered real persons and behaviors" (p. 312). Television network programs' stereotypical portrayal of women could similarly "prime" viewers' schemas and reinforce stereotypical attitudes. Feminist Theory The fundamental tenet of Feminist Theory focuses on "a theoretical acknowledgment of women's traditional devaluation . . . in relation to men with the assumption that the relationship needs to change" (Steeves, 1987, p. 96). Liberal feminism assumes that this change can be accomplished within existing democratic political structures because they promote an ideal of equal opportunities for all members. In applying Feminist Theory to stereotypical sex role portrayals, some liberal feminists argue that irrational prejudice can be overcome simply by rational argument (Steeves, 1987, p. 100). Other liberal feminists view stereotypes as more complex issues resulting from socialization. By adopting a cognitive (Kohlberg, 1966) or social learning (Bandura, 1977) approach, liberal feminists interpret children's sex role development and stereotypes as a pattern of modeling behavior and subsequent reinforcement that can be altered via rational argument, legal action, and the creation of alternative female role models (Steeves, 1987, p. 100). Stereotypical portrayals of women during prime time television programs, therefore, could perpetuate stereotypes through modeling behaviors; these same stereotypes may be undermined by presenting alternative televised role models of women. Cultural feminism similarly believes that media portrayals of women should be "deconstructed" (identified as patriarchal oppression against women) in order to advance women's equality. Susan Bordo describes the body as a cultural medium that is a locus of practical, direct social control (1993, p. 165). Essentially, there is no "natural" body; the body always exists within a cultural context. In Western culture, the mind or spirit has long been regarded as superior to the physical or material body (Cirksena and Cuklanz, 1992, p. 33). Influenced by such philosophers as Plato, Descartes, and Augustine and the tenets of Christianity, Western culture views the mind as separate from the body: The mind is conceived as the will, the soul, creativity, and freedom imprisoned by the animalistic, ravenous, sinful body (Bordo, 1993). Since women have long been defined by their bodies through its reproductive and caregiving nature, this mind/body dualism is a gendered one in patriarchal society. Women's plight in Western culture has been dictated by their inextricable link to the body, implying "the ideological belief that women represent a lower order of beings, that they are 'less transcendental of nature than men'" ( Wooley, 1994, p. 20). Many feminist scholars view the body as a battleground where women's bodies and their representation have long been controlled by men (Wooley, 1994). Another way feminist theorists deconstruct images of women in the media is by identifying the media's patriarchal disposition to depict women as objects. Liesbet van Zoonen describes this concept as the "display of woman as spectacle" resulting in "the 'objectification' of women" (1994, p. 87). Due to this objectification, mediated portrayals of women tend to be more stereotypical compared to their male counterparts. Female characters, indeed, are much more likely to be portrayed as sexual objects or physically beautiful than male characters. These "appearance norms" impose a submissive status to women who are to be looked at and subsequently judged; appearance is a form of social control over women practiced by a patriarchal society (Rothblum, 1994, p. 71). Prior Content Analyses A vast body of research has been generated to study stereotypical portrayals of women on television. Bretl and Cantor (1988) summarized 15 years of content analyses in television commercials in the U. S. Since 1971, findings indicated that male characters consistently appeared with greater frequency than female characters and female characters were delegated to traditional settings, roles, and occupations. Male narrators in advertisements were also the norm. Although Bretl and Cantor (1988) indicated that these gaps between male and female character portrayals may be declining, recent content analyses do not support this hypothesis [Smith (1994), Signorielli, McLeod, & Healy (1994), Craig (1992), Lovdal (1989)]. In a study of advertisements on MTV, Signorielli, McLeod, and Healy (1994) found that women were consistently underrepresented and stereotypically portrayed. The researchers specifically compared female and male characters based on frequency of portrayal, physical fitness, attractiveness, type of attire (provocative vs. neutral), and frequency of a character portrayed as the object of another character's stare (or gaze) (Signorielli, McLeod, & Healy, 1994, p. 93). The findings confirmed that stereotypical portrayals of women existed in every category: Female characters appeared less often, were more physically fit, more attractive, wore more provocative attire, and were objects of another character's stare more often than male characters. Comparable to the current study, Donald Davis (1990) conducted a content analysis of prime time network television programs to determine sex role stereotyping. Prior content analyses of network programming [Head (1954), DeFleur (1964), Tedesco (1974), Turow (1974), Signorielli (1989)] all encountered underrepresented and stereotypical portrayals of women. In his more recent study, Davis categorized male and female prime time television characters by frequency of portrayals, apparent age, hair color, parental and marital status, and provocative attire (1990, p. 329). His findings indicated that female characters were portrayed less often and were more likely to be younger, blonde, and dressed more provocatively than their male counterparts (Davis, 1990, p. 329). The marital and parental status of female characters was also more likely to be identified than for male characters (Davis, 1990, p. 330). Davis concluded that the prime time television portrayals of female demographic data had not significantly changed since the 1950s. Based on content analyses generated over the past 40 years, portrayals of women on network television programs appear to have remained consistently stereotypical. This study's first hypothesis, that female characters will appear less frequently than male characters on prime time network television programs, corresponds with the trends evidenced in previous content analytic research findings [Smith (1994), Signorielli, McLeod, & Healy (1994), Craig (1992), Lovdal (1989)]. The theoretical definition of the frequency of female character portrayals will be based on the number of female actresses portrayed in major, minor, and voice-over announcer roles on the networks' prime time programming compared to their male counterparts in the same roles. With the recent launch of three new television networks in this decade,[1] stereotypical portrayals of women may actually have increased due to the young, male audiences targeted by these networks. This study also hypothesizes that television programs on networks that target a younger male audience will contain more stereotypical sex role portrayals than other networks' programming. According to television network audience data, FOX and UPN were identified as the two networks that target young male viewers.