Content-Type: text/html Fall Colors 2000 - Fall Colors 2000 - FALL COLORS 2000: THE STATE OF DIVERSITY IN BROADCAST NETWORK PRIME TIME TELEVISION by Katharine E. Heintz-Knowles, Ph.D. Children's Media Research and Consulting 24124 N.E. Sixth Place Sammamish, Washington 98074 [log in to unmask] And Jennifer Henderson, Ph.C. University of Washington School of Communications Box 353740 Seattle, Washington 98195 [log in to unmask] August 2001 Paper presented to the Entertainment Studies Interest Group at the annual meeting of the Association for Educators in Journalism and Mass Communication, Washington, D.C., FALL COLORS 2000: THE STATE OF DIVERSITY IN BROADCAST NETWORK PRIME TIME TELEVISION INTRODUCTION[1] The 1999-2000 prime time TV season became the subject of considerable controversy after NAACP President Kweisi Mfume, in a July 1999 speech, described the new fall season as a "virtual whitewash" and threatened a boycott of network television. Mfume called the 1999-2000 season "a shameful display by network executives, who are either clueless, careless, or both."[2] Studies of network programming from that season confirmed comments of critics like Mfume that prime time network programming was overwhelmingly a white world, with people of color appearing in mostly secondary and guest roles. One study discovered that eighty percent of characters appearing on prime time network entertainment were white, African Americans made a visible presence (13% of characters), and characters from other racial groups were virtually nonexistent (Heintz-Knowles, 2000). The Fall 1999 season then, was the continuation of a trend, topping off, "a steady decline in the percentage of ethnic minoritie s on prime time" since the beginning of the 1990.[3] Mfume's concern of television's "whitewash" is not a new one. The pattern of under and misrepresentation of minority groups during prime time is not simply a problem of the 1990s. It has persisted in varying forms since the beginning of television. MacDonald's, Blacks and White TV: African Americans in Television Since 1948, describes the stereotypes, limited roles and attempts at assimilation on prime time television during its first forty years.[4] Baughman explains that blacks were "rarely seen" on network programming until the mid-1960s" when shows such as I Spy and Julia appeared.[5] Not only do people of color appear infrequently on prime time television, researchers have also found these limited portrayals to be inaccurate, derived primarily from a white perspective. Gates contends there is little commonality between "the social status of Black Americans and the fabricated images of black people Americans consume everyday."[6] Gray explains, "Whiteness is the privileged yet unnamed place from which to see and make sense of the world_blackness simply works to reaffirm, shore up and police the cultural and moral boundaries of the existing social order."[7] Researchers have long contended that messages from entertainment fare become part of our belief systems about the world around us. Television is a socializing force, a "common storyteller" that prescribes what is "normal" and what is outside of the mainstream.[8] As Hall explains, television images are significant as they produce and perpetuate a social ideology, one that becomes our "taken for granted_world of common sense."[9] The lack of realistic portrayals on television, then, can have a substantially negative effect on all viewers in a multicultural society as television "amplifies real-life stereotypes."[10] As Gordon notes, "the most well-adjusted people are those who have some realistic perspective about what other groups are like." It is also "important for people to see others like themselves on TV. It validates you."[11] As "black viewers are heavy consumers of all television,"[12] watching "50 percent more television than any other group,"[13] the absence of r ealistic black characters on TV is even more troubling. In response to pressure from the NAACP, network executives promised to increase the racial diversity in program casts and on writing staffs as early as the 2000-2001 season. According to an article in the October 2000 issue of Ebony, "there is more color in TV land this fall" but Mfume contends that it's "not enough_While there is more diversity in the [upcoming] fall TV season, we still really have a long, long way to go."[14] Yvette Lee Bowser, executive producer and creator of WB's For Your Love, says the portrayals of African Americans on TV are getting worse in the Fall 2000 season, not better: "We are taking giant steps backward because [Black] images are limited and regurgitated."[15] Fredrick L. McKissak Jr., agrees with Bowser noting, "Blacks are increasingly pigeonholed in simpleminded comedies."[16] With these contradictory assessments of the most recent television season in hand, this study set out to determine the current state of diversity on prime time broadcast network television by examining the following research questions: 1. Does level of racial diversity vary by genre? 2. Does level of racial diversity vary by network? 3. Does level of racial diversity vary by time program airs? 4. Does level of racial diversity vary by gender? 5. Does the level of racial diversity vary by occupation? METHODOLOGY This study examined the first two episodes of each prime time entertainment series airing in Fall 2000 on the six broadcast networks. "Entertainment series" was defined as fictional programming in a serial format. The sample did not include sports programming, news magazines, game shows, made-for-TV movies or specials. Programs in the sample aired between 8 - 11 p.m. PST Monday through Saturday and 7 - 11p.m. PST Sunday. Programs were recorded between September 4 and November 12, 2000 by the two authors. Since this project focused on 2000 series programming, only first run episodes were recorded. All content was subjected to two levels of analysis: ù Macro-level analysis which examined the following program characteristics: 1. Genre: defined as situation comedy, drama (other than science fiction), science fiction, variety, and other. In cases where programs could be identified as more than one type of program (i.e. NBC's Ed and Fox's Ally McBeal are comedy-drama hybrids), coders referred to network promotional material to decide how to code for genre. 