Sponsored by

Office of National Drug Control Policy

and

Department of Health and Human Services

Substance Abuse and Mental Health Services Administration

Research conducted by

with

April 1999

This study examines the frequency and nature of substance use in the most popular movie rentals and songs of 1996 and 1997. The intent was to determine the accuracy of public perceptions about extensive substance use in media popular among youth. Because teenagers are major consumers of movies and music, there is concern about the potential for media depictions of tobacco, alcohol, and illicit drugs to encourage use. For instance, portrayals that tend to legitimize, normalize, trivialize, or glorify substances might suggest to young people that this behavior is without negative consequences. Careful examination of media content is a crucial first step in determining what role media may play in promoting substance use and abuse.

This study examined the 200 most popular movie rentals and 1,000 of the most popular songs from 1996 and 1997. The source for the movie sample selection was the Video Software Dealers Association, which rank orders home video rentals; the music sample was based on song rankings produced by Billboard, Radio and Records magazine, and the College Music Journal. In order to encompass young people’s divergent tastes in music, the sample included top songs from five genres: Country-Western, Alternative Rock, Hot-100 (also referred to as Top-40 or Mainstream), Rap, and Heavy Metal (which includes Hard Rock and Heavy Rock).

Substances included in the study were illicit drugs, alcohol, tobacco, and over-the-counter and prescription medicines. Researchers examined what was used, by whom, how often, under what circumstances, and with what consequences. The study considered whether movies and songs involved substance use as an important theme, contained pro- or anti-use behavior or statements, conveyed limit-setting messages, or associated substance use with positive or negative contexts. Also examined was the extent to which substance use portrayals varied among different types of movies and movies with different ratings.

Findings revealed that 98 percent of movies studied depicted illicit drugs, alcohol, tobacco or over-the-counter/prescription medicines. Alcohol and tobacco appeared in more than 90 percent of the movies and illicit drugs appeared in 22 percent. About one-quarter (26 percent) of the movies that depicted illicit drugs contained explicit, graphic portrayals of their preparation and/or ingestion. Substance use was almost never a central theme, and very few movies ever specified motivations for use. Less than one-half (49 percent) of the movies portrayed short-term consequences of substance use, and about 12 percent depicted long-term consequences. Of the 669 adult major characters featured in the 200 movies, 5 percent used illicit drugs, 25 percent smoked tobacco, and 65 percent consumed alcohol. One or more major characters used illicit drugs in 12 percent of the movies, tobacco in 44 percent, and alcohol in 85 percent.

All movies in which illicit drugs appeared received restricted ratings (PG-13 or R). However, 45 percent of the movies in which illicit drugs were used did not receive specific remarks identifying drug-related content from the Motion Picture Association of America (MPAA). Fifteen movies depicting illicit drug use (albeit briefly), were not identified as such in the Motion Picture Rating Directory, nor were an additional 10 movies that portrayed drug sales or trafficking.

The major finding from the song analysis is the dramatic difference among music categories, with substance references being particularly common in Rap. Illicit drugs were mentioned in 63 percent of Rap songs versus about 10 percent of the lyrics in the other categories. Similarly, alcohol references appeared in almost half of the Rap lyrics, but in 13 percent or fewer of the other genres. In song lyrics that mentioned illicit drugs, marijuana was by far the most frequent of the illicit drugs mentioned (63 percent). In general, 27 percent of the 1,000 songs contained a clear reference to either alcohol or illicit drugs. There were almost no references to tobacco. Substance use formed a central theme in only 2 percent of the songs and substance use was rarely associated with any motivations or consequences. There were few references that could be considered either explicitly pro-use or anti-use.

Neither movies nor music provided much information about motives for substance use. However, the two media depicted the consequences quite differently, especially for illicit drugs. In movies that portrayed drug use, 48 percent showed one or more consequences of drug use. By contrast, 19 percent of the songs that referred to illicit drugs mentioned any consequence.

Following are additional highlights from the study.

The preceding results show the differences in the frequency of substance use portrayals between movies and songs. However, since songs are only a few minutes in length and movies often last 2 or more hours, another useful comparison was made by dividing the movies into 4,372 segments of 5 minutes and equating these shorter intervals to songs. This approach provides a more accurate comparison of the frequency of substance references in equivalent time periods of movie viewing or music listening.

• Song lyrics contained a greater concentration of illicit drug references than did
  5-minute movie segments. Illicit drugs appeared in nine times more songs (18 percent) than 5-minute movie segments (2 percent).
• The difference between the frequency of alcohol references in movies and songs was reduced. Alcohol appeared in about half as many songs (17 percent) as 5-minute movie segments (31 percent).

This research on substance use in popular movies and music was prompted by two facts: that substance use and abuse constitute a serious problem among America’s youth, and that American teenagers are heavy consumers of motion pictures and popular music. Also influencing the study were widely held public perceptions that media content incorporates a great many messages and images related to substance use, and that it plays a significant role in the creation and perpetuation of America’s substance use problem. Documenting the frequency and nature of substance use portrayals in movies and music is a necessary first step toward understanding the possible connection between symbolic media representations of substances and real-world substance use.