[2] Based on a uses and gratifications theoretical perspective, television networks will provide programming that appeals to the perceived needs of their target audiences. Since the newer networks target a youthful male audience, their programs have a greater likelihood of containing stereotypical portrayals of women to appeal to this specific audience segment. Various types of character portrayals measured in previous content analyses (e.g., physical attractiveness, provocative attire, physical fitness, hair color, and age) will provide the theoretical definitions for stereotypical role portrayals in this study. A panel of experts was utilized to establish stereotypical levels for these variables. Method In order to test the study's hypotheses that (1) female characters will not appear as frequently as male characters and (2) the network's target audience will determine the level of female stereotyping in television network programs, a content analysis was developed to quantify portrayals and to identify stereotyping in prime time television network programs. Like other content analytic research, this study objectively defined, categorized, and compared various aspects of a communication message, specifically female depictions in prime time network programming. In Content Analysis: An Introduction to its Methodology, Klaus Krippendorff provides a definitive reference volume for content analytic research. In this text, Krippendorff defines content analysis as "a research technique for making replicable and valid inferences from data to their context" (1980, p. 21). To clarify the procedures for this method of research, Krippendorff details unitizing, sampling, and recording techniques. He also identifies reliability and validity tests appropriate to the research method. These guidelines, therefore, were carefully utilized in the current research study. Other texts describing content analyses methodology [Babbie (1995), Weber (1990), Holsti (1969), Budd, Thorp, & Donohew (1967)], provided additional references for the current work. Sampling Promotional advertisements were videotaped for prime time network television programs broadcast Monday through Sunday from 8 PM to 11 PM on five networks (ABC, CBS, NBC, FOX, and UPN) in a mid-size, northeastern U. S. television market.[3] Because these promotional spots are designed to entice viewers, it is likely that the most exciting (and probably most stereotypical) content within the program will appear during these advertisements. It was assumed that networks were more likely to air these promotional spots during prime time; therefore, the 8 PM to 11 PM time period was chosen for the study. Promotional advertisements were recorded on consecutive evenings from September 11 to September 17, 1995. Due to the recent or imminent launch of the networks' new television season, this week was deemed particularly germane for the study as a time when networks would be heavily promoting their new programs. Network promotional spots were defined as any advertisement that promotes a prime time network program. In order to avoid including syndicated and local programs within the sample, TV Guide's Fall Preview issue (Sept. 16, 1995) was used to identify the titles of network programs to be included in the study. All network promotional advertisements broadcast during this sample were content analyzed, coding female stereotyping within the promotional spots. Repeat promotional advertisements were coded as they appeared in the sample. Operational Definitions As stated previously, representatives from each network were contacted in order to identify specific target audiences. Network target audiences, therefore, have been defined by the networks themselves and confirmed through popular and trade press articles. During prime time, ABC, CBS, and NBC target an adult audience from 18 to 49 years of age. Prior to this year, FOX presented programs to appeal to a predominately younger, male audience. The FOX network is currently in a transitionary phase in an effort to expand its audience to more closely resemble the traditional three networks (e.g., adults aged 18 to 49). Modeling itself after FOX, UPN is initially pursuing a male audience between the ages of 18 and 34. In order to identify potential sex stereotyping in the promotional spots coded, this study measured the proportional representation of women in major and minor character roles. The race and age of each character were also recorded to identify the diversity of portrayals between men and women. Characters' attire, in addition, was coded to determine whether female characters were more often portrayed in provocative dress than male characters. Based on the stereotypical assumption that female characters will be presented as more attractive and physically fit, these two characteristics were also coded. In order to replicate findings from a previous content analysis of prime time network programs (Davis, 1990), hair color was also identified to determine if female characters are more often portrayed as blondes than male characters. In order to identify if certain program formats portray more females than males and vice versa, the genre of each program was coded (e.g., comedy, drama, news, etc.). If a voice-over announcer was utilized in the promotional spot, the gender of the announcer was coded to determine if male voices were more frequently used than female voices. For each of the promotional advertisements coded in the study, therefore, stereotypical sex role portrayals were analyzed for the following characteristics: frequency of portrayal; character role (e.g., primary character, minor character); character race; apparent age; type of attire (e.g., provocative); physical attractiveness; physical fitness; and hair color. Promotional announcements were also coded for genre (e.g., drama, comedy, news program, etc.) and announcer (voice-over) gender. The coding instruction manual (Appendix A) provides the complete listing of operational definitions for each of these characteristics. In order to classify certain variables coded in the content analysis as stereotypical, a panel of "experts" validated the level of stereotyping contained in each of the following variables: character attire, attractiveness, physical fitness, hair color, and age. A purposive sample of undergraduate, graduate, and faculty members at a communications school in the northeastern United States was asked to rank each variable indicator on a five-point Likert scale ranging from very stereotypical to not at all stereotypical. Results were then used to create an index of "stereotypicality" for each variable measured. Coders and Inter-Coder Reliability In order to train coders for this content analysis, the researcher developed a coding instruction manual detailing every aspect of coding procedures (Appendix A). Two coders received extensive pre-test training: studying the coding instruction book, perfecting individual understanding of each variable definition, and coding several sample programs in order to establish inter-coder reliability prior to coding the promotional advertisements for the main study. As a result of these pre-test coding sessions, the coding sheets were revised in order to perfect variable definitions and facilitate accurate coding. Holsti's coefficient of reliability (( 2M / N1 + N2) and Scott's pi[4] were calculated to determine an acceptable level of inter-coder reliability for the study (Holsti, 1969, p. 140). The statistical package SPSS Version 6.1 was utilized for statistical data analysis. Frequency distributions, cross tabulations, and analyses of variance were used for the statistical analysis of the data. Results A total of 811 promotional announcements and 969 characters were coded from the sample. Based on Holsti's coefficient of reliability, inter-coder reliability levels ranged between .74 and .96, all within acceptable reliability standards. Scott's pi or index of reliability was computed for the five stereotyping variables, with an average value of .81.