2. Network 3. Time program starts 4. Program length 5. Format: defined as live action, animated, claymation, or some combination of live action and animation/puppets 6. Age-Based Rating: TV-G, TV-PG, TV-14/TV-M: as determined by networks and presented on screen in the beginning of each program 7. Content Rating: V (violence), D (suggestive dialog), S (sex), L (harsh language): as determined by networks and presented on screen in the beginning of each program ù Micro-level analysis which identified each speaking character and examined the following characteristics: Character Role: defined as primary recurring if character appears in opening credit sequence and is integral to the plot; secondary recurring if character does not appear in opening credit sequence but appears regularly and contributes to the plot; primary non-recurring if character does not appear regularly but is vital to one episode or plot; secondary non-recurring if character does not appear regularly but plays a supporting role in one episode; tertiary if character does not appear regularly and speaks but does not move the plot. Age: defined as child (pre-pubescent); adolescent (approx. 13-18); young adult (approx. 19-29); adult (approx. 30-39); settled adult (approx. 40-49); older adult (approx. 50-69); elderly (over 70). Social/life experiences were used to help determine approximate age for most adult characters. For example, a character who is established in his/her career and has a young family was identified as an "adult" while a character who is shown just starting out in a career was identified as a "young adult." Gender: for non-human characters (i.e. aliens, borgs), physical markers, clothing, and/or verbal identification was used for coding purposes Race: defined as White/Caucasian; African American; Asian Pacific American; Native American; Latino/Hispanic; Multiracial (if specific information given); Apparent racial minority (if race was unidentifiable); Other as determined primarily by physical features such as skin color, hair type, and eye shape and secondarily by character name and verbal identification. Note: race refers to character's race, and not actor's race Marital Status: defined as single/apparently never married; currently married; separated or divorced; widowed as determined through observation and verbal identification. Characters identified as children or adolescents were not coded for marital status, nor were they included in analysis of this data. Parental Status: defined by clear depiction of parental responsibility for dependent children Occupation: identified from observation or verbal identification using an exhaustive set of categories developed by the authors in previous studies of prime time characters. Disabled: defined as character being clearly shown as exhibiting a chronic disability (physical or mental) All content was coded by the authors and one other trained coder. To ensure reliability among coders, ten percent of the sample was coded by each of the coders independently. The percent of agreement between coders was calculated. All variables included in this analysis received a level of agreement of at least 96%. KEY DEFINITIONS Character Role Type All characters with speaking roles were included for analysis in this study. However, all speaking roles are not created equal. To determine if there are differences in the ways characters of different racial groups contribute to the story lines of the sample episodes, each character was classified according to his/her importance to the plot and frequency of occurrence. The following character role types describe the contributions of different characters to the story lines of program episodes: ù Primary Recurring: Actor appears in opening cast credits, is involved in the primary story line of most episodes. ù Primary Non-Recurring: Actor does not appear in opening cast credits, is a guest star involved in the primary story line of one or two episodes (clearly identified as a non-recurring character) ù Secondary Recurring: Actor does not appear in opening cast credits but appears regularly on the series and contributes to story line development ù Secondary Non-Recurring: Actor does not appear in opening cast credits but plays a supporting guest role ù Tertiary: Character with minor speaking part who does not affect the outcome of the story line. For example, on the NBC hit, Law and Order, the characters may be defined as: Character Role Type Actor Character Description Primary Recurring Jesse L. Martin Ed Greene One of two primary detectives; appears in opening credits cast. Primary Non-Recurring Megan Follows Megan Parnell Mother accused of murdering her disabled son; guest stars in one episode Secondary Recurring J.K. Simmons Dr. Emil Skoda Psychiatrist who appears regularly to consult on defendants' mental health. Appears in secondary credits. Secondary Non-Recurring Brian Delate Colin Parnell Husband of accused murderer; plays secondary role in one episode Tertiary Rita Gardner Jury Foreperson One line reading the verdict. Program Character Set The Character Role Types allow us to examine the level of diversity in different subgroups in the sample. For this report, analyses will be conducted on the following Program Character Sets: ù Entire Cast of Characters: Includes all the speaking characters appearing in the two episodes examined ù Plot-Relevant Characters Only: Includes primary and secondary characters only; excludes characters identified as tertiary who did not contribute to the development of the story line ù Recurring Characters Only: Includes all primary and secondary characters who have recurring roles ù Primary Recurring Characters Only or Opening Credits Cast: Includes the main characters who appear in the programs' opening credits sequences Program Diversity Index In addition to looking for diversity across the sample as a whole, this study aims to assess the level of diversity within individual programs. Each program in the sample was evaluated and labeled for Program Diversity according to the following definitions: ù All White: All Characters in the program character set identified racially as "white" ù All Black: All Characters in the program character set identified racially as "African American" ù Only 1: All characters in the program character set but one identified as belonging to the same racial group ù Mixed: Program character set includes a mix of racial and ethnic characters that is not all white, all black, or Only 1. DESCRIPTION OF THE SAMPLE This method of data collection yielded 2251 characters