It is important to acknowledge that the mere existence of a certain type of media portrayal does not ensure that audiences will be influenced by it. The ultimate effects of media exposure depend on multiple factors: how individuals interpret messages, the extent to which the messages are contradicted or supported by other sources, the dynamics of parent-child interaction, peer influence, social and cultural background, and so forth. Still, if it is true that substance use appears frequently and is portrayed positively in movies and music, then it is reasonable to hypothesize that these portrayals may be influencing young people to use alcohol, tobacco, and illicit drugs.

On the other hand, if substances are portrayed relatively rarely in movies and music lyrics, or if they are depicted in ways that deglamorize or otherwise discourage use, then it makes little sense to attribute any of society’s substance abuse problems to the media. Equally important, if movies and music do contribute to the problem—that is, if they legitimize, glamorize, or otherwise promote the use of substances—then, logically, they could also help solve the problem by depicting substance use realistically with consequences, or as deviant, unglamorous, and socially unacceptable. In other words, although a variety of other factors may alter the size and nature of media effects, media content clearly matters.

Beyond question, the U.S. faces an epidemic of underage and illegal substance use. In 1997, more than 54 percent of U.S. high school seniors had used an illegal drug at least once, as had more than 29 percent of eighth graders. Among adolescents ages 12 to 17, the average ages of first use of marijuana, cocaine, and heroin were 13.7, 14.7, and 14.4 years, respectively. Youth tobacco smoking rates are higher now than at any time in the past 17 years. Two-thirds of U.S. high school students have tried cigarettes and more than one-third currently smoke. Every day another 3,000 American children and teenagers become regular smokers.

Alcohol consumption among adolescents remains at unacceptably high levels. More than 80 percent of U.S. high school students have tried alcohol; in 1997, more than 31 percent of 12^th graders, 25 percent of 10^th graders, and 14 percent of 8^th graders claimed to have consumed five or more alcoholic drinks in the preceding 2 weeks. Perhaps most disturbing, among 12- to 17-year-olds who exhibit no other problem behaviors, those who have used marijuana, alcohol, or cigarettes in the past month are 17 times more likely to consume illegal drugs such as cocaine, heroin, or LSD than those who have not used these drugs.

Movies and music are extremely popular among adolescents (see Appendix A). Although teens make up only 16 percent of the U.S. population, they account for 26 percent of all movie admissions. Ninety percent of 12- to 20-year-olds report going to movies at least occasionally, and their viewing is not limited to theaters. Sixty-three percent of 9- to 17-year-olds watch at least one rented video per week. Watching videos is one of America’s favorite leisure time activities, with video revenues almost triple that of theatrical box office receipts.

Popular music is the backdrop for much of adolescent life. Teenagers name music listening as their most preferred non-school activity. Moreover, when attention is paid to "background" listening (listening while working, doing homework, driving, etc.), estimates of adolescents’ exposure to music average as high as 4 to 6 hours daily. Recording industry figures indicate that 87 percent of all Rock music sales, 65 percent of Hot-100, 90 percent of R&B and Rap, and 64 percent of Country-Western are accounted for by people age 24 and under.

Of the systematic reviews of the frequency of substance use portrayals in media published since 1980, almost all have focused on television (17 studies); only four have looked at movies, and music lyrics have been ignored. For the most part, content analyses have concentrated exclusively on alcohol or tobacco. Illicit drugs have received little attention (two studies), and only one study looked at all three substances simultaneously (see Appendix B). Most studies have examined media portrayals simply by reporting the percentage of programs or movies in which a substance appears or is "consumed." Few studies have attempted to examine the deeper issues explored in this research, such as the types of characters involved in substance use or the consequences attached to use.

Social learning theory also provided guidance for the study. According to this theory, media messages influence young people by providing explicit, concrete "models" for behaviors (e.g., smoking marijuana), attitudes (e.g., taking an anti-drug point of view), and feelings (e.g., fearing the effects of drug use). Whenever a child or adolescent encounters a media depiction or portrayal as in a movie or song, the potential exists for the behavior to be imitated. Research on social learning theory also demonstrates that the likelihood of imitation depends on the context surrounding the portrayal, particularly consequences attached to the behavior. Generally, perceived negative consequences (e.g., someone dying of an overdose) decrease the probability of a modeling effect, and perceived positive consequences (e.g., gaining social acceptance by drinking at a party) increase the probability.

Further, young audience members are more likely to learn and imitate behaviors performed by attractive, successful, or powerful role models or associated with positive outcomes such as approval, money, power, romance, and sex. Interestingly, even the absence of a negative outcome—such as when a teen character is not punished for using drugs—often has the same influence as an explicit positive consequence or reward. Thus, a second goal of this study was to describe the contexts in which substance use occurs.

The samples for this content analysis consisted of the 200 most popular movie rentals and 1,000 of the most popular songs from 1996 and 1997.

The Video Software Dealers Association rank-ordered list of home video rental income identified the 200 most popular movies for 1996 and 1997. Nine movie titles appeared in the top 100 for both years. In these cases, the titles remained in the year in which they ranked highest, and alternate titles were selected (beginning at rank 101) in order to derive a sample of 100 different movies for each year (see Appendix C). To simplify sample descriptions and analyses, movies were categorized into three genres: action adventure (30 percent); comedy, including romantic and dark/macabre comedies (35 percent); and drama (35 percent).

Because adolescents tend to listen to particular types of music rather than simply to music in general, the songs in the music sample were distributed evenly among five favorite genres (see Appendix D): Country-Western; Alternative Rock; Hot-100 (or Top-40); Rap; and Heavy Metal.