[5] Table 1 provides percentage results for variables coded for all television networks. The predominant program formats were comedy (51.3%) and drama (27%). Male voices dominated as voice-over announcers comprising 95.7%; the remaining promotional advertisements featured no announcer (2.5%), both male and female announcers simultaneously (1.6%) or female announcers (0.2%). The majority of characters were male (55.6%) and Caucasian (85.2%). The characters portrayed also tended more frequently to be young adults (61.7%) or middle-aged (30.7%). The majority of characters were shown in major roles (67.8%). Most of the variables measuring the level of stereotyping of characters tended to cluster around "neutral" or "average" values. The most frequently occurring category for character attire, for example, was neutral (63.2%), with somewhat unprovocative (16.9%) and somewhat provocative (13.6%) ranking second and third. The character attractiveness variable was mainly split between two categories: average (45.2%) and somewhat attractive (43.4%). The physical fitness of characters again heavily favored the average category (80.7%). Finally, the majority of characters' hair color was brown (47.5%), blonde (20.1%), and black (16.4%). Crosstabulation results in Table 4 show that the majority of characters portrayed were white young adults, regardless of gender. Female characters were more likely to be coded in a major role (70.5%) than male characters. All characters were predominately classified in the neutral attire category, although more women were found in the somewhat provocative (24.0% for women, 5.4% for males) and very provocative (6.7% for women and 5.9% for males) categories. Female characters were also more likely to be somewhat attractive (57.2%) than male characters (32.5%). The majority of men (81.3%) and women (80.0%) were in average physical shape, although a larger percent of female characters (14.0%) were in somewhat good shape than male characters (2.2%). The predominate hair color among male characters was brown (64.2%) and blonde (38.8%) for female characters. Crosstabulation results for female characters categorized by network representation is presented in Table 5. These findings have been reported without corresponding statistics due to the distinct nature of the results: Most of the crosstabulation tables yielded results with more than 20 percent of cells containing expected frequencies of less than five. Rather than collapse categories and recode variables to correct this problem, it was determined that the empty cells themselves provided significant results for analyses that would be lost if variables were recoded. All networks portrayed predominately Caucasian characters. FOX portrayed the highest percentage of African-American women (26.7%), while ABC and UPN did not portray any African-American female characters.[6] Young adult was the most frequent age category for female characters on all networks except CBS, with half of its characters coded as middle-aged. ABC, CBS, and FOX were the only networks to portray female children (3.7%, 0.6%, and 12.0%, respectively) while ABC ranked heaviest on the elderly age category (15%). NBC, FOX, and UPN did not portray any elderly women characters. All television networks except FOX portrayed more women in major roles than minor ones. For the character attire variable, all of the networks most frequently portrayed characters dressed in neutral attire. FOX and UPN, however, more frequently portrayed female characters in very provocative dress (21.3% and 18.8%, respectively). No networks portrayed any female characters in the very unprovocative attire category. ABC and NBC were more likely to portray female characters as average in attractiveness (46.7% and 56.4%, respectively) compared to the somewhat attractive category for CBS (70.2%), FOX (69.3%), and UPN (62.5%). UPN and ABC ranked highest in very attractive female portrayals (15.6% and 12.1%, respectively), while CBS and NBC ranked highest in the somewhat unattractive portrayal category (6.2% and 5.5%). No networks portrayed any women in the very unattractive category. For female characters' physical fitness, all networks most often portrayed average body types. FOX and UPN were the only networks to portray women who were very fit and ABC was the only network to portray females who were very unfit. All of the networks portrayed predominately blonde characters, although NBC portrayed almost equal numbers of blonde (36.4%) and brown-haired (34.5%) characters. Two-way analyses of variance were run for all stereotypical variables coded. In order to assess the level of "stereotypicality" for these variables, a panel of experts consisting of 25 undergraduate, graduate, and faculty members at a communications school in the northeastern United States determined each variable's level of stereotyping (Appendix C). Individuals who study mass communication and television were deemed more likely to be "experts" in the area of identifying stereotypical television portrayals of women. The purposive sample included approximately equal numbers of men and women within each age group. Values for each stereotypical variable were calculated by taking the mean response from the 25 experts (Table 6). For the variable character hair color, for example, the value "blonde" received a rating of 4.36 and the value "brown" was rated at 2.88. This permitted the five stereotyping variables (e.g., attire, attractiveness, age, physical fitness, and hair color)[7] to be used as the dependent measures in the analyses of variance (Figures 1 through 5). Two-way analyses of variance of the stereotypical variables by network and character gender generated some interesting interactions. Figure 1 depicts stereotypical attire and demonstrates a clear two-way interaction between the veteran three networks and FOX and UPN. The two newer networks demonstrate much higher levels of stereotypical female attire portrayals than male compared to ABC, CBS, and NBC. Figure 2 depicts a very similar relationship, with FOX and UPN ranking female characters higher on the attraction stereotype scale than males. The other three networks show more congruent portrayals among male and female characters. Figure 3 shows NBC and FOX as clearly depicting female characters' by stereotypical ages more than the other networks. Males on ABC actually ranked higher in stereotyping for this variable than women. UPN ranked both men and women