across 176 episodes of 88 different programs. Table 1 shows the number of male and female characters in the entire sample. A list of the programs sampled is included in Appendix A. TABLE 1 GENDER OF CHARACTER Gender Total Sample (n=2251) Primary Recurring Characters (n=564) Frequency Percent Frequency Percent Male 1462 65 325 58 Female 789 35 239 42 Total 2251 100 564 100 The following tables show the distribution of programs by network and genre. CBS and NBC aired the largest number of programs in the sample and featured the largest number of characters. The smallest number of programs and characters came from the UPN network. TABLE 2 DISTRIBUTION OF PROGRAMS AND CHARACTERS BY NETWORK Network Number of Programs Number of Characters ABC 12 227 CBS 18 541 NBC 19 574 FOX 13 341 WB 17 365 UPN 9 203 Total 88 2251 Nearly half of the programs in the sample were situation comedies, yet the majority of characters appeared in dramatic programs. This discrepancy is due to the substantially larger casts in most dramas than situation comedies. TABLE 3 DISTRIBUTION OF CHARACTERS BY GENRE Genre Number of Programs Number of Characters Situation comedy 43 757 Drama 38 1347 Science Fiction 5 125 Variety 2 22 Total 88 2251 A majority of characters in the sample played non-recurring roles. The most frequently occurring type was tertiary. TABLE 4 FREQUENCY OF CHARACTER ROLE TYPES Character Role Type Frequency Percent of total sample Primary Recurring 564 25 Primary Non-Recurring 178 8 Secondary Recurring 338 15 Secondary Non-Recurring 359 16 Tertiary 812 36 Total 2251 100 RESULTS The world of prime time entertainment television is overwhelmingly populated by white adults and children. As Table 4 indicates, three-fourths of the characters in the sample (n=1687) are white. African Americans make up roughly 17% of the sample (n=383). Less than one percent of characters were Native American (n=5, 0.2%). This picture does not change when examining the race of primary recurring characters only. TABLE 5 RACE OF CHARACTER Race Total Sample (n=2251) Primary Recurring Characters Only (n=564) Frequency Percent Frequency Percent White 1688 75 431 76 African American 383 17 103 18 Asian/Pacific Islander 61 3 11 2 Native American 5 0.2 1 0.2 Latino, Hispanic 47 2 9 2 Other 44 2 5 1 Don't know/Can't tell 23 1 4 1 Total 2251 100 564 100 NETWORKS Racial diversity is not equivalent across the six networks. When examining diversity of all characters in the sample, UPN had the highest proportion of nonwhite characters (41%) while ABC had the lowest (17%). The racial picture changes only slightly when examining all recurring characters or primary recurring characters only. UPN programs still feature the most nonwhite characters (approx. 40%). The most interesting change occurs on CBS when non-recurring characters are excluded from the analysis. The proportion of African American characters on that network jumps from 16% of the overall character composition to 25% of the primary recurring character composition. TABLE 6 RACIAL DIVERSITY OF ALL CHARACTERS BY NETWORK (N=2251) (percent of each network's population) Network White African American Asian/Pacific Islander Native American Latino/ Hispanic Other* ABC 83% 13% 2% 0.4% 1% 0.4% CBS 76% 16% 2% 0.2% 4% 1% NBC 79% 14% 3% - 2% 2% Fox 73% 13% 3% 1% 3% 8% WB 72% 20% 2% - 1% 5% UPN 59% 33% 5% 0.5% 0.5% 3% *includes non-humans with ambiguous race TABLE 7 RACIAL DIVERSITY OF RECURRING CHARACTERS BY NETWORK (N=903) (percent of each network's population) Network White African American Asian/Pacific Islander Native American Latino/ Hispanic Other* ABC 81% 14% 1% - 3% 1% CBS 71% 26% 1% - 3% 1% NBC 80% 13% 4% - 2% 0.5% Fox 81% 11% 3% 1% 1% 4% WB 76% 20% 2% - 2% 1% UPN 62% 32% 4% - - 2% *includes non-humans with ambiguous race TABLE 8 RACIAL DIVERSITY OF PRIMARY RECURRING CHARACTERS BY NETWORK (N=564) (percent of each network's population) Network White African American Asian/Pacific Islander Native American Latino/ Hispanic Other* ABC 83% 11% - - 5% 2% CBS 71% 25% - - 3% 1% NBC 83% 12% 3% - 2% - Fox 82% 10% 4% - 1% 4% WB 78% 19% 1% - - 2% UPN 59% 35% 4% - - 3% *includes non-humans with ambiguous race In addition to overall racial diversity on prime time, this study examined the racial diversity within each program. Each program was assigned a rating on the Program Diversity Index (see page 3 for definitions) based on the racial make up of the cast of characters appearing across the two episodes. Each program received four ratings, based on the Character Set (see page 3 for definitions) examined. Diversity in network prime time entertainment most often comes from the inclusion of non-recurring characters. When examining all of the characters in the sample, a vast majority of programs were identified as having mixed casts. 72 of the 88 programs contain mixed casts when all speaking characters are included. However, that number drops dramatically when tertiary (non-plot relevant) characters are excluded. By honing in further on central characters, the number of programs with mixed casts drops to 25 when examining opening credits casts only. Over 40% of the programs (n= 38) have opening credits casts that are either all- white or all- black. TABLE 9 PROGRAM DIVERSITY INDEX FOR OVERALL PRIME TIME PDI Rating Total Cast Primary and Secondary Characters Only (Plot Relevant) Recurring Characters Only Opening Credits Cast Only Mixed 72 54 42 25 Only 1 10 16 24 25 All White 6 16 19 33 All Black - 2 3 5 Total Programs 88 88 88 88 When examining the total cast of characters, there is not much difference between the networks in the level of diversity in individual programs. At least three-fourths of the programs on each network were identified as mixed when total casts were examined. However, as we look at the diversity of central characters, some differences between the networks emerge. Eliminating tertiary characters alone reduces the diversity of casts significantly on two networks - ABC and the WB. While 75% of ABC programs and 72% of WB programs were identified as mixed when looking at entire casts, this percentage drops to less than half for each network when tertiary characters are not included (ABC = 5 of 12 programs; WB= 8 of 17 programs). Focusing further on central characters, we see a decrease in the number of programs on all networks with mixed casts and an increase in the number of programs with all white and all black casts. Again, ABC and the WB feature the fewest programs with