Just as with the video sample, considerable yearly and genre "crossover" of titles complicated the picture. For example, several hits that made the top 100 in Rap for 1996 also made the top 100 for 1997. In addition, a number of songs appeared on the charts for more than one genre. To ensure a total of 1,000 unique titles overall, alternates from the charts were selected when crossover occurred (see Appendix D for explanation).

Music industry charts were used to establish the lists of top songs for the various categories. For four of the music genres—Country-Western, Hot-100, Alternative Rock, and Rap—the year-end rankings from Billboard magazine were used to establish the sample (Billboard uses the term "Modern Rock" to refer to Alternative Rock music). Since Billboard does not publish a Heavy Metal or Heavy Rock chart, a hybrid list was constructed by combining year-end singles from Radio & Records Magazine’s "Active Rock" chart and selections from College Music Journal’s top "Loud Rock" albums (for further explanation see Appendix D).

Describing substance portrayals in movies, which are visual and verbal, and music lyrics, which are only verbal, required different procedures. Specially trained coders watched all 200 movies or read the lyrics of all 1,000 songs included in the study, paying particular attention to the following:

• Alcohol (beer, malt liquor, wine/champagne, hard liquor/mixed drinks, including fictional name brands).
• Tobacco (cigarettes, cigarillos, cigars, pipes, chewing tobacco, including fictional name brands).
• Illicit drugs (controlled substances, such as marijuana, cocaine, crack cocaine, heroin, PCP, crank, LSD).
• 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).
• Inhalants (legal, ordinary household products used for the purpose of getting high, such as paint thinner, glue, lighter fluid, spray paint, aerosols, helium and laughing gas, also used to propel commercial whipping cream).
• Unidentified pills (any pills or capsules of unknown origin or purpose).

Coders were instructed to ignore medicines administered to patients by medical personnel in a hospital or other settings. Also excluded were fictitious drugs and substances with unrealistic, seemingly impossible effects, such as transforming one character into another, or permitting a character to experience other people's memories.

All content analyses are fundamentally concerned with counting and describing particular content. In this study, counting procedures differentiated substance use from substance appearance. Substance use included explicit portrayals of consumption (drinking alcohol, lighting up or puffing on a cigarette, snorting cocaine, injecting drugs, swallowing pills), and depictions that implied consumption, such as buying, ordering, accepting, or possessing alcohol, tobacco, or other substances. Substance appearance was noted whenever substances or related paraphernalia (references to brands of alcohol, tobacco, or over-the-counter medicines, generic bar or cocktail signs, ashtrays, syringes, and the like) were seen, absent any indication of use. In either case, the counting procedure provided a conservative estimate of substance use, since it did not include every verbal reference. For example, a conversation between two characters recalling some past substance-use episode ("Boy, did I tie one on last week") was not coded.

In addition to counting the proportion of movies in which substances appeared, coding procedures attempted to describe dominant messages about substance use. Specifically, coders identified whether movies:

• Involved substance use or trafficking as important themes.
• Conveyed pro-use messages by expressing desire or longing, or advocating positive attributes of substances and their use.
• Modeled anti-use behavior by including characters that expressly refused offers to drink, smoke, or take drugs, or by statements that emphasized rules that govern use or characterize use or users in negative ways.
• Conveyed limit-setting messages that restricted where, when, and how often alcohol, tobacco, and illicit drugs were consumed.
• Associated substance use with positive (e.g., parties, humor) or negative (e.g., crime or violence, rape, risky behaviors) contexts.
• Depicted consequences of substance use to self or others that are short-term (closely linked, brief outcomes) or long-term (removed in time, enduring).

Consequences included any outcomes linked to substance use that showed what can happen to the body when substances are consumed (e.g., coughing, vomiting, blurred vision), or other significant outcomes such as social disapproval, physical harm, or arrest. Coders indicated whether consequences pertained to substance users and/or others (e.g., a woman is beaten when her husband has too much to drink).

The process of counting and describing movie content was applied to several distinct elements: the movie as a whole (as explained above), particular kinds of scenes, major characters, and 5-minute time intervals.

A closer examination of substance use portrayals was provided by a detailed analysis of two types of scenes: 1) those depicting illicit drug use by any character, and 2) those depicting substance use by characters known to be under 18 or who appeared to be high school age or younger. Scenes were defined as a series or sequence of dialogue and action at a single location or point in time. These scene analyses examined why substances were used, in what contexts, and with what, if any, consequences. Specifically, coders identified:

• Apparent motivations for use.
• Physical and social settings of use.
• Positive or negative associations with use.
• Short- or long-term consequences of use (to self or others).

Because scenes showing substance use by an underage character and those depicting illicit drug use by any character are not mutually exclusive, a few scenes are included in both sets of results.

All 748 major characters (adults and youth) were described in terms of role (protagonist vs. antagonist), gender, apparent age group, occupation, ethnicity, and socioeconomic status. In the absence of specific information about a character’s ethnic background, apparent ethnicity was coded. Socioeconomic status (SES) was coded by identifying characters who were obviously well-to-do (high SES) or destitute (low SES). All other characters were coded with a moderate SES.

