Abstract
OBJECTIVE:
This study used prospective data to test the hypothesis that exposure to alcohol advertising contributes to an increase in underage drinking and that an increase in underage drinking then leads to problems associated with drinking alcohol.
METHODS:
A total of 3890 students were surveyed once per year across 4 years from the 7th through the 10th grades. Assessments included several measures of exposure to alcohol advertising, alcohol use, problems related to alcohol use, and a range of covariates, such as age, drinking by peers, drinking by close adults, playing sports, general TV watching, acculturation, parents’ jobs, and parents’ education.
RESULTS:
Structural equation modeling of alcohol consumption showed that exposure to alcohol ads and/or liking of those ads in seventh grade were predictive of the latent growth factors for alcohol use (past 30 days and past 6 months) after controlling for covariates. In addition, there was a significant total effect for boys and a significant mediated effect for girls of exposure to alcohol ads and liking of those ads in 7th grade through latent growth factors for alcohol use on alcohol-related problems in 10th grade.
CONCLUSIONS:
Younger adolescents appear to be susceptible to the persuasive messages contained in alcohol commercials broadcast on TV, which sometimes results in a positive affective reaction to the ads. Alcohol ad exposure and the affective reaction to those ads influence some youth to drink more and experience drinking-related problems later in adolescence.
KEY WORDS: alcohol advertising, alcohol drinking, adolescent, statistical model
What’s Known on This Subject:
The influence of alcohol advertising on underage drinking has been demonstrated in both cross-sectional and prospective studies. What is not well known is whether this increase in drinking leads to more problems related to alcohol consumption.
What This Study Adds:
Exposure to alcohol advertising and liking of those ads in grade 7 has a significant influence on the severity of alcohol-related problems in grade 10 and that influence is mediated by growth in alcohol use from grades 7 to 9.
Alcohol use among adolescents and young adults is a major health concern in the United States. According to a Substance Abuse and Mental Health Services Administration report published in 2004,1 ∼10.9 million (29%) adolescents reported drinking alcohol in the past month, 16.6% reported problem behaviors related to alcohol use, and 6.2% met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria for substance abuse or dependence.2 Because of the risks involved, considerable attention has been given to the influence of alcohol advertising on underage drinking. Cross-sectional studies have consistently shown a small but significant association between exposure to alcohol ads and alcohol use.3–6 More importantly, prospective studies have shown similar findings providing support for a temporal relationship between exposure to ads and alcohol use,7–13 which has been confirmed in a systematic review of 13 longitudinal studies.14 Few studies, however, have successfully used prospective data to demonstrate the temporal relationship among exposure to alcohol ads, alcohol consumption, and problem behaviors associated with alcohol use.
The current study examined the effects of alcohol ad exposure on consumption and problem behaviors across 4 years of data collection to test 2 hypotheses. First, the influence of exposure to alcohol ads on underage drinking was hypothesized to interact with an effect modifier (or moderator): an affective reaction to alcohol ads, self-reported as a liking of alcohol ads.5,15 It was anticipated that adolescents who like alcohol advertisements will be more likely to elaborate on the content of the ads (eg, imagine themselves in the scene), and as a result, they will be more likely to be persuaded to try the product.16,17 Studies on copy testing by advertisers have shown that liking of advertisements is predictive of sales for consumer products.18 In addition, drinking among adolescents and young adults is associated with desirability and identification with characters in alcohol ads5 and with liking of alcohol ads.10,19 Second, it was hypothesized that the growth in alcohol use over the first 3 years of the study would significantly mediate the relationship between exposure to alcohol ads in year 1 and alcohol-related problems in year 4 (see paths a and b in Fig 1). That is, effects of Year 1 alcohol ads on the growth in alcohol consumption over time (path a) was expected to translate into later (Year 4) levels of alcohol problems (path b). Figure 1 depicts a conceptual model that incorporates both key hypotheses within a moderated-mediation model.
FIGURE 1.
Conceptual model of primary hypothesized paths tested in the moderated-mediation models.