mixed casts. Just two programs on ABC (17%) and three programs on the WB (18%) were classified as mixed when the analysis included just the opening credits casts. Half of the programs on ABC (n=6) and 53% of the programs on the WB (n=9) were identified as having racially homogeneous opening credits casts. The UPN network featured the highest proportion of programs with mixed opening credits casts (n=4 of 9, 44% of UPN programs). TABLE 10 PROGRAM DIVERSITY INDEX BY NETWORK Network Mixed Only 1 All White All Black Total ABC Total Cast 9 2 1 - 12 Plot-Relevant 5 3 4 - 12 Recur Only 4 4 4 - 12 Opening Credit 2 4 6 - 12 CBS Total Cast 16 1 1 - 18 Plot Relevant 14 - 4 - 18 Recur Only 9 5 4 - 18 Opening Credit 7 5 6 - 18 NBC Total Cast 16 1 2 - 19 Plot Relevant 12 3 4 - 19 Recur Only 10 5 4 - 19 Opening Credit 5 7 7 - 19 FOX Total Cast 11 1 1 - 13 Plot Relevant 9 2 2 - 13 Recur Only 7 4 2 - 13 Opening Credit 5 2 6 - 13 WB Total Cast 13 4 - - 17 Plot Relevant 8 6 2 1 17 Recur Only 6 6 4 1 17 Opening Credit 3 4 5 3 17 UPN Total Cast 7 2 - - 9 Plot Relevant 6 2 - 1 9 Recur Only 5 2 - 2 9 Opening Credit 4 3 - 2 9 GENRE With the exception of Native American characters, the majority of characters from each racial group appear in dramatic programs when the entire cast is analyzed. When we hone in on central characters only, the picture changes somewhat. If we consider science fiction programming (i.e. Star Trek: Voyager, UPN; Dark Angel, Fox) to be dramatic in nature, then the majority of primary recurring characters from each racial group except African American appear in dramatic programming. African American characters appeared with equal frequency in dramatic and comedy programming. Primary recurring characters from all other minority racial groups are far more likely to appear in dramatic programs than comedies. TABLE 11 RACIAL DIVERSITY OF ALL CHARACTERS BY GENRE (N=2251) (percent of each racial group) Genre White (n=1687) African American (n=383) Asian/Pacific Islander (n=61) Native American (n=5) Latino/ Hispanic (n=47) Situation Comedy 32% 41% 26% 60% 28% Drama 61% 55% 64% 20% 72% Science Fiction 6% 4% 10% 20% - Variety 1% 1% - - - Total 100% 100% 100% 100% 100% TABLE 12 RACIAL DIVERSITY OF RECURRING CHARACTERS BY GENRE (N=903) (percent of each racial group) Genre White (n=685) African American (n=166) Asian/Pacific Islander (n=21) Native American (n=2) Latino/ Hispanic (n=16) Situation Comedy 39% 42% 29% 50% 25% Drama 53% 54% 57% - 75% Science Fiction 6% 2% 14% 50% - Variety 2% 2% - - - Total 100% 100% 100% 100% 100% TABLE 13 RACIAL DIVERSITY OF PRIMARY RECURRING CHARACTERS BY GENRE (N=564) (percent of each racial group) Genre White (n=431) African American (n=103) Asian/Pacific Islander (n=11) Native American (n=1) Latino/ Hispanic (n=9) Situation Comedy 40% 48% 27% 22% Drama 53% 49% 46% 78% Science Fiction 5% 1% 27% 100% - Variety 2% 3% - - Total 100% 100% 100% 100% 100% There is a great disparity in levels of diversity for programs of different genres. Dramas are more than three times as likely as situation comedies to feature mixed primary casts (45% of dramas vs.14% of situation comedies). When examining total casts, all but one of the dramas were identified as mixed (n=37, 97%) while 70% of the situation comedies were so identified (n=30). For nearly 40% of the situation comedies (n=16, 37%), the story lines were moved entirely by characters of the same racial group (n=14 programs identified as all white and 2 programs identified as all black). For 10 of these programs (63%), characters of a different racial group appeared only in tertiary roles, often with just one line. Nearly two-thirds of the situation comedies (n=28, 65%) feature opening credits casts that are racially homogeneous. Fourteen percent of the situation comedies (n=6) did not contain a single character of color in a speaking role in either of the episodes examined. Four out of five of the science fiction programs were identified as mixed when examining the total cast of characters, but this number dropped to just 1 in 5 (Star Trek: Voyager, UPN) when examining opening credit casts only. The two variety programs in the sample were split between having a mixed cast (Hype, Fox) and a cast with only one racial minority character (Whose Line is it Anyway?, ABC). TABLE 14 PROGRAM DIVERSITY INDEX BY GENRE Genre Mixed Only 1 All White All Black Total SitCom Total Cast 30 7 6 - 43 Plot-Relevant 17 10 14 2 43 Recur Only 14 10 16 3 43 Opening Credit 6 9 23 5 43 Drama Total Cast 37 1 - - 38 Plot Relevant 32 5 1 - 38 Recur Only 25 11 2 - 38 Opening Credit 17 13 8 - 38 Sci Fi Total Cast 4 1 - - 5 Plot Relevant 4 - 1 - 5 Recur Only 2 2 1 - 5 Opening Credit 1 2 2 - 5 Variety Total Cast 1 1 - - 2 Plot Relevant 1 1 - - 2 Recur Only 1 1 - - 2 Opening Credit 1 1 - - 2 PROGRAM AIR TIME Another way to assess diversity is to look at all programs across the TV dial at the same hour. We can ask the question: how diverse is the world pictured on prime time network entertainment from 8-9pm? 9-10pm? When we examine program diversity by program hour, an interesting pattern emerges. Program diversity dramatically increases as the evening goes on. While 73% (n=29) of the programs on during the 8 o'clock hour are defined as mixed when considering the total cast, just 13% of those programs (n=5) are so classified when just opening credit casts are included in the analysis. Indeed, over half of all programs airing in the 8 o'clock hour (n=22, 55%) were classified as either all white or all black when just opening credit casts were analyzed. The picture becomes slightly more diverse during the 9 o'clock hour, with 85% of programs (n=27) identified as mixed when considering the total cast, and nearly one-third of the programs (n=10, 31%) classified as such when examining exclusively the opening credit casts. Still, at this hour, more programs featured opening credit casts that were either all-white or all-black (n=13, 41%) than racially mixed. The 10 o'clock hour is the most diverse time on prime time entertainment. Every program analyzed at this hour was classified as mixed when examining both total cast and plot relevant characters only. When the analysis was narrowed to recurring characters and opening credit cast only, the level of diversity dropped slightly. However, at this hour, two-thirds of the programs (n=8, 67%) were identified as having mixed opening credit casts. The few programs that aired during the 7 o'clock hour on Sundays were