The prevalence of substance use was determined by calculating the proportion of major characters who used illicit drugs, tobacco, or alcohol. In addition, the number of major characters experiencing consequences of use, attempting to quit, or describing themselves as former users, was noted.

Since movies and scenes vary in length (and characters differ in screen time and importance), none of the preceding coding units—movies, scenes, characters—is ideally suited to studying the frequency of substance use within movies. Rather, this goal is best served by defining and employing a standard coding unit that remains constant from one movie to the next. Therefore, the frequency of substance appearance was described for all 5-minute intervals of the movies, a procedure that is typical of other content analyses identified in Appendix B. The presence or absence of illicit drugs, alcohol, tobacco, and other legal drugs was coded for each 5-minute interval, beginning with the audio and/or video that uniquely identified each movie (typically, after the credits for production/distribution studios) and ending when final credits rolled.

Coders analyzed written transcripts of lyrics for the presence and nature of substance references. The coding process was similar in many respects to that used for movies, as substances of interest were identical and many of the variables overlapped, at least in broad conceptual terms. The different nature of the two media, however, led to certain differences in both coding procedures and variables.

Because lyrics contain no visual information and generally lack the narrative structure, time element, and identifiable cast of characters contained in movies, the basic unit of analysis was the complete lyric. Nothing in the music analysis corresponded to the separate analysis of movies according to scene, time interval, and characters. No attempt was made to analyze the demographic characteristics of individuals. In addition, the complete reliance on verbal cues in the analysis of song lyrics precluded the examination of physical settings, location of the "action," the historical time frame, and so on.

Conversely, the nature of contemporary popular music and youth culture led to the inclusion of certain issues for the music analysis but not for movies. For example, based on the perceived link between Heavy Metal music and Satanic/occult beliefs, music lyrics were examined for any association between such beliefs and substance use.

The first task in the analysis of lyrics was simply to identify any verbal references to illicit drugs, alcohol, or tobacco. Given the ever-changing slang that characterizes both popular music and the drug culture, this process was not as simple as it may sound. Whereas many of the substance-related terms encountered in music are obvious ("champagne," "marijuana," "stoned," "cigarette," and so on), it is difficult to interpret contemporary slang terms. Marijuana, for instance, goes by a variety of street names—"blunt," "chronic," "ganja," "lah lah," "Phillies," "sinsemilla," and "Thai," to name a few. This problem was addressed by employing coders familiar with popular music and its terminology and by consulting published sources (many on the World Wide Web) and experts in the music and substance abuse treatment communities.

Substance references were recorded at several different levels:

• Figurative use of language (e.g., "I’m high on you").
• Mention of places or activities often or almost always associated with substance use ("painting the town," "bar-hopping").
• Literal references to substances or their use ("I’m drinkin’ tonight").

Literal references were further broken down into substance categories (illicit drugs, alcohol, tobacco) and specific substances within those categories, then judged in terms of whether they were:

• "Wallpaper" references in which terminology appears but is not associated with past, present, or intended use ("the girl on the Budweiser billboard").
• Behavior or attitudes related to past, present, or intended use—that is, references made in the "normal" context of consumption ("I got wasted last night").

These categories and distinctions are not mutually exclusive; many songs contained figurative and literal mentions, wallpaper references, and actual use.

For each identified substance, a variety of contextual issues were examined. As with the movie coding, lyrics were examined for references to dealing or trafficking, pro-use and anti-use messages, refusal behavior, limit setting, brand information, motivations, consequences, and associations with use. Motivations and consequences were broken down into specific types. Motivations included peer pressure, mood management (cheering up), relief of troubles or depression, and addiction or craving. Consequences were scored on a 5-point scale ranging from 1 (very negative) to 5 (very positive), with 3 being neutral. In addition, lyrics were examined for consequences within each of these separate categories: mental, emotional, physical, social, legal, monetary/material.

Lyrics were coded for references to intoxication, expressions of a desire to quit use or seek treatment, and condemnation of the effects of substance use on the community at large. They were also reviewed for the presence of associations with sex/romance, rape, violence and crime, driving or other high-risk behaviors, images of wealth or luxury, expressions of bravado or power, and suicide. At the end of the process, coders were asked to consider the lyric as a whole and to judge, on the
5-point scale, whether the portrayal or image of substances or their use was positive or negative.

Substance Use in Popular Movies and Music explored the frequency and context of tobacco, alcohol, and legal and illicit drugs in the most popular home video rentals and music recordings of 1996 and 1997. Key questions asked in the study were: How do popular movies and songs portray these substances? How often is their use depicted? Who uses and in what context? What are the motivations and consequences?

It is important to note that substance use was differentiated from substance appearance in the analysis of movies. Substance use included portrayals of actual consumption or implied consumption; appearance was noted when substance related signs or paraphernalia (billboard ads, ashtrays, cocktail glasses, liquor bottles, syringes) were seen.

The movie analysis examined movies as a whole, scenes, characters, and time intervals. These intervals, 5-minute segments from the films, enabled an analysis of the frequency with which illicit drugs, tobacco, and alcohol appeared within movies of different lengths. They also provided more equivalent units of time so that comparisons could be made between individual songs and movies.

This section begins with some general findings that compare how movies and music treated substances.

