Although there is much speculation about whether television influences adolescent substance use, careful examination of television content is a crucial first step in determining what influence, if any, might be attributed to the medium. This study addresses a single research question: How are illicit drugs, tobacco, and alcohol portrayed in the most popular television series? Specifically, this study examined the frequency and nature of substance use portrayals in the top-rated, prime-time shows of the fall 1998-1999 season. The study sample, based on Nielsen Media Research ratings, was comprised of four consecutive episodes from 42 top-rated situation comedies and dramas - a total of 168 episodes. Thus, the sample compares substance use portrayals in top-rated shows among teens versus adults, and among white, African-American, and Hispanic teens.

Nielsen Media Research provided television ratings for prime-time viewing by five audience groups: African-American teens (ages 12-17); Hispanic teens (ages 12-17); white teens (ages 12-17); all teens (ages 12-17); and all adults (ages 25-54). Rank-ordered lists of the Nielsen ratings for the period during which shows were taped identified the most popular shows for each audience segment. Programs other than situation comedies and serial dramas were removed from the lists until the top-20 programs within these genres were identified for each audience group. Excluded from the sample were movies (e.g., "Wonderful World of Disney"), sports events (e.g., "FOX NFL Football"), newsmagazines (e.g., "60 Minutes"), and non-fiction or "reality TV" programs (e.g., "World's Wildest Police Videos"). Such shows were excluded to avoid making inferences about substance portrayals in programs with non-comparable formats, such as non-fiction versus fiction, 2-hour movies versus 30-minute sitcoms, real people versus fictional characters. This selection procedure eliminated 9 of the highest-rated programs for African-American teens, 4 for Hispanic teens, 3 for white teens, 4 for teens as a whole, and 1 for adults. (See Appendix A.)

Due to overlapping preferences between subgroups, a total of 42 shows was sufficient to fill out the various top-twenty lists. Thus, the complete sample consists of four consecutive episodes from each of these series broadcast during the fall premiere season, October - December, 1998. Cable shows were not included because no comedy or drama programs on cable drew a teen audience as large as or larger than the teen audiences of broadcast shows. The sample includes 13 one-hour dramas and 29 half-hour situation comedies - a total of 168 program episodes. Appendix A identifies the title, network, and broadcast time for each series. In particular, we examined substance use references and portrayals in four consecutive episodes of:

• the 20 most popular shows among adults
• the 20 most popular shows among teens
• the 20 most popular shows among African-American teens, among Hispanic teens and among white teens
• the 42 shows that appeared in any top-twenty list

Figure 1 compares the unique and shared program preferences for each audience group. According to the Nielsen ratings, African-American, Hispanic and white teens watched many of the same prime-time series. Thirty-four shows accounted for the top-20 preferences among the three teen audiences:

• 6 appeared on all 3 lists
• 7 appeared on both the white and Hispanic lists
• 7 appeared on both the Hispanic and African-American lists
• 7 appeared only on the African-American list
• 7 appeared only on the white list
• no shows appeared only on the Hispanic list

Figure 1: To what extent do multiple audiences watch the same or different programs? Percent of top-rated shows that are unique favorites among: Percent of top-rated shows that are shared preferences among: (View larger image) Percentages based on the 20 top-rated shows for each audience.

Content analyses are fundamentally concerned with counting and describing particular features of mass media content. In this study, specially trained coders examined the episodes for any visual or verbal content related to:

• Illicit drugs (controlled substances, such as marijuana, cocaine, heroin, LSD, etc.)
• Alcohol (beer, wine, champagne, hard liquor and mixed drinks)
• Tobacco (cigarettes, cigars, chewing tobacco)
• Over-the-counter medicines (legally purchased, such as aspirin, diet pills, antacids, laxatives, cough and cold serums, nicotine gum/patches)
• Prescription medication (self-administered prescription medicines, such as sleeping pills, muscle relaxants, anti-depressants, pain relievers)
• For comparison purposes, use of non-alcoholic beverages was also examined

A distinction was made between substance use: visual portrayals of actual consumption or implied consumption; and substance references: any time illicit drugs, alcohol, or tobacco were either seen or mentioned. Thus, the latter category is broader, including both visual portrayals and conversational references to substances.