all classified as mixed when examining total casts and were evenly split between mixed and homogeneous (either all white or all black) when examining just opening credit casts. TABLE 15 PROGRAM DIVERSITY INDEX BY TIME PROGRAM AIRS Program Air Time Mixed Only 1 All White All Black Total 7-8pm Total Cast 4 - - - 4 Plot-Relevant 3 - - 1 4 Recur Only 2 - 1 1 4 Opening Credit 2 - 1 1 4 8-9pm Total Cast 29 8 3 - 40 Plot Relevant 19 12 9 - 40 Recur Only 14 14 11 1 40 Opening Credit 5 13 19 3 40 9-10pm Total Cast 28 1 3 - 32 Plot Relevant 20 5 6 1 32 Recur Only 15 9 7 1 32 Opening Credit 10 10 11 1 32 10-11pm Total Cast 12 - - - 12 Plot Relevant 12 - - - 12 Recur Only 10 2 - - 12 Opening Credit 8 3 1 - 12 GENDER Consistent with most studies of television content, this sample consisted of more male than female characters (see Table 1). The racial diversity of the sample of female characters reflects the diversity in the total sample. TABLE 16 RACE OF FEMALE CHARACTERS Race Frequency Percent White 584 74 African American 145 18 Asian/Pacific Islander 25 3 Native American - 0.1 Latina/Hispanic 18 2 Other 15 2 Don't Know 2 0.3 Total 789 100 Female characters are slightly more likely to be included in guest or tertiary roles than in recurring roles. However, female recurring characters are almost twice as likely to play primary roles. TABLE 17 ROLE TYPE OF FEMALE CHARACTERS Role Type Frequency Percent Primary Recurring 239 30 Primary Non-Recurring 63 8 Secondary Recurring 133 17 Secondary Non-Recurring 117 15 Tertiary 237 30 Total 789 100 If we compare the role types of males and females of different races, we see that females of all races are more likely than males to play recurring roles. This difference is greatest for Asian/Pacific Islander characters. More than half of the female Asian characters played recurring roles, while less than a quarter of male Asian characters did. Male Asian characters were overwhelmingly more likely to be included in tertiary roles than in any other role type. Latina recurring characters were more likely to play secondary roles than primary roles. Female recurring characters of all other races were more likely to play primary roles than secondary. TABLE 18 RACE AND ROLE TYPE OF FEMALE CHARACTERS (N=789) (percent of each racial group) Race Role Type Primary Recurring Primary Non-Recurring Secondary Recurring Secondary Non-Recurring Tertiary White (n=584) 31 8 16 16 30 African American (n=145) 31 9 20 13 27 Asian/Pacific American (n=25) 32 4 24 4 36 Latina(n=18) 17 17 21 12 33 Other/Don't Know (n=10) 12 6 6 18 59 TABLE 19 RACE AND ROLE TYPE OF MALE CHARACTERS (n=1462) (percent of each racial group) Race Role Type Primary Recurring Primary Non-Recurring Secondary Recurring Secondary Non-Recurring Tertiary White (n=1104) 23 8 15 17 38 African American (n=238) 24 7 14 15 40 Asian/Pacific American (n=36) 8 11 11 14 56 Native American (n=5) - 40 - 20 20 Latino (n=29) 21 7 10 21 41 Other/Don't Know (n=50) 14 8 6 26 46 OCCUPATIONS Television adults tend to engage in professional careers with a good deal of freedom. TV adults were most often employed as professionals (i.e. architects, stock brokers, computer analysts). Among the top ten occupations of TV adults[17] were entertainer/performer/artist, physician, and small business owner. Nine percent of the TV adult population was employed in law enforcement at all levels (from beat cop to detective to chief of police) and eight percent were attorneys. Table 23 shows the top occupations for the adult characters in the sample. TABLE 20 TOP OCCUPATIONS FOR TV ADULTS WITH IDENTIFIABLE OCCUPATIONS (N=1659) Occupation Frequency Percent Professional 168 10 Law Enforcement 147 9 Attorney 127 8 Service, Retail, Restaurant 129 8 Physician 92 6 Entertainer/Performer/Artist 102 6 Criminal 75 5 Student 67 4 Small Business Owner/Manager 66 4 Clerical 62 4 White characters were more likely than African American characters to be shown in professional occupations, and African Americans were shown more often than whites in law enforcement occupations. Both whites and African Americans appeared with about equal frequency as physicians, attorneys, and in service/retail/restaurant jobs. The top 10 occupations for female characters showed a mix of both traditionally female occupations and non-traditional occupations. Females in the sample were shown working as professionals, attorneys, and physicians with about equal frequency as they were shown working as nurses, secretaries, homemakers, and teachers. TABLE 21 TOP OCCUPATIONS FOR TV ADULTS BY RACE AND GENDER (n=1659) All White Adults (n=1497) All African American Adults (n=347) All Male Adults (n=1289) All Female Adults (687) Professional (9%) Law Enforcement (11%) Professional (9%) Professional (7%) Attorney (7%) Service (7%) Attorney (7%) Attorney (6%) Law Enforcement (7%) Attorney (6%) Service (6%) Nurse/Physician's Assistant (6%) Entertainer (6%) Physician (6%) Physician (5%) Service (6%) Service (6%) Nurse/Physician's Asst. (6%) Entertainer (5%) Clerical (6%) Criminal (4%) Student (5%) Criminal (5%) Homemaker (5%) Physician (4%) Professional (5%) Manager/Small Business Owner (4%) Entertainer (5%) Executive/CEO (3%) Clerical (4%) Executive/CEO (3%) Student (5%) Journalist/Media Professional (3%) Manager/Small Business Owner (4%) Military (3%) Physician (4%) Clerical (3%) Teacher (3%) Craftsman (3%) Journalist/Media Professional (4%) Manager/Small Business Owner (3%) Entertainer (3%) Security (3%) Teacher (4%) Paramedic/Fire Fighter (3%) Security (3%) To assess the relative importance of characters appearing in different occupational categories, the data were analyzed by Character Role Type. This allows us to see if there is a difference in the occupations shown for primary and secondary, recurring and non-recurring characters. White recurring characters most frequently cast appeared as business professionals, while African American recurring characters were most frequently shown working in health care or law enforcement. Primarily due to the introduction of two new hospital series (City of Angels (CBS), Gideon's Crossing (ABC)), the most common occupations for African American recurring adult characters were physicians and their assistants. The second most common occupation was in law enforcement at all levels, from patrol cop to undercover officer to detective to chief of detectives. The few Asian and Latino recurring characters were shown in a variety of occupations from high status