At first glance, the preceding results appear to show large differences in the frequency of substance use portrayals between movies, where substances appeared often, and the lyrics of popular songs, in which substances appeared considerably less often. Illicit drugs appeared in about one-fifth of the movies, and alcohol and tobacco in almost all. Illicit drugs and alcohol also appeared in about one-fifth of the lyrics, but references to tobacco were virtually non-existent.

However, it should also be noted that individual songs are only a few minutes long, while movies often last 2 hours or more, raising a legitimate question about whether a single drug reference in a brief song should be compared with a single drug reference in an extended film. This time difference was taken into account by dividing movies into 5-minute segments (still somewhat longer than most popular songs) and comparing these shorter intervals to songs. Analysis using this method produced a very different pattern of results, showing that the prevalence of alcohol in songs was higher than it had originally appeared and that song lyrics contained a greater concentration of illicit drug references than movies (see Figures 1 and 4). Specifically, when 5-minute movie segments were compared with songs:

• Illicit drugs appeared nine times more frequently in lyrics (18 percent) than in
  5-minute movie segments (2 percent).
• Alcohol appeared almost twice as often in movie segments (31 percent) as in songs (17 percent).
• Tobacco appeared eight times as frequently in movie segments (24 percent) as in songs (3 percent).

Considering that young people are likely to listen to at least 25 to 30 hours of music a week, the results indicate they may easily encounter 40 or 50 songs with alcohol or illicit drug references in that time.

Because the nature of movies (audio-visual) and song lyrics (verbal), as well as the frequency and nature of substance use depicted in each, are so fundamentally different, the findings for the two media, for the most part, are presented separately.

In the following pages, text and figures describing the movie analysis specify whether findings pertain to:

• All 200 movies.
• Movies that portray illicit drugs (43), tobacco use (172), or alcohol use (183).
• 5-minute segments of movies.
• Major characters.
• Scenes portraying illicit drug use by any character, major or minor.
• Scenes portraying substance use by characters who appear to be underage.

Results for song lyrics are described at three different levels of analysis:

• All 1,000 songs.
• Approximately 200 songs in each of five genres.
• Songs that refer to illicit drug (156) and alcohol use (149).

1. Of songs referring to illicit drug use, 30 percent associated use with sexual activity or romantic relationships, 20 percent with wealth or luxury, and 20 percent with crime or violence. (Figure 2)
2. Of songs referring to alcohol use, 34 percent associated drinking with sex or romance, 24 percent with wealth or luxury, 13 percent with crime or violence, and 21 percent with expressions of bravado or power.
3. Driving or other risky behavior was associated with substance use in only three songs, suicide in two, and rape in one. There were no examples of songs that connected substance use with Satanic or occult practices or beliefs.

Twenty-one content analyses that examine portrayals of illegal drugs, tobacco, alcohol, and/or legal drugs in media have been published since 1980. As the following annotated bibliography indicates, the research encompasses a variety of approaches and concerns.

The substance of primary concern has been alcohol, examined by 15 of the 21 studies. Eleven studies focused exclusively on alcohol, five on tobacco, and one on legal and over-the-counter drugs. Two studies examined both alcohol and tobacco; one looked at illegal drugs, legal drugs, and alcohol; and one looked at all four substances.

The medium most frequently studied has been television. Seventeen of the studies analyzed television content. Most concentrated on prime-time fictional content, but daytime soap operas, news and documentary programs, commercials, and music videos also received attention. Of the four studies on movies made for theatrical release, three examined only tobacco and one looked at all four substance categories. No work concerned with substance use in popular song lyrics has been located, although one study reports on tobacco and alcohol use in music videos.

Comparisons among studies are somewhat difficult to make due to differences in units of analysis. Movies do not conveniently compare to television programs (i.e., half-hour television programs cannot be compared directly to much longer movies), but even within television, program units may range from 90-minute made-for-television movies to hour dramas to half-hour situation comedies and soap operas to even briefer music video and commercial formats. Compounding the problem, different studies often use different units of analysis within what might otherwise be comparable units. For example, tobacco use has been examined in a variety of ways, in terms of whether or not tobacco appeared in a movie or a television program, the number of scenes in which tobacco appeared or was used, its appearance per 5-minute interval, the total amount of time tobacco appeared on screen, or the number of incidents (defined as individual camera shots in which a character explicitly used tobacco). Similarly, alcohol use has been examined in movies in a variety of ways.

On still another level, some studies coded verbal references to various substances; some concentrated on the appearance of a substance regardless of whether or not it was used; others focused on use; still others paid attention to a variety of character and context attributes associated with use or the frequency of use.

Of the two studies concerned with illegal drugs, one tallied the number of television news programs, public service announcements, or documentaries that mentioned illegal drugs, and the other the number of 5-minute intervals per film in which illegal drug use appeared.

Despite large differences in variables coded and units of analysis, data from the combined studies enable a few preliminary generalizations. In general, tobacco and alcohol are consumed at relatively high rates; smoking is more prevalent in television and film than in the real world, and it is increasing still further on television. In both television and film, alcoholic beverages are more likely to be consumed than non-alcoholic beverages. Drinkers and smokers tend to be leading characters, often protagonists, and usually successful. Underage drinking and smoking are relatively uncommon in both media.

Illegal drug use is seldom portrayed on television; the one prior study that examined movies also found low rates of illegal drug references. Illegal drug users tend to be unattractive, low-status, and/or criminals. More detailed findings can be found in the following sources.