The coding procedures pertained to four units of analysis: episodes, major characters, screen time and commercials.

Episodes

For each episode, coders paid special attention to a number of factors relevant to the context of substance references. These included:

• the settings in which use occurred (bar, restaurant, party, and so on)
• substance use by major versus minor characters
• whether substance use constituted a major theme or sub-plot
• the parental advisory assigned to each episode (TVG, TVPG, TV14)
• humorous references to substances
• any negative consequences attached to substance use, such as loss of friends, health effects, difficulty in keeping or getting a job, impaired functioning
• negative or critical statements concerning substance use (anti-use statements)
• positive or advocacy statements concerning substance use, such as expressions of desire or longing, or positive attributes of substances and their use (pro-use statements)
• references to intoxication, alcoholism or addiction
• attempts to quit use
• characterizations or portrayals of substance use as positive or negative, such as relaxation, social acceptance, having fun, escaping problems

These coding procedures yielded data describing each of the 168 episodes (i.e., 4 episodes each of 42 shows) in all-or-nothing terms, indicating whether a given episode mentioned or portrayed use, contained a humorous reference, described negative consequences, and so on. Results are reported in percentage terms based on all episodes in a given list of shows. For instance, we will report the percentage of all episodes popular with teens referring to or showing substance use, and compare that to the percentage of substance use references or portrayals in episodes popular with adults. Similarly, we will compare the percentages of episodes mentioning or showing substance use in episodes popular with African-American youth, with Hispanic youth, and with white youth.

Major Characters

Additional analyses were made of the substance-related behavior and statements of ongoing major characters in the shows. A major character was defined as anyone who appeared or was credited in a show's opening or title sequence. A total population of 270 major characters over the 42 series was identified. Information about this population is reported in terms of the proportion of characters who engaged in substance use in all 168 episodes and in the different lists of favorites. The gender, age and ethnic group for each character was identified. For comparison purposes, major character consumption of non-alcoholic beverages was also noted.

Screen Time

The mere appearance of drinking or of alcoholic beverage containers is an imprecise measure of the true salience or centrality of alcohol in a program. Accordingly, two measures of the amount of screen time taken up with alcohol portrayals were created.

Major character drinking time. Using a stopwatch, coders calculated the total time major characters were shown actually drinking or appeared with a container (glass, pitcher, bottle, cup, etc.) from which the character was presumed to be drinking an alcoholic beverage. The stopwatch was stopped whenever alcohol was off screen, as during a close-up of the actor's face. Thus, this was a conservative measure of the prominence of alcohol in a program. A similar measure of screen time for major characters consuming non-alcoholic beverages was taken for comparison purposes.

Alcohol venue time. Viewers may be aware of the presence of substance use or assume its presence even when it is briefly off-screen. Thus, a measure was also taken of the total amount of time spent in locations or engaged in activities typically associated with drinking - in bars, taverns, restaurants, and at parties, wedding receptions, celebrations, and so on. When these venues were encountered, the stopwatch began and kept going as long as the action continued at that venue, even if the alcohol was temporarily off-screen. In other words, timing continued as long as it was reasonable to assume that drinking was still occurring. Only those venues in which drinking was depicted by at least one character were timed. Thus, a teen party with no alcohol consumption shown would not be timed.

Commercials and Other Non-Program Content

Although the major focus of this study was program content, we were also interested in knowing what messages relating to substance use appeared in the non-program content surrounding each episode. Accordingly, coders noted the frequency of:

• ads for alcohol and, for comparison, non-alcoholic beverages
• ads for prescription and over-the-counter medicines
• references and appearances of alcohol and other substances in other product ads and network spots promoting upcoming programs
• ads with an anti-substance message, such as anti-drug and anti-smoking ads.