highly (although men slightly higher than women). ABC and CBS portrayed similar levels of fitness between female and male characters in Figure 4, while NBC demonstrated a tendency to depict females more stereotypically than male characters regarding physical fitness. The two newer networks, however, again portrayed the most striking divergence between male and female characters compared to the veteran networks. Both FOX and UPN rank females significantly higher than males for this stereotypical portrayal. The hair color stereotype in Figure 5 was the only variable measured which demonstrated no interaction. Female characters consistently ranked higher than males for this stereotypical portrayal across all networks except UPN. The portrayal of females and males for stereotypical hair color was more comparable on UPN than the other networks. Discussion The "symbolic annihilation of women" described by Gaye Tuchman over 15 years ago in Hearth and Home (1978) appears to be alive and well in prime time television portrayals today. This content analysis of prime time television network promotional advertisements demonstrated that female characters are still significantly underrepresented compared to male characters. It is also surprising that television networks were more likely to use no voice-over announcer than female announcers. In fact, women are practically invisible in the role of announcer for every network. The first hypothesis that female characters will not appear as frequently as male characters, was supported. This finding demonstrates a persistent trend of women's underrepresentation on television, as evidenced in the work of Signorielli, McLeod, and Healy (1994), Davis (1990), Bretl and Cantor (1988), and many others. Although not specifically hypothesized in this research, the role of minorities on prime time television may have faired even worse than women, with whites comprising well over three-fourths of all portrayals. African-Americans account for approximately ten percent, while all other races are almost non-existent. The "other ethnicity" category (comprising 3.4% of all characters coded) is somewhat misleading; it does not represent diversity in ethnic portrayal because it consists predominately of cartoon characters that did not represent any race. The fact that these "non-race" characters rank third in ethnic portrayals in prime-time programming is remarkable. These results are consistent with previous content analytic studies conducted by mass communication researchers and suggest that prime time television representations of minorities have not changed significantly in recent years. Results showed some support for the study's second hypothesis, that the level of female stereotyping in television network programs depends on the network's target audience. FOX and UPN were predicted to depict more stereotypical portrayals of women in programming designed to target their predominately young, male audiences. Across all networks, female characters were generally portrayed in more provocative attire, as more attractive, and more physically fit than male characters. These findings directly support the previous work of Signorielli, McLeod, and Healy (1994) who also found that women on television were portrayed as more attractive, more provocatively dressed, and more physically fit than male characters. Women were also more likely than men to be blondes, which confirms a previous study conducted by Davis (1990). This study's results, therefore, support forty years of content analytic research describing female characters as consistently more stereotyped than male characters on television [Head (1954), DeFleur (1964), Tedesco (1974), Turow (1974), Signorielli (1989)]. Apparently, there has been little or no development in televised female portrayals despite suggestions to the contrary by network executives and the popular press. The study's second hypothesis actually contributes a new dimension to the content analytic stereotyping literature by comparing female portrayals across five networks. When specifically analyzing FOX and UPN portrayals compared to the other three networks, crosstabulation results indicate that FOX was more likely to portray women in minor character roles, dressed in provocative attire, and more physically fit than the other networks. UPN was also more likely to depict women as more attractive, physically fit, and provocatively dressed than the three veteran networks. The two-way interactions depicted in Figures 1 through 4 show clear support that FOX, and to a lesser extent UPN, depict females more stereotypically across the majority of variables tested. These results support the uses and gratifications approach defining audiences as choosing television content based upon perceived needs; television networks, subsequently, produce specific content in their programs in order to attract target audience members. With FOX and UPN targeting a younger, male audience, it is logical that these networks' portrayals of women will be somewhat more stereotypical than the other three networks that generally target male and female viewers between the ages of 18 and 49. One key limitation of the current study concerns the sampling of promotional advertisements rather than entire prime time programs. Future research perhaps should not restrict analysis to promotional announcements. It is possible that the study's assumption that promotional announcements are more likely to contain more exciting (and more stereotypical) content than the programs themselves may have been an erroneous one. Future studies should analyze entire program content in order to replicate these findings. Table 1. Percentages for promotional announcement variables. Variables % ------------------------------------------------------------------------------- Promo format Action Adventure 5.3 Comedy 51.3 Drama 27.0 News 3.7 Other 12.7 . 100.0% (N=811) Voice-over announcer gender Male 95.7 Female 0.2 Both 1.6 None . 2.5 . 100.0% (N=811) Table 2. Percentages for character demographic variables. Variables % ------------------------------------------------------------------------------- Character sex Male 55.6 Female 44.4 . 100.0% (N=969) Character race Caucasian 85.2 African American 10.7 Hispanic 0.5 Asian 0.1 American Indian 0.0 Other . 3.4 . 100.0% (N=969) Character role Major 67.8 Minor 32.2 . 100.0% (N=969) Table 3. Percentages for character stereotype variables. Variables % ------------------------------------------------------------------------------- Character age Child 3.1 Teenager 0.9 Young adult 61.7 Middle-aged 30.7 Elderly . 3.6 . 100.0% (N=969) Character attire Very provocative 6.3 Somewhat provocative 13.6 Neutral 63.2 Somewhat unprovocative 16.9 Not provocative . 0.0 . 100.0% (N=969) Character attractiveness Very attractive 4.5 Somewhat attractive 43.4 Average 45.2 Somewhat unattractive 6.8 Not attractive . 0.0 . 100.0% (N=969) Character physical fitness In good shape 1.9 In somewhat good shape 7.4 Average 80.7 Somewhat out of shape 7.6 Very out of shape . 2.4 . 100.0% (N=969) Character hair color Brown 47.5 Black 16.4 Blonde 20.1 Red/Auburn 10.6 Gray/White 3.5 Bald 0.9 Other . 0.9 . 