Breed, W., and De Foe, J.R. (1981). "The portrayal of the drinking process on prime-time television." Journal of Communication, 31, pp. 58-67.
Substance(s) of interest: Alcohol. Sample: Top television situation comedies and dramas from 1976-77 season. Unit of analysis: Programs; "significant alcohol scenes" (scenes in which heavy drinking, evaluations of drinking, consequences, etc. appear). Selected findings: TV characters are more likely to drink alcohol than all other beverages combined. Situation comedies and dramas portray drinking/drinkers differently. Drinkers are generally "good" characters, but when they drink too much, they seldom suffer censure or other consequences. Youth drink very rarely, but sometimes express a longing for alcohol.

Breed, W., and De Foe, J.R. (1984). "Drinking and smoking on television, 1950-1982." Journal of Public Health Policy, 5, pp. 257-270.
Substance(s) of interest: Alcohol, tobacco. Sample: Television dramas and situation comedies from the 1950’s through 1982. Unit of analysis: Programs, drinking scenes, drinking acts. Selected findings: Tobacco use declined and alcohol use increased over the 3 decades. Very little smoking was portrayed in the 1981-82 season. Alcohol use increased steadily.

Cafiso, J., Goodstadt, M.S., Garlington, W.K., and Sheppard, M.A. (1982). "Television portrayal of alcohol and other beverages." Journal of Studies on Alcohol, 43, pp. 1232-1243.
Substance(s) of interest: Alcohol. Sample: Representative week of television programming and commercials from west coast television affiliates of ABC, CBS, and NBC, summer, 1975. Unit of analysis: "Drinking events" (alcoholic and non-alcoholic beverages), programs, time periods. Selected findings: Alcohol appears most frequently in motion pictures/dramas, followed by situation comedies, and daytime serials; overall alcohol consumption rate is low, but slightly more alcoholic beverages than non-alcoholic beverages are consumed during prime time; relief was the most common reason given for drinking.

Cruz, J., and Wallack, L. (1986). "Trends in tobacco use on television." American Journal of Public Health, 76, pp. 698-699.
Substances(s) of interest: Tobacco. Sample: Composite 2-week sample of regularly scheduled entertainment prime-time television programs from fall, 1984. Unit of analysis: Smoking acts. Selected findings: One smoking act per hour of programming; more smoking in dramas than situation comedies; two-thirds of smokers were lead characters with 70 percent cast in strong, enduring roles.

De Foe, J.R., and Breed, W. (1988). "Response to the alcoholic by ‘the other’ on prime-time television." Contemporary Drug Problems, 15(2), pp. 205-228.
Substance(s) of interest: Alcohol. Sample: Systematic samples of eight seasons of top-ranked prime-time television entertainment programs (dramas and situation comedies) between 1976 and 1986. Unit of analysis: Scenes in which "alcoholic" characters elicit responses from one or more other characters. Selected findings: Problem drinkers seldom portrayed (74 scenes in 1,417 episodes); responses of "others" were mixed. Tendency to focus on jokes, various forms of enabling and sequences of "game playing." A few scenes did portray responses that might move an alcoholic individual toward reality.

De Foe, J.R., and Breed, W. (1988). "Youth and alcohol in television stories, with suggestions to the industry for alternative portrayals." Adolescence, 23, pp. 533-550.
Substance(s) of interest: Alcohol. Sample: Systematic samples of eight seasons of top-ranked prime-time television entertainment programs (dramas and situation comedies) between 1976 and 1986. Unit of analysis: "Significant alcohol scenes" (scenes in which heavy drinking, evaluations of drinking, consequences of drinking, etc. appear). Selected findings: Less than 2 percent of drinking on television was done by underage drinkers; depictions of young persons drinking were associated with portrayals of gangs or criminal activity. Exception is portrayal of a troubled youth, who ultimately learns alcohol does not solve problems. Qualitative analysis of a few individual programs that portrayed underage drinking.

DuRant, R.H., Rome, E.S., Rich, M., Allred, E., Emans, S.J., and Woods, E.R. (1997). "Tobacco and alcohol use behaviors portrayed in music videos: A content analysis." American Journal of Public Health, 87, pp. 1131-1135.
Substance(s) of interest: Alcohol, tobacco. Sample: 518 music videos recorded from MTV, VH1, CMT, and BET in May and June, 1994. Unit of analysis: Videos; individual occurrences of smoking-related and alcohol-related behavior. Selected findings: Tobacco use was highest on MTV, with alcohol use similar across networks. Tobacco and alcohol use were slightly higher in Rap videos. Lead performers most often use tobacco and alcohol; alcohol use is associated with a high degree of sexuality.

Fedler, F., Phillips, M., Raker, P., Schefsky, D., and Soluri, J. (1994). "Network commercials promote legal drugs: outnumber anti-drugs PSAs 45-to-1." Journal of Drug Education, 24(4), pp. 291-302.
Substance(s) of interest: Illegal drugs, alcohol, legal drugs. Sample: One week of television commercials from ABC, CBS, CNN, and NBC in September 1990. Unit of analysis: Commercials; public service announcements (PSAs); news stories; documentaries. Selected findings: 6 percent of commercials were for over-the-counter drugs, and there were few commercials for alcoholic beverages (0.6 percent). There were 17 anti-drug PSAs, 2 news stories about the problem of illegal drugs, and 1 news story about alcohol. Commercials promoting legal drugs and alcohol outnumbered networks’ news stories, documentaries, and PSAs about illegal drugs by a ratio of 39:1.

