Abstract
A content analysis of eight week-long samples of network prime time programs broadcast between 2000 and 2008 found a significant decreasing linear trend in the proportion of Black characters, due primarily to the decrease in the number of situation comedies broadcast at the end of the decade. There was a corresponding, significant increasing linear trend in the proportion of White characters. Other minority groups and Hispanics appeared very infrequently. Although Blacks, in the middle of the decade reached a level of parity with their numbers in the U.S. population, it was a representation of segregation and isolation, particularly for Black women. Overall, messages about diversity and tolerance on prime time television are almost nonexistent, fostering a continued sense that minorities and Whites have little in common. Prime time network broadcast programs at the end of this decade were less racially representative of the U.S. population than they were at the beginning of the decade.
Notes
Note. Hispanic and Latino identity was coded separately, and it was not included in the racial categories. ∗p ≤ .01; ∗∗p ≤ .001.
a Data were not collected for the fall 2007 season.
b N = 301,621 million.
c N = 2,926.
d United States: N = 148,659 million; prime time: N = 1,691.
e United States: N = 152,962 million; prime time: N = 1,230.
Note. R%: row percentages. N = 2,921; White: n = 2,351 (80%); Black: n = 433 (15%); Asian: n = 39 (1%); other race: n = 98 (4%); Hispanic: n = 96 (5%).
a Data were not collected for the fall 2007 season.
Note. R%: row percentages. N = 706; situation comedy: n = 241 (34%); action and crime: n = 172 (24%); drama: n = 204 (29%); reality: n = 89 (13%).
a Data were not collected for the fall 2007 season.
Note. R%: row percentages. N = 706; all White: n = 125 (18%); mostly White: n = 319 (45%); mixed: n = 206 (29%); minority: n = 56 (8%).
a Data were not collected for the fall 2007 season.
Note. R%: row percentages; C%: column percentages. N = 2,926; White: n = 2,353 (80%); Black: n = 433 (15%); other: n = 140 (5%); Hispanic: n = 96 (5%). Race (all): χ2(6, N = 433) = 94.77 p < .001; men: χ2(6, N = 258) = 23.72, p < .001; women: χ2(6, N = 175) = 95.69, p < .001. Hispanic (all): χ2(6, N = 96) = 30.09, p < .001; men: χ2(6, N = 52) = 17.17, p < .001; women: χ2(6, N = 44) = 15.56, p < .001.
a Data were not collected for the fall 2007 season.
b Hispanic and Latino: N = 969.
c N = 1,691; White: n = 1,358 (80%); Black: n = 258 (15%); other: n = 75 (5%); Hispanic: n = 52 (4%).
d N = 1,230; White: n = 933 (81%); Black: n = 175 (14%); other: n = 62 (5%); Hispanic: n = 44 (5%).
Note. R%: row percentages; C%: column percentages. N = 2,926. White: n = 2,353 (80%); Black: n = 433 (15%); other: n = 140 (5%); Hispanic: n = 96 (5%). Race (all): χ2(6, N = 433) = 921.66, p < .001; men: χ2(6, N = 258) = 421.65, p < .001; women χ2(6, N = 175) = 516.04, p < .001. Hispanic and Latino (all): χ2(6, N = 96) = 74.16, p < .001; men: χ2(6, N = 52) = 51.02, p < .001; women: χ2(6, N = 44) = 31.27, p < .001.
a Data were not collected for the fall 2007 season.
b Data from 2002–2008 only: N = 2,049.
c N = 1,691; White: n = 1,358 (80%); Black: n = 258 (15%); other: n = 75 (5%); Hispanic: n = 52 (5%).
d N = 1,230; White: n = 993 (81%); Black: n = 175 (14%); other: n = 62 (5%); Hispanic: n = 44 (5%).
Note. R%: row percentages; C%: column percentages. N = 706; all White: n = 125 (18%); mostly White: n = 319 (45%); mixed: n = 206 (29%); minority: n = 56 (8%). χ2(6, N = ) = 184.80, p < .000.
a Data were not collected for the fall 2007 season.
Prime time programs were those broadcast between 8 p.m. and 11 p.m. Eastern Standard Time (EST) on Monday through Saturday and between 7 p.m. and 11 p.m. EST on Sunday. News, magazine, and award programs were not included in the analysis to provide comparability across the yearly samples.
Data were not collected in the fall of 2007.
CW was added to the analysis in the fall of 2006 when WB and UPN ceased to broadcast.
A sampling exercise found that a one-week sample provides a reasonable approximation of the different genres of prime time programs broadcast during the fall of the year. The programs in seven weeks of program listings published by Broadcasting & Cable in September and October, 2003 were categorized into one of five program genres. A comparison of the numbers of programs in each genre in each of these seven weeks found no statistically significant difference from week to week, χ2(4, N = 755) = 2.70, p = .306. Moreover, the distribution by genre in a composite week of programs selected from these seven weeks was not statistically different from the genre distribution of the programs in the fall 2003 sample in this study, χ2(4, N = 229) = 2.72, p = .605.
The coders were students in an upper level communication course that typically has around 25 students each time it is taught. The data were collected by ten different sets of coders, who were predominantly White and middle class; between two thirds and three fourths of the coders for each sample were women.
The coders also had an out-of-class training assignment of a program whose coding was discussed in a later class meeting.
Characters who could not be categorized by gender were eliminated from the analysis.
Ethnicity (Hispanic and Latino) was added to the recording instrument in the fall of 2002; consequently, this part of the analysis is based on six samples of programming. Ethnicity is considered separate from race in this research. The U.S. Census Bureau notes that persons of Hispanic and Latino origin may be of any race (e.g., see www.census.gov/prod/2007pubs/acs-03.pdf).
See Table (“Resident Population by Sex, Race, and Hispanic Origin Status: 2000 to 2007”) at www.census.gov/popest/national/asrh/NC-EST 2006/NC-EST2007–03.xls or www.census.gov/compendia/statab/tables/o6s0013.x15