100.0% (N=969) Table 4. Crosstabulation of character variables by gender. Gender Variable Male Female -------------------------------------------------------------------------------- ---------- Character race Caucasian 81.1 90.5 African American 14.3 6.3 Other 4.6 3.3 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 17.90, df = 2, p < .001 Cramer's V = .14 Variable Male Female -------------------------------------------------------------------------------- ---------- Character age Child 3.0 3.3 Teenager 1.1 0.7 Young adult 62.7 60.5 Middle-aged 30.1 31.4 Elderly 3.2 4.2 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 1.55, df = 4, ns Cramer's V = .04 Variable Male Female -------------------------------------------------------------------------------- ---------- Character role Major 65.7 70.5 Minor 34.3 29.5 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 2.52, df = 1, ns Cramer's V = .05 Table 4 (continued). Crosstabulation of character variables by gender. Variable Male Female -------------------------------------------------------------------------------- ---------- Character attire Very provocative 5.9 6.7 Somewhat provocative 5.4 24.0 Neutral 63.6 62.6 Somewhat unprovocative 25.0 6.7 Not provocative 0.0 0.0 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 108.20, df = 3, p < .001 Cramer's V = .33 Variable Male Female -------------------------------------------------------------------------------- ---------- Character attractiveness Very attractive 2.2 7.4 Somewhat attractive 32.5 57.2 Average 55.5 32.3 Somewhat unattractive 9.8 3.0 Unattractive 0.0 0.0 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 92.67, df = 3, p < .001 Cramer's V = .31 Variable Male Female -------------------------------------------------------------------------------- ---------- Character physical fitness In good shape 2.2 1.4 In somewhat good shape 2.2 14.0 Average 81.3 80.0 Somewhat out of shape 11.9 2.3 Very out of shape 2.4 2.3 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 73.77, df = 4, p < .001 Cramer's V = .28 Table 4 (continued). Crosstabulation of character variables by gender. Variable Male Female -------------------------------------------------------------------------------- ---------- Character hair color Brown 64.2 26.5 Black 20.4 11.4 Blonde 5.2 38.8 Red/Auburn 1.7 21.9 Other 8.5 1.4 -------------------------------------------------------------------------------- ---------- 100.0% 100.0% (N=539) (N=430) X2 = 332.35, df = 4, p < .001 Cramer's V = .59 Table 5. Crosstabulation of female characters by network. Variable ABC CBS NBC FOX UPN -------------------------------------------------------------------------------- ------------------------- Character race Caucasian 100.0 96.3 94.5 61.3 90.6 African American 0.0 2.5 5.5 26.7 0.0 Other 0.0 1.2 0.0 12.0 9.4 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) -------------------------------------------------------------------------------- ------------------------- Character age Child 3.7 0.6 0.0 12.0 0.0 Teenager 1.9 0.6 0.0 0.0 0.0 Young adult 47.7 46.6 85.5 88.0 65.6 Middle-aged 31.8 50.9 14.5 0.0 34.4 Elderly 15.0 1.2 0.0 0.0 0.0 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) -------------------------------------------------------------------------------- ------------------------- Character role Major 83.2 73.3 81.8 42.7 59.4 Minor 16.8 26.7 18.2 57.3 40.6 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) -------------------------------------------------------------------------------- ------------------------- Character attire Very prov. 0.0 2.5 5.5 21.3 18.8 Somewhat prov. 25.2 21.7 16.4 34.7 18.8 Neutral 73.8 65.8 72.7 40.0 43.8 Somewhat unprov. 0.9 9.9 5.5 4.0 18.8 Very unprov. 0.0 0.0 0.0 0.0 0.0 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) Table 5 (continued). Crosstabulation of female character variables by gender. Variable ABC CBS NBC FOX UPN -------------------------------------------------------------------------------- ------------------------- Character attractiveness Very attract. 12.1 2.5 7.3 8.0 15.6 Somewhat attract. 41.1 70.2 30.9 69.3 62.5 Average 46.7 21.1 56.4 22.7 21.9 Somewhat unattract. 0.0 6.2 5.5 0.0 0.0 Very unattract. 0.0 0.0 0.0 0.0 0.0 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) -------------------------------------------------------------------------------- ------------------------- Character fitness Very fit 0.0 0.0 0.0 1.3 15.6 Somewhat fit 7.5 7.5 16.4 36.0 12.5 Average 81.3 90.7 83.6 61.3 59.4 Somewhat unfit 1.9 1.9 0.0 1.3 12.5 Very unfit 9.3 0.0 0.0 0.0 0.0 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) -------------------------------------------------------------------------------- ------------------------- Character hair color Brown 31.8 28.0 34.5 12.0 21.9 Black 7.5 9.3 5.5 21.3 21.9 Blonde 43.0 36.0 36.4 38.7 43.8 Red/Auburn 17.8 26.7 23.6 25.3 0.0 Other 0.0 0.0 0.0 2.7 12.5 -------------------------------------------------------------------------------- ------------------------- 100.0% 100.0% 100.0% 100.0% 100.0% (N=107) (N=161) (N=55) (N=75) (N=32) _ Table 6. Mean expert rating for character stereotype variables. Mean expert rating Variables of stereotype* ------------------------------------------------------------------------------- Character attire Very provocative 4.28 Somewhat provocative 3.88 Neutral 2.84 Somewhat unprovocative 2.32 Not provocative 2.52 Character attractiveness Very attractive 4.52 Somewhat attractive 3.80 Average 2.84 Somewhat unattractive 2.60 Not attractive 2.68 Character age Child 2.76 Teenager 3.60 Young adult 4.28 Middle-aged 3.28 Elderly 2.40 Character physical fitness In good shape 4.32 In somewhat good shape 3.84 Average 3.04 Somewhat out of shape 2.52 Very out of shape 2.96 Character hair color Brown 2.88 Black 2.72 Blonde 4.36 Red/Auburn 3.32 Gray/White 2.36 *Scale was coded: 1 = not at all stereotypical, 2 = somewhat not stereotypical, 3 = neutral, 4 = somewhat stereotypical, 5 = very stereotypical. Main effects: Network, F = 3.94, df = 4, 959, p < .01 Gender, F = 82.33, df = 1, 959, p < .001 Interaction: Network by gender, F = 12.31, df = 4, 959, p < .001 Main effects: Network, F = 6.77, df = 4, 959, p < .001 Gender, F = 81.97, df = 1, 959, p < .001 Interaction: Network by gender, F = 9.69, df = 4, 959, p < .001 Main effects: Network, F = 11.39, df = 4, 959, p < .001 Gender, F = .57, df = 1, 959, ns Interaction: Network by gender, F = 5.85, df = 4, 959, p < .001 Main effects: Network, F = 1.98, df = 4, 959, ns Gender, F = 51.34, df = 1, 959, p < .001 Interaction: Network by gender, F = 14.56, df = 4, 959, p < .001 Main effects: Network, F = 3.30 df = 4, 941, p < .05 Gender, F = 307.19, df = 1, 941, p < .001 Interaction: Network by gender, F = 3.37, df = 4, 941, ns References Babbie, E. R. (1995). The practice of social research. Belmont, CA: Wadsworth Publishing Company. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall. Bordo, S. (1993). Unbearable weight: Feminism, western culture, and the body. Berkeley, CA: University of California Press. Bretl, D. J. & Cantor, J. (1988). The portrayal of men and women in U. S. television commercials: A recent content analysis and trends over 15 years. Sex Roles, 18, 595-609. Budd, R. W., Thorp, R. K., & Donohew, L. (1967). Content analysis of communications. New York, NY: Macmillian. Cirksena, K., & Cuklanz, L. (1992). Male is to female as _____ is to _____: A guided tour of five feminist frameworks for communications studies. In L. F. Rakow (Ed.), Women making meaning: New feminist directions in communication (pp. 18-44). New York, NY: Routledge. Coe, S. (1995, January 2). CBS makes major midseason shifts. Broadcasting & Cable, 125, 37. Craig, R. S. (1992). The effect of television day part on gender portrayals in television commercials: A content analysis. Sex Roles, 26, 197-211. Davis, D. M. (1990). Portrayals of women in prime-time network television: Some demographic characteristics. Sex Roles, 23, 325-332. DeFleur, M. (1964). Occupational roles as portrayed on television. Public Opinion Quarterly, 28, 57-74. Fall Preview '95. (1995, September 16-22). TV Guide, 43 (37), 12-73. Finn, S. (1992). Television addiction? An evaluation of four competing media-use models. Journalism Quarterly, 69, 422-435. Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA: Harvard University Press. Goffman, E. (1979). Gender advertisements. Cambridge, MA: Harvard University Press. Hansen, C. H. & Hansen, R. D. (1988). How rock music videos can change what is seen when boy meets girl: Priming stereotypic appraisal of social interactions. Sex Roles, 19, 287-316. Head, S. (1954). Content analysis of television drama programs. Quarterly of Film, Radio, and Television, 9, 175-194. Holsti, O. R. (1969). Content analysis for the social sciences and humanities. Reading, MA: Addison-Wesley Publishing Company. Katz, E., Blumler, J. G., & Gurevitch, M. (1974). Utilization of mass communications by the individual. In J. G. Blumler and E. Katz (Eds.), The uses of mass communications: Current perspectives on gratifications research, (pp. 19-32). Beverly Hills, CA: Sage Publications. Kohlberg, L. (1966). A cognitive developmental analysis of sex-role concepts and attitudes. In E. Maccoby (Ed.), The development of sex differences. Stanford, CA: Stanford University Press. Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage Publications. Lovdal, L. T. (1989). Sex role messages in television commercials: An update. Sex Roles, 21, 715-724. Rothblum, E. D. (1994). "I'll die for the revolution but don't ask me not to diet": Feminism and the continuing stigmatization of obesity. In P. Fallon, M. Katzman, and S. Wooley (Eds.), Feminist perspectives on eating disorders (pp. 53-76). New York, NY: Guildford Press. Rubin, A. M. (1994). Media uses and effects: A uses-and-gratifications perspective. In J. Bryant and D. Zillman (Eds.), Media effects: Advances in theory and research (pp. 417-436). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Seiter, E. (1986). Stereotypes and the media: A re-evaluation. Journal of Communication, 36 (2), 14-26. Signorielli, N. (1989). Television and conceptions about sex roles: Maintaining conventionality and the status quo. Sex Roles, 21, 341-360. Signorielli, N., McLeod, D. & Healy, E. (1994). Gender stereotypes in MTV commercials: The beat goes on. Journal of Broadcasting & Electronic Media, 38, 91-101. Smith, L. J. (1994). A content analysis of gender differences in children's advertising. Journal of Broadcasting & Electronic Media, 38, 323-337. Steeves, H. L. (1987). Feminist theories and media studies. Critical Studies in Mass Communication, 4, 95-135. Tedesco, N. (1974). Patterns in prime-time. Journal of Communication, 24, 119-124. Tuchman, G. (1978). Introduction: The symbolic annihilation of women by the mass media. In G. Tuchman, A. E. Daniels, and J. Benet (Eds.), Hearth and home: Images of women in the mass media. New York, NY: Oxford University Press. Turow, J. (1974). Advising and ordering: Daytime, prime-time. Journal of Communication, 24, 138-141. Weber, R. P. (1990). Basic content analysis. Beverly Hills, CA: Sage Publications. Wooley, O. W. (1994). ...And man created 'woman': Representations of women's bodies in western culture. In P. Fallon, M. Katzman, and S. Wooley (Eds.), Feminist perspectives on eating disorders (pp. 17-52). New York, NY: Guildford Press. Zoonen, L. van (1994). Feminist media studies. London, England: Sage Publications. Appendix A Content Analysis Coding InstructionsProgram Code Sheet Instructions Code all television program promotional announcements for prime time network programs that appear on the videotape between 8 PM and 11 PM (refer to D. below for timing instructions). Do not code promos for any shows that are not prime time (e.g., do not code promos for local or network news, talk shows like Dave Letterman or Oprah, daytime soap opera and the like). Complete one program code sheet for each promotional spot coded. The following represent some unique promotional spots and how to treat them: 1. "Split Promo" - a promo where two shows are advertised simultaneous with cuts between characters from each show -- code with two program code sheets for each show represented and then match the characters with the show they belong to. 2. "Montage Promo" - a promo where images from a dozen or so shows are cut together into one ad -- code each show that is mentioned on an individual program code sheet and then code only those characters that clearly belong to that show and meet the coding restrictions (10 seconds on the screen and/or at least three spoken words). 3. "Nestled Promo" - if promo is edited into the credits of a program (e.g., a split screen or box is used to broadcast the program's credits and a promo at the same time) do not code the promo. If the promo is cut in between an advertisement (e.g., TV Guide has a contest and shows a promo of a network show during its TV Guide ad), then don't code the promo. Only code promos that stand alone and are not part of some other format (e.g., credits or advertisement). A. Match the program title with the code numbers listed on pages 5 through 8 of this instruction book. Write the code number of the promotional spot's program title in the space provided. B. Match the television station network and call letters listed on the videocassette tape you are coding from with the station network and call letters provided on the program code sheet. C. Match the day of week listed on the videocassette tape you are coding from with the day of week provided on the program code sheet. D. When you first place videocassette tape in videocassette machine, tape will be cued up to 8 PM. Zero out the machine's timer and indicate the time each promotional aired based upon the timer reading at the start of the promotional spot. E. Define program format categories by the following: 1 Action Adventure - main theme of program is based upon characters that physically explore and move about in their environment (e.g., westerns, police shows, science fiction adventure shows, etc.). 2 Comedy - main theme of program is based upon comic dialogue and situations (e.g., situation comedies). 3 Drama - main theme of program is dramatized serious fictional content (e.g., programs dealing with court drama, family drama, etc.). 4 News - main theme of program is factual reporting of the news (e.g., news magazine programs like 20/20 and 60 Minutes.). 