careers like physician and attorney to lower status positions as domestic servants, secretaries, or homemakers. Tables 26 through 29 show the top occupations for male and female adult characters of each racial group by role type. TABLE 22 TOP 5 OCCUPATION S FOR WHITE ADULTS (N= 1497) Primary Recurring (n=371) Primary Non-Recurring (n=118) Secondary Recurring (n=215) Secondary Non-Recurring (n=250) Tertiary (n=543) Professional (13%) Criminal (18%) Professional (14%) Attorney (13%) Service (8%) Attorney (9%) Executive/CEO (10%) Clerical (8%) Criminal (7%) Professional (7%) Professional Wrestler (7%) Professional (9%) Student (6%) Professional (6%) Entertainer (7%) Law Enforcement (7%) Law Enforcement (6%) Service (6%) Law Enforcement (6%) Law Enforcement (7%) Entertainer (6%) Attorney (5%) Physician (5%) Service (6%) Attorney (5%) TABLE 23 TOP 5 OCCUPATION S FOR AFRICAN AMERICAN ADULTS (N= 347) Primary Recurring (n=97) Primary Non-Recurring (n=25) Secondary Recurring(n=55) Secondary Non-Recurring (n=48) Tertiary (n=122) Physician (12%) Attorney (12%) Nurse/Physician's Assistant (16%) Law Enforcement (17%) Nurse/Physician's Assistant (15%) Law Enforcement (10%) Criminal (12%) Law Enforcement (9%) Service (12%) Law Enforcement (13%) Student (9%) Professional (8%) Physician (7%) Attorney (10%) Service (7%) Professional (8%) Service (8%) Student (7%) Nurse/Physician's Assistant (6%) Paramedic/Fire Fighter (6%) Attorney (7%) Military (4%) Service (7%) Student (4%) Entertainer (5%) Teacher (7%) Retired (4%) Clerical (7%) Criminal (4%) Security (4%) TABLE 24 TOP 5 OCCUPATION S FOR ASIAN/PACIFIC AMERICAN ADULTS (N= 55) Primary Recurring (n=9) Primary Non-Recurring (n=5) Secondary Recurring(n=7) Secondary Non-Recurring(n=6) Tertiary(n=28) Professional (n=3) Manager/Small Business Owner (n=1) Physician (n=1) Service (n=2) Service (n=5) Attorney (n=1) Entertainer (n=1) Nurse/Physician's Assistant (n=1) Professional (n=1) Law Enforcement (n=4) Physician (n=1) Criminal (n=1) Military (n=1) Security (n=1) Physician (n=3) Service (n=1) Clerical (n=1) Law Enforcement (n=1) Teacher (n=2) Student (n=1) Domestic Servant (n=1) Student (n=2) Homemaker (n=1) Student (n=1) Criminal (n=2) Homemaker (n=1) TABLE 25 TOP 5 OCCUPATION S FOR LATINO ADULTS (N= 43) Primary Recurring (n=8) Primary Non-Recurring (n=4) Secondary Recurring(n=6) Secondary Non-Recurring(n=9) Tertiary(n=16) Law Enforcement (n=2) Criminal (n=1) Physician (n=1) Criminal (n=2) Service (n=3) Paramedic/Fire Fighter (n=1) Unskilled Laborer (n=1) Manager/Small Business Owner (n=1) Service (n=2) Entertainer (n=2) Professional (n=1) Nurse/Physician's Assistant (n=1) Nurse/Physician's Assistant (n=1) Security (n=1) Physician (n=1) Clerical (n=1) Law Enforcement (n=1) Journalist/Media Professional (n=1) Executive/CEO (n=1) Domestic Servant (n=1) Unskilled Laborer (n=1) Law Enforcement (n=1) Clerical (n=1) Domestic Servant (n=1) Unskilled Laborer (n=1) Student (n=1) THE STATE OF DIVERSITY ON PRIME TIME BROADCAST TELEVISION Prime time broadcast television does not reflect the diversity that is apparent in the world outside the screen. There are fewer Latinos, Asian Americans, and Native Americans than in the general population, especially among the child characters. The world of prime time network entertainment is primarily a white world. Whites made up _ of the sample population. African Americans were a visible presence, but all other racial minority groups were virtually invisible. These proportions remained nearly the same when examining female and youth subpopulations. When programming does include people of color, it frequently does so in an exclusionary manner. The Program Diversity Index - measuring the level of diversity within individual programs - showed that most programs feature primary casts that are either all-white or all-black. There was a striking reduction in programs labeled mixed as we honed in on the central characters in the programs. Eighty-two percent of programs (n=72) were identified as mixed when examining total casts. However, that number drops substantially when we examine plot-relevant characters only - fifty-four programs, just 61% of the sample were identified as mixed when tertiary characters were eliminated from the analysis. And looking at opening credits casts only, the Program Diversity Index identified just 28% (n=25) of the programs as having mixed casts. By contrast, over 40% (n=38) of the programs were identified as having opening credits casts that were either all white or all black. The racial divers ity in prime time, then, comes in the form of secondary and guest characters. Network All major television networks present more images of whites than people of color in their prime time programming. Some networks, however, are doing a much better job of providing audiences with diverse character roles in their shows. UPN's prime time programming contained the greatest percentage of people of color in primary recurring roles. More than 40% of the primary recurring roles on that network were written for people of color, with 35% going to African Americans. ABC and NBC provided the least diverse programming during prime time when examining primary recurring characters. Each of these two networks presented shows where 83% of the primary recurring casts were white. The Fox Network did not fare much better, with whites comprising 82% of primary recurring characters in their prime time programming UPN also leads the field in racial diversity when all characters are considered, maintaining a 41% people of color ratio. When all character roles are included in analysis, Fox improves character diversity, increasing the percentage of people of color from 18% to 27%. ABC and NBC, however, make little progress, with ABC remaining 83% white and NBC improving only slightly. Genre Racial diversity is found most often in dramatic programming during prime time. A subset of this the drama genre, science fiction, has the greatest level of program diversity in the study. Situation comedies are more likely to feature all-white or all-black casts. Prime time comedies are nearly three times more likely than dramas to have a completely homogenous cast. African Americans were featured more frequently in situation comedy programs than any other racial group. Actors from this racial group appeared with equal frequency in dramatic and comedy programs, while actors from all other racial groups appeared with greater frequency in dramatic