Methods
Participants
The current data were collected as part of a prospective study on the influence of alcohol advertising on underage drinking.12,13 Participants recruited from public schools were surveyed during regular school hours from the 7th through 10th grades. Of the 4186 students recruited to participate in the study, 3890 (93% of consented) students completed the survey in at least 1 wave: 2986 (77%) were surveyed in 7th grade, 2849 (73%) in the 8th grade, 2093 (54%) in the 9th grade, and 1609 (41%) in the 10th grade. Dropout in the 9th and 10th grades was largely because of failure of entire schools to remain in the study after initial agreements by the schools to participate. Thus, most dropouts were not because of subject self-selection factors that could confound results. Further, the data analysis (outlined below) thoroughly addresses missing data. A total of 23 public middle schools, randomly selected from all middle schools in Los Angeles County, agreed to participate in the study. The goal was to recruit a sample representative of students attending Los Angeles County high schools.
Procedures
All seventh grade students in each school at the time of the study were invited to participate. Data collectors visited classrooms to distribute consent and assent forms to students about 2 weeks before administering the surveys. Parents of the students either signed a consent form brought home from school by the student or gave verbal consent to data collectors via telephone if the consent forms were not returned. Students signed assent forms before completing the surveys. The surveys and all procedures were approved by the University of Southern California Institutional Review Board. Students completed paper-and-pencil questionnaires during regular classroom hours at their school.
Outcome Measures
Current alcohol use was assessed with a total of 9 self-report items. Five items20 assessed on how many days during the past 30 days the participant drank beer, wine, or liquor; drank 3 or more beers in a row; drank 3 or more glasses of wine or liquor; and drank enough to get drunk. An additional 4 items asked how often in the past 6 months participants drank beer, drank wine or wine coolers, drank liquor, or got drunk. An index was formed from all 9 items (coefficient α = 0.91). Problems due to alcohol use were assessed with 8 self-report items.21 Participants indicated how often their alcohol use caused them problems, such as not being able to do their homework, getting into fights, neglecting responsibilities, or causing someone shame or embarrassment. An index score was formed from the 8 items (coefficient α = 0.93).
Independent Variables
Four measures of exposure to alcohol advertising were assessed: (1) Exposure to alcohol advertising on popular shows. Participants indicated how frequently they watched 20 popular TV shows during the past month on a 6-point scale ranging from 1 (never) to 6 (every day). The frequency of watching each show was multiplied by the average frequency of alcohol advertising broadcast on each show during the 10 months before the survey.22 Data on televised alcohol advertising during the popular shows was purchased from Nielsen Media Research (New York, NY). The weighted items were summed to yield an index score for the number of alcohol ads each participant was exposed to during a typical day of watching popular shows (coefficient α = 0.79). This measure of exposure does not directly ask about exposure to alcohol ads, and it has been predictive of alcohol use in past studies.13,22 (2) Exposure to alcohol advertising on sports programs. This measure was similar to the popular shows assessment except that it asked about the frequency of watching college and professional sports programs (coefficient α = 0.80), which often include a higher frequency of alcohol advertisements than other programming.23 (3) Memory for alcohol ads: cued recall. Surveys included still pictures captured from TV advertisements including 2 example and 15 test ads.24 The still pictures extracted from advertisements did not contain brand names or logos. An open-ended item asked participants to write down what product was being advertised. Independent judges coded the responses as being related to the advertisement or not (κ = 0.88). (4) Self-reported observation of alcohol advertising. Participants were asked 4 items25 about how often they saw alcohol commercials on TV (coefficient α = 0.72).
The survey included 3 items assessing how much participants like alcohol ads on TV.26 The items assessed whether participants thought that alcohol ads are funny or sexy, and whether they like the alcohol ads better than other ads (coefficient α = 0.78). These items measured an affective or emotional reaction to alcohol ads that has been useful in both the study of alcohol advertising5,15,19 and by the advertising industry in general to estimate the potential effectiveness of advertising copy