Hazan, A.R., and Glantz, S.A. (1995). "Current trends in tobacco use on prime-time fictional television." American Journal of Public Health, 85, pp. 116-117.
Substance(s) of interest: Tobacco. Sample: Three composite weeks of fall 1992 prime-time programming on ABC, CBS, and NBC. Unit of analysis: Tobacco events including anti-smoking messages. Selected findings: 24 percent of programs contain at least one tobacco event; 92 percent were pro-tobacco events, and 8 percent were anti-smoking messages. In terms of character analysis, 55 percent of smokers were "good guys;" high-status characters were more likely than medium- or low-status characters to smoke. Higher rates of smoking occur on television than in real life.

Hazan, A.R., Lipton, H.L., and Glantz, S.A. (1994). "Popular films do not reflect current tobacco use." American Journal of Public Health, 84, pp. 998-999.
Substance(s) of interest: Tobacco. Sample: Two randomly selected, feature length films from the top 20 list each year from 1960 to 1990. Unit of analysis: Five-minute intervals of film time. Selected findings: Rate of tobacco use did not change over the 30-year period. Smokers were generally successful, attractive white males; smoking was three times as prevalent in films as in actual population.

Heilbronn, L.M. (1988). "What does alcohol mean? Alcohol’s use as a symbolic code." Contemporary Drug Problems, 15(2), pp. 229-248.
Substance(s) of interest: Alcohol. Sample: 77 prime-time continuing episodes from 1984 television season containing "alcohol appearances" (see Wallack, Breed, and Cruz, 1986). Unit of analysis: Program. Selected findings: A qualitative (semiotic) analysis of how alcohol-related behavior is used to signify various "meanings" in television programming. Used to establish types of settings (e.g., different types of alcohol signify different life styles), types of characters (e.g., preference for beer often indicates working class status).

Lowery, S.A. (1980). "Soap and booze in the afternoon: An analysis of the portrayal of alcohol use in daytime serials." Journal of Studies on Alcohol, 41, pp. 829-838.
Substance(s) of interest: Alcohol. Sample: Four consecutive weeks of 14 daytime soap operas and 12 daytime game shows from spring 1977. Unit of analysis: half-hour programs;
1-minute intervals. Selected findings: Alcohol-related events occurred at an average rate of 3 per program in soap operas and .3 per program in game shows. About half the alcohol-related events in soap operas were actual drinking events; drinking occurred most frequently in the home.

Mathios, A., Avery, R., Bisogni, C., and Shanahan, J. (1998). "Alcohol portrayal on prime-time television: Manifest and latent messages." Journal of Studies on Alcohol, 59, pp. 305-310.
Substance(s) of interest: Alcohol. Sample: Two weeks of broadcast television programs from ABC, CBS, FOX, and NBC from fall 1994 and spring 1995. Unit of Analysis: Food and beverage episodes within programs; characters. Selected Findings: Alcoholic beverages were the most frequently portrayed food or drink; alcohol appeared with characters of all ages, accounting for a significant percent of food and drink incidents for adolescents. When adolescents are involved in alcohol episodes, they are portrayed with significantly more negative personality characteristics than older characters.

Signorielli, N. (1987). "Drinking, sex, and violence on television: The cultural indicators perspective." Journal of Drug Education, 17(3), pp. 245-260.
Substance(s) of interest: Alcohol. Sample: 17 annual week-long samples of prime-time and weekend daytime network dramatic TV programs from 1969 to 1985. Unit of analysis: Program; major characters. Selected findings: References to alcohol and drinking increased steadily from 1969 to 1985. The harmful effects of drinking alcohol were rarely mentioned. Drinking was often associated with sexual behavior. About 37 percent of major characters drink, and they do not differ significantly from major characters who do not drink. Alcoholics were treated quite negatively.

Stockwell, T.F., and Glantz, S.A. (1997). "Tobacco use is increasing in popular films." Tobacco Control, 6, pp. 282-284.
Substance(s) of interest: Tobacco. Sample: Five randomly selected feature length films from the top 20 list from 1990 through 1996; two randomly selected, feature length films from the top 20 list each year from 1960 to 1990. Unit of analysis: 5-minute intervals of film time. Selected findings: Film portrayals of tobacco use bottomed out in the eighties, and have since increased to levels last seen in the sixties. Films continue to portray smokers as successful, white males, while portrayal of smoking among women is increasing.

Teens take a look at tobacco use in the top 250 movies from 1991-1996. (1997). American Lung Association of Sacramento-Emigrant Trails, Sacramento, CA.
Substance(s) of interest: Tobacco. Sample: Top 50 box office movies each year from 1991 through 1996. Unit of analysis: Individual camera shots of tobacco use. Selected findings: 77 percent of the movies contained at least one tobacco incident; 23 percent had no incidents; 50 percent had 10 or more incidents. Tobacco use was portrayed as attractive in 33 percent of the movies, relaxing in 38 percent of the movies, and as a means of rebellion in 16 percent. Anti-smoking messages appeared in 29 percent of the movies. Leading and/or supporting actors smoked in 75 percent of movies that portrayed smoking; men lit up more than twice as often as women.