programming. Program Time Racial diversity is concentrated in late-evening programs. Therefore, children and viewers of early evening programming are most likely to see programs with all-white casts. Almost one-half (48%) of all programs aired at 8:00 EST have white-only primary recurring casts. An additional 33% of shows during this time period contained only one person of color in the recurring cast. Gender Across all prime time programming, women comprise only 35% of the population. There are half as many white women as white men on prime time, a striking 2:1 ratio. Women of color also lag behind men in prime time appearances with men comprising about 60% of all roles that feature people of color. Of the 2251 roles on prime time last season, only 198 (8.8%) were written for women of color. Even more disconcerting, no Native American women appeared during the Fall 2000 prime time season. Occupation Primary recurring characters of color generally are employed in high income/high status jobs such as physician, attorney or other professional. When secondary and non-recurring characters of color are considered, however, stereotypes emerge. Of the top five occupations for all racial groups, only people of color filled roles of domestic worker, homemaker, nurse/physician's assistant and unskilled laborer. Whites, on the other hand, were overwhelmingly cast in high status and professional occupations whether recurring, non-recurring, primary or secondary. WHY THE STATE OF DIVERSITY ON PRIME TIME BROADCAST TELEVISION MATTERS Through patterns of inclusion and exclusion, television content sends implicit messages about the relative cultural importance of different groups, behaviors, and ideologies.[18] Under representation and negative portrayals of different groups on TV can influence the self-concepts and images of viewers from these groups. These portrayals can influence the attitudes and beliefs about such groups among members of the general public. Even when viewers recognize that the content they are viewing is fictional, its messages and images gradually shape expectations and beliefs about the real world.[19] Since children are more likely to watch television in the earlier hours, then the type of world presented to them on network prime time entertainment is a very homogenous one. The central characters who reappear week after week, and who are involved in the primary plots, tend to be surrounded by people of the same racial group as themselves. Some of the most popular programs among young viewers airing during early prime time feature primary casts that are either all-white (Sabrina, Popular, The Simpson's, Dawson's Creek, The Drew Carey Show) or all-black (Moesha, The Steve Harvey Show, The Jamie Foxx Show). Clearly, the diversity that exists in young viewers' real worlds is not reflected on their TV screens. The Cultivation hypothesis, developed by Gerbner and his colleagues to explain the effect of violent content on audiences, provides a conceptual framework for the study of television content. The hypothesis posits that heavy viewers of television cultivate perceptions about the "real world" that are based on the television "world" they see. In other words, heavy viewers are more likely to believe that television content imitates and represents the world at large.[20] The world presented in the Fall 2000 television season is substantially different than that of real world, consisting primarily of whites, men, and people in professional occupations. People of color in this world appear and disappear, but rarely stay. When they do appear, they often occupy lower income or lower status occupations. The reality proposed by prime time television virtually ignores the contributions of, and in many ways virtually ignores, people of color. For example, of the total prime time population, only 3 % of the characters are Asian Pacific American, only 2% Latino or Hispanic and only .2% Native American. The crux of the cultivation hypothesis is that television is ubiquitous and the types of messages are relatively similar and frequently repeated, regardless of program type. Similar stories and situations get played out over and over and the repetition of these messages leads to accumulative effects in audiences. Cultivation effects do not occur immediately or with one television program; they are the result of long-term and heavy viewing. Jefferes explains: "Like the steady drip of a faucet, measurable effects at any one moment may be small, but they accumulate over time."[21] Unfortunately, the Fall 2000 prime time television season has built upon an already poor tradition of minority representation in television. For example, Latinos shown as criminals and unskilled laborers, Asian Americans as small business owners and African Americans cast as comic relief are stereotypes developed years ago and repeated here. TV has also been shown to be a powerful source of behavioral models. Cognitive social learning theory, developed by psychologist Albert Bandura,[22] proposes that social behavior is determined by a combination of mental and physical abilities and desires. We learn new behaviors through observing others perform them. The consequences we observe others receiving influence our willingness to perform the behavior ourselves. Television characters - and the actors who play them - are often taken as role models by viewers, young and old alike. Numerous studies document the influence of television viewing on the aggressive behavior of children and adolescents.[23] Conversely, there is substantial evidence that television characters are effective models of prosocial behavior.[24] Television characters who are attractive, successful, and of high status are more likely to be imitated than those who are unattractive, unsuccessful, or ridiculed.