5 Other - for any type of format not included above, indicate program format in space provided. F. Identify the gender of voice-over announcer (voice whose image is never visually presented) as male or female. If the announcer's voice is computer generated or some other non-gender source, then code as other. If no voice-over occurs during the program, then code as none. Prime Time Stereotyping Character Descriptive Code Sheet Instructions Code all characters that are depicted for at least ten seconds and/or speak at least three words during the 30 second promotional spot. Do not code characters in instances where you think that characters may have spoken but announcer's voice drown them out (if you cannot clearly identify the character and the voice -- do not guess). Code a character with a "detached voice" (e.g., voice is heard without seeing a face speak) only if you are sure that the voice and body go together). Compete one character descriptive code sheet for each character to be coded. A. Character number: code two digit program number first (listed on pages 5 through 8 of this instruction book) followed by a two-digit character number you will systematically assign to each character (starting at 01). Do not repeat assigned numbers; each character will have a unique number assigned to it. B. Character name or description: list any formal name and/or nickname of the character or list a description of the character if the name is not identifiable. C. Character Role 1 Major - major characters share the majority of dialogue during the promotional spot, play the largest role in the dramatic action, and appear on the screen for the longest period of time during the spot. 2 Minor - minor characters have less or no dialogue, play a less significant role in the dramatic actions, and appear on the screen for a shorter period of time than major characters during the spot. For ensemble cast shows (e.g., ER, Central Park West, Chicago Hope), code all characters as minor because all are equally supporting roles to each other. D. Gender 1 Male 2 Female E. Ethnicity 1 Caucasian 2 African American 3 Hispanic 4 Asian 5 American Indian 6 Other - for any race not included above, indicate race in space provided. F. Approximate Age 1 Child (approx. 0 to 13 years of age) 2 Teenager (approx. 13 to 19 years of age) 3 Young adult (approx. 20 to 39 years of age) 4 Middle-aged adult (approx. 40 - 60 years of age) 5 Elderly adult (approx. 60 years or older) 6 Indeterminate (use this category sparingly -- estimate from above 6 categories if possible. G. Provocative Attire 1 Very provocative - character's clothing is scanty or sexy, exposing a generous amount of skin and/or is extremely tight-fitting (e.g., partially exposed cleavage, extremely short "mini" skirt). 2 Somewhat provocative - character's clothing is moderately scanty or sexy, exposing a moderate amount of skin and/or is somewhat tight-fitting (e.g. bare shoulders, somewhat short skirt). 3 Neutral - character's clothing is "average" (exhibiting no distinguishable characteristics that would place it in the provocative or conservative category; child usually coded here). 4 Somewhat not provocative - character's clothing is classically conservative (e.g., bland business suit or military uniform -- note: business suits do not include blazer and blouse on women, men with no tie, or men with tie and no coat -- characters must be wearing entire suit with coat and tie to meet #4 criteria). 5 Not provocative - character's clothing covers the majority of the body at a level considered prudish (e.g., shirt buttoned up to neck without tie or suit). H. Attractiveness 1 Very attractive - character's facial features are strikingly attractive or beautiful (e.g., character is striking with "movie star" looks). (Note: if you see only part of a face, code as attractive if the lips or nose is attractive.) 2 Somewhat attractive - character's facial features are above average in attractiveness or beauty (e.g., character is attractive or "cute" but not striking). 3 Average - character's facial features exhibit no distinguishable characteristics that would place it in the attractive or unattractive categories (e.g., character is "normal" looking -- not attractive or unattractive; child usually coded here). 4 Somewhat unattractive - character's facial features are somewhat ugly or homely (e.g., character has unattractive trait like large nose). 5 Very unattractive - character's facial features are ugly or homely (e.g., overall facial appearance is ugly with many homely attributes). I. Body Type/Physical Fitness 1 In good shape - character's body appears to be very muscular and toned (e.g., actually see large size muscles or extremely toned or rippled muscles). 2 In somewhat good shape - character's body appears to be somewhat muscular and toned (e.g., clothes are tight enough to display firm body underneath). 3 Average - character's body is not overweight but does not show obvious indications of toned muscles (e.g., clothes give appearance that body is thin but non-descript in muscle tone; child usually coded here). 4 Somewhat out of shape - character's body is somewhat overweight and flabby (e.g., character has "bulge" around stomach area but otherwise not fat). 5 Very out of shape - character's body is very overweight and flabby (e.g., character is obese). J. Hair Color - chose the predominate hair color of the character from the list provided (e.g., if a character has black hair with a little gray, categorize as black). If hair color of the character is not apparent (e.g., character's hair is covered with hat), then list as indeterminate. 1 Brown 2 Black 3 Blonde 4 Red/Auburn 5 Gray/White 6 Bald 7 Other (SPECIFY ____________________________) Program Coding Numbers ABC Saturday 001 - The Jeff Foxworthy Show 002 - Maybe This Time 003 - ABC Family Movie (Encino Man) 108 - ABC Family Movie (Great Mom Swap) Sunday 004 - America's Funniest Home Videos 005 - Lois & Clark 006 - ABC Sunday Night Movie (Stranger Beside Me) 109 - ABC Sunday Night Movie (Lethal Weapon III) Monday 007 - The Marshall 008 - ABC's NFL Monday Night Football Tuesday 009 - Roseanne 010 - Hudson Street 011 - Home Improvement 012 - Coach 013 - NYPD Blue Wednesday 014 - Ellen 015 - The Drew Carey Show 016 - Grace Under Fire 017 - The Naked Truth 018 - PrimeTime Live Thursday 112 - Day One 019 - Charlie Grace 020 - The Monroes 021 - Murder One Friday 022 - Family Matters 023 - Boy Meets World 024 - Step By Step 025 - Hangin' With Mr. Cooper 026 - 20/20 CBS Additionals 102 - Her Deadly Rival 103 - Too Hot to Skate 104 - Whose Daughter is She? 113 - Color Me, Barbra 114 - Sneak Peak Saturday 027 - Dr. Quinn, Medicine Woman 028 - Touched By An Angel 029 - Walker, Texas Ranger Sunday 030 - 60 Minutes 031 - Cybill 032 - Almost Perfect 033 - CBS Sunday Night Movie Monday 034 - The Nanny 035 - Can't Hurry Love 036 - Murphy Brown 037 - If Not For You 038 - Chicago Hope Tuesday 039 - John Grisham's The Client 040 - CBS Tuesday Night Movie (The Secretary) Wednesday 041 - Bless This House 042 - Dave's World 043 - Central Park West 044 - Courthouse Thursday 045 - Murder, She Wrote 046 - New York News 047 - 48 Hours Friday 048 - Dweebs 049 - Bonnie Hunt 050 - Picket Fences 051 - American Gothic Fox Saturday 052 - Martin 053 - The Preston Episodes 054 - Cops 055 - America's Most Wanted Sunday 056 - Space: Above and Beyond 057 - The Simpsons 058 - Too Something 059 - Married With Children 060 - Misery Loves Company 105 - Fox Football -- NFL Sunday Monday 061 - Melrose Place 062- Partners 063 - Ned and Stacey 106 - Emmy Awards Special Tuesday 064 - The Fox Tuesday Night Movie (Divas) 