Terre, L., Drabman, R.S., and Speer, P. (1991). "Health-relevant behaviors in media." Journal of Applied Social Psychology, 21, pp. 1303-1319.
Substance(s) of interest: Illegal drugs, legal drugs, tobacco, alcohol. Sample: 20 top box-office films each year from 1977-78 through 1987-88 (200 films). Unit of analysis: 5-minute intervals of film time. Selected findings: Overall, smoking increased over the time period. Smoking and illegal drug use were more common in R-rated movies. Alcohol consumption was greater in non-R-rated movies. Males were more likely than females to use tobacco or alcohol; low-status characters were more likely than high-status characters to engage in illegal drug use.

Tsao, J.C. (1997). "Informational and symbolic content of over-the-counter drug advertising on television." Journal of Drug Education, 27(2), pp. 173-197.
Substance(s) of interest: Over-the-counter drugs. Sample: 150 commercials for over-the-counter drugs recorded from ABC, CBS, and NBC in 1993. Unit of analysis: Individual ads. Selected findings: Drug disclosures and drug performance most common type of information, followed by drug quality, drug ingredients, and drug popularity. Of the ads, 88 percent present over-the-counter drugs as a simple solution to relieve symptoms; 25 percent depict them casually rather than as products to be used carefully.

Wallack, L, Breed, W., and De Foe, J.R. (1985). "Alcohol and soap operas: Drinking in the light of day." Journal of Drug Education, 15(4), pp. 365-379.
Substance(s) of interest: Alcohol. Sample: 4 ½ years of the daytime soap opera "All My Children," including 30 consecutive episodes from 1984. Unit of analysis: Characters. Selected findings: Depicted drinking and drinking problems realistically and accurately. There were three patterns of alcohol use identified: social facilitation; crisis management; and escape from reality. Negative discussions or consequences were associated only with drinking to escape from reality, and program may encourage drinking for purposes of social facilitation and crisis management.

Wallack, L., Breed, W., and Cruz, J. (1987). "Alcohol on prime-time television." Journal of Studies on Alcohol, 48(1), pp. 33-38.
Substance(s) of interest: Alcohol. Sample: Composite 2-week sample for all regularly scheduled, network, prime-time dramatic programming, fall 1984 (127 episodes). Unit of analysis: Program episodes; scenes; alcohol "appearances." Selected findings: 80 percent of the episodes contained one or more appearances of alcohol; 90 percent of dramas referred to alcohol; alcohol was ingested in 60 percent of programs. There were more than 10 drinking acts per hour. Alcohol was consumed almost three times more frequently than non-alcoholic beverages.

Wallack, L., Grube, J.W., Madden, P.A., and Breed, W. (1990). "Portrayals of alcohol on prime-time television." Journal of Studies on Alcohol, 51(5), pp. 428-437.
Substances(s) of interest: Alcohol. Sample: Composite 3-week sample of fictional prime-time network television programming from 1986 fall season. Unit of analysis: Program episodes; scenes. Selected findings: Alcohol appeared in 64 percent of episodes and was consumed in 50 percent; alcohol drinking acts occurred more than eight times per hour; most drinking occurred in made-for-television movies, followed by situation comedies, theatrical movies, and dramas. Regularly appearing characters were more likely to drink than non-regularly appearing characters; drinkers tended to be higher status, white, upper-class professionals.

Because adolescents tend to listen to a certain type of music rather than simply "popular music" in general, the 1,000 songs in the music sample were distributed evenly among 5 popular genres: Country-Western, Alternative Rock, Hot-100 (or Top-40), Rap, and Heavy Metal. The study design called for the inclusion of the top 200 songs (100 for each year) within each of these categories; however, the existence of considerable yearly and genre "crossover" complicated the general picture. For example, a list of songs that made the top 100 in Rap for 1996 and 1997 included several that were popular in both years, thus yielding an overall sample somewhat smaller than 200.

Billboard charts were used for all music types except Heavy Metal. Because Billboard does not publish a Heavy Metal chart, song titles in this category were derived from a two-step process. The first songs included were those that appeared on Radio and Records magazine’s Active Rock singles chart, without at the same time appearing (or "crossing over") on Billboard’s Modern Rock chart (used to define the Alternative Music list for this study). This process yielded 94 songs. The remaining songs—the number needed to achieve the total Heavy Metal sample—were taken from College Music Journal’s list of top Loud Rock albums. Cuts were randomly selected from the top 58 albums of 1996 and the top 59 of 1997. While this hybrid selection procedure deviated from the method used for other music categories, it nonetheless achieved the end result of accumulating a list of songs representative of the core of rock music’s "heavy" edge.

Billboard’s Hot-100 chart represents mainstream or pop music as well as certain songs from other genres that became popular enough to break into the Hot-100 charts for 1996 or 1997. The following figures indicate the extent of this "crossover" phenomenon.

• 31 songs appeared on both the Hot-100 and Rap lists (15 percent crossover).
• 21 songs appeared on both the Hot-100 and Alternative Rock lists (10 percent crossover).
• 2 songs appeared on both the Hot-100 and Heavy Metal lists (1 percent crossover).
• 1 song appeared on both the Hot-100 and Country lists (0.5 percent crossover).

The following pages identify all of the song titles that were included in this study.