[25] Television programmers are not under any contractual obligation to provide viewers with representative pictures of their world. However, as our culture's primary storytellers, TV creators provide stories and images that contribute to the world views of millions of viewers. When certain groups are privileged in stories and others are excluded, that can send messages to viewers - especially young ones - that these groups are valued differently by the culture. This can, in turn, affect the ways viewers feel about themselves. Such a powerful medium would do well to provide a variety of images for its' diverse audience to learn from and enjoy. SUGGESTIONS FOR FURTHER RESEARCH This study provides valuable information about the state of the cultural landscape in the beginning of the 21st century, but content analyses must be coupled with examinations of both the systems of production and the process of reception to allow for a richer understanding of the role of the entertainment media in the larger cultural life of the people in the U.S Future research should examine the behind-the-scenes construction of these images. Is the lack of diversity on the small screen reflective of the lack of diversity among program creative staffs? What is the level of diversity of decision-makers at the broadcast networks? What are the policies regarding diversity, if any, that exist at the broadcast networks? Future research should also continue to pursue the question of audience preferences and meaning construction. Do audiences prefer to see racially homogeneous casts? Is there a differential impact on viewers' world view when programs feature characters of color in racially homogeneous environments vs. racially mixed environments? APPENDIX A Fall Colors 2000 Program List 1. Ally McBeal 2. Angel 3. Becker 4. Bette 5. Boston Public 6. Buffy the Vampire Slayer 7. C.S.I. 8. Charmed 9. City of Angels 10. Cursed 11. Daddio 12. DAG 13. Dark Angel 14. Dawson's Creek 15. Deadline 16. Dharma and Greg 17. Diagnosis Murder 18. E.R. 19. Ed 20. Everybody Loves Raymond 21. Family Law 22. Felicity 23. For Your Love 24. Frasier 25. Freakylinks 26. Freedom 27. Friends 28. Futurama 29. Gideon's Crossing 30. Gilmore Girls 31. Girlfriends 32. Grosse Pointe 33. Hype 34. JAG 35. Judging Amy 36. Just Shoot Me 37. King of Queens 38. King of the Hill 39. Law and Order 40. Law and Order: Special Victims' Unit 41. Level 9 42. Madigan Men 43. Malcolm in the Middle 44. Moesha 45. Nash Bridges 46. Nikki 47. Normal, Ohio 48. Once and Again 49. Popular 50. Providence 51. Roswell 52. Sabrina, the Teenaged Witch 53. Seven Days 54. Seventh Heaven 55. Spin City 56. Star Trek: Voyager 57. That 70s Show 58. That's Life 59. The District 60. The Drew Carey Show 61. The Fugitive 62. The Geena Davis Show 63. The Hughleys 64. The Jamie Foxx Show 65. The Michael Richards Show 66. The Norm Show 67. The Parkers 68. The PJ's 69. The Practice 70. The Simpsons 71. The Steve Harvey Show 72. The Street 73. The Trouble with Normal 74. The West Wing 75. The X-Files 76. Third Rock from the Sun 77. Third Watch 78. Titans 79. Titus 80. Touched by an Angel 81. Tucker 82. Two Guys and a Girl 83. Walker, Texas Ranger 84. Welcome to New York 85. Whose Line is it Anyway? 86. Will and Grace 87. WWF Smackdown 88. Yes, Dear Bostrom, Margaret (2000). The 21st Century Teen: Public Perspectives and Teen Reality. Report prepared for the Frameworks Institute, Washington, D.C. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with television: The cultivation perspective. In J. Bryant and D. Zillmann (eds.), Media Effects: Advances in Theory and Research (pa. 17-42). Hillsdale, NJ: Erlbaum. [1] This research was funded by a grant to the first author by Children Now, Oakland, CA. [2] Aucoin, Don. "TV Networks Under Fire For Diversity Gap," The Boston Globe, A1. [3] Aucoin, Don. "TV Networks Under Fire For Diversity Gap," The Boston Globe, A1. [4] MacDonald, J. Fred. Blacks and White TV: African Americans in Television Since 1948, 2nd Ed. (Chicago: Nelson-Hall Publishers, 1992). [5] Baughman, James L. The Republic of Mass Culture, 2nd Ed. (Baltimore: The Johns Hopkins University Press, 1997): 109. [6] Gates, Jr., Henry Louis. "TV's Black World Turns - but Stays Unreal," in Race, Class and Gender: An Anthology, Margaret L. Anderson and Patrician Hill Collins, Eds. (Belmont: Wadsworth Publishing, 1992): 316. [7] Gray, Herman. Watching Race: Television and the Struggle for "Blackness." (Minneapolis: University of Minnesota Press, 1997): 71, 86-87. [8] Signorelli, Nancy and Bacue, Aaron. Recognition and Respect: A Content Analysis of Prime-Time Television Characters Across Three Decades. Sex Roles: A Journal of Research. 40 (1999): 527-544. [9] Hall, Stuart. "The Whites of Their Eyes: Racist Ideologies and the Media." In Gail Dines and Jean Humez, Eds, Gender, Race and Class in Media, (Thousand Oaks: Sage, 1995): 18-22. [10] Renn, Jennifer A. and Calvert, Sandra L. "The Relation Between Gender Schemas and Adults' Recall of Stereotyped and Counterstereotyped Television Information." Sex Roles: A Journal of Research. 28 (1999): 449-53. [11] Brand-Willimas and Tiska, Detroit News, July 13, 1999. [12] Matabane, Paula W. "Television and the Black Audience: Cultivating Moderate Perspectives on Racial Integration. Journal of Communication, 38 (1988): 21-31. [13] McKissack, Jr. Fredrick L. "The Problem with Black T.V.," The Progressive, February 1997: 39. [14] Hughes, Zondra. The New TV Season: "What's New, What's Black, What's Back," Ebony, October 2000. [15] Hughes, Zondra. The New TV Season: "What's New, What's Black, What's Back," Ebony, October 2000. [16] McKissack, Jr. Fredrick L. "The Problem with Black T.V.," The Progressive, February 1997: 38. [17] For seventeen percent of the adult population, occupation was unidentifiable. [18] Clark, C. (1972. Race, identification, and television violence (pp. 120-184). In G.A. Comstock, E.A. Rubinstein, & J.P. Murray (Eds.) Television and Social Behavior (vol. 5): Television 's Effects: Further Explorations. Washington, D.C.: U.S. Government Printing Office. [19] Huston, A., Donnerstein, E., Fairchild, H., Feshbach, N., Katz, P., Murray, J. Rubinstein, E., Wilcox, B., Zuckerman, D. (1992). Big World, Small Screen. University of Nebraska Press. [20] Gerbner, G., Gross, L., Morgan, M., and Signorielli, N. (1981). A curious journey into the scary world of Paul Hirsch. Communication Research, 8: 39-72. [21] Jefferes, L. (1997). Mass Media Effects. Prospect Heights, IL: Waveland Press, Inc: 87. [22] Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, N.J.: Prentice Hall. [23] For a review, see Josephson, W. (1995, February). Television Violence: A Review of the Effects on Children [24] For a review, see Johnston, J. & Ettema, J. (1986). Using Television to Best Advantage: Research for Prosocial TV. In Bryant, J. & Zillman, D. (eds.) Perspectives on Media Effects. LEA. Pp. 143-164. [25] Bandura, A. (1994). Social cognitive theory of mass communication. In J. Bryant and D. Zillman (eds.), Media Effects: Advances in Theory and Research (pp. 61-90). Hillsdale, NJ: Earlbaum.