116 - The Fox Tuesday Night Movie (In the Name of Love) Wednesday 065 - Beverly Hills, 90210 066 - Party of Five Thursday 067 - Living Single 068 - The Crew 069 - New York Undercover Friday 070 - Strange Luck 071 - The X-Files NBC Saturday 072 - JAG 073 - The John Larroquette Show 074 - The Home Court 075 - Sisters Sunday 076 - Brotherly Love 077 - Minor Adjustments 078 - Mad About You 079 - Hope & Gloria 080 - NBC Sunday Night at the Movies (Zoya) 110 - NBC Sunday Night at the Movies (Unspoken Truth) 107- Miss America Pageant Monday 081 - Fresh Prince 082 - In the House 083 - NBC Monday Night Movie (Various/multiple movie promo) 111 - NBC Monday Night Movie (Beauty's Revenge) 117 - NBC Monday Night Movie (Thelma and Louise) Tuesday 084 - Wings 085 - NewsRadio 086 - Fraiser 087 - The Pursuit of Happiness 088 - Dateline NBC Wednesday 089 - seaQuest DSV 090 - Law and Order Thursday 091 - Friends 092 - The Single Guy 093 - Seinfeld 094 - Caroline in the City 095 - ER Friday 096 - Unsolved Mysteries 097 - Homicide: Life in the Streets UPN Monday 098 - Star Trek: Voyager 099 - Nowhere Man Tuesday 100 - Deadly Games 101 - Live Shot Saturday 115 - UPN Movie (There Goes the Neighborhood) Appendix B Content Analysis Coding Instrument Program Code Sheet A. Program code (code the number of the promotional spot's program title as listed on pages 5-8 in the coding instruction book) ____________ B. Station code 1 ABC - WIXT (Channel 9) 2 CBS - WTVH (Channel 5) 3 FOX - WSYT (Channel 68) 4 NBC - WSTM (Channel 3) 5 UPN - WSNR (Channel 43) ____________ C. Day of Week Promotional Spot Aired 1 Monday 5 Friday 2 Tuesday 6 Saturday 3 Wednesday 7 Sunday 4 Thursday ____________ D. Time Promotional Spot Aired 1 7 PM - 7:59 PM 2 8 PM - 8:59 PM 3 9 PM - 9:59 PM 4 10 PM - 10:59 PM ____________ E. Program Format 1 Action Adventure 2 Comedy 3 Drama 4 News 5 Other (SPECIFY ____________________________) ____________ F. Gender of Voice-Over Announcer 1 Male 2 Female 3 No Announcer 4 Other (SPECIFY ____________________________) ____________ Character Descriptive Code Sheet A. Character number ____________ B. Character name or description ____________________ C. Character Role 1 Major 2 Minor ____________ D. Gender 1 Male 2 Female ____________ E. Ethnicity 1 Caucasian 2 African American 3 Hispanic 4 Asian 5 American Indian 6 Other (SPECIFY ____________________________) ____________ F. Approximate age 1 Child 4 Middle-aged adult 2 Teenager 5 Elderly adult 3 Young adult 6 Indeterminate ____________ G. Provocative Attire 1 Very provocative 2 Somewhat provocative 3 Neutral 4 Somewhat unprovocative 5 Not provocative ____________ H. Attractiveness 1 Very attractive 2 Somewhat attractive 3 Average 4 Somewhat unattractive 5 Very unattractive ____________ I. Body Type/Physical Fitness 1 In good shape 2 In somewhat good shape 3 Average 4 Somewhat out of shape 5 Very out of shape ___________ J. Hair Color 1 Brown 2 Black 3 Blonde 4 Red/Auburn 5 Gray/White 6 Bald 7 Other (SPECIFY ____________________________) ____________ Appendix C Stereotyping Expert Questionnaire Stereotyping has been defined as making broad generalizations about a group of people based on a very narrow assumption about that group's role or identity in society. Seeing certain groups of people performing stereotypical activities may cause reinforcement of these stereotypes. For example, seeing women on television as always in a traditional, domestic role in the home may support the stereotype that "a woman's place is in the home." Or seeing young, attractive women on television who seem to decorate the program more than advance the plot may reinforce the stereotype of women as "sex object." Listed below are various ways to describe female television characters. For each question, give your opinion if this description is: 1 = not at all stereotypical 2 = somewhat not stereotypical 3 = neutral 4 = somewhat stereotypical 5 = very stereotypical __________________________________________________________________ The way the female character is generally dressed: Very Provocative 1 2 3 4 5 Somewhat Provocative 1 2 3 4 5 Average 1 2 3 4 5 Somewhat Unprovocative 1 2 3 4 5 Not Provocative 1 2 3 4 5 __________________________________________________________________ KEY: 1 = not at all stereotypical 2 = somewhat not stereotypical 3 = neutral 4 = somewhat stereotypical 5 = very stereotypical _________________________________________________________________ The general attractiveness of the female character: Very Attractive 1 2 3 4 5 Somewhat Attractive 1 2 3 4 5 Average 1 2 3 4 5 Somewhat Unattractive 1 2 3 4 5 Very Unattractive 1 2 3 4 5 __________________________________________________________________ __________________________________________________________________ How physically fit the female character generally is: In good shape 1 2 3 4 5 In somewhat good shape 1 2 3 4 5 Average 1 2 3 4 5 Somewhat out of shape 1 2 3 4 5 Very out of shape 1 2 3 4 5 __________________________________________________________________ KEY: 1 = not at all stereotypical 2 = somewhat not stereotypical 3 = neutral 4 = somewhat stereotypical 5 = very stereotypical __________________________________________________________________ The hair color of the female character generally is: Brown 1 2 3 4 5 Black 1 2 3 4 5 Blonde 1 2 3 4 5 Red/Auburn 1 2 3 4 5 Gray 1 2 3 4 5 __________________________________________________________________ __________________________________________________________________ Age or how old the female character generally is: Child 1 2 3 4 5 Teenager 1 2 3 4 5 Young Adult 1 2 3 4 5 Middle Aged 1 2 3 4 5 Elderly 1 2 3 4 5 __________________________________________________________________ [1] These three new television networks are Fox, Paramount, and Warner Brothers. Due to its lack of representation in the market studied for this analysis and significantly smaller number of affiliates nationwide than UPN, the Warner Brothers Network has been excluded from this study. [2] Representatives from each network were contacted directly in order to identify specific target audiences. Definitions of network target audiences, therefore, were based on network sources, popular and trade press articles. [3] The Warner Brothers Network has been excluded from this study for the reasons discussed previously. Due to the non-commercial nature of public television, the PBS affiliate was also excluded in the sample. Because a public television station does not t arget certain audiences in order to sell commercial advertising, it was deemed inappropriate for inclusion in the study. [4] pi = % observed agreement - % expected agreement 1 - % expected agreement [5] The Scott's pi reliability results for each variable were: promo format = .91, promo announcer gender = .92, character role = .83, character gender = .95, character race = .93, character age = .95, character attire = .87, character attractiveness = .67, character physical fitness = .72, character hair color = .82. [6] Although 12% of female ethnic portrayals on FOX were coded in the "other" category, most of these represent cartoon characters (e.g., The Simpsons) of indeterminate race. [7] An overall stereotypical index scale was also computed yielding a reliability coefficient standardized item alpha of .66, which was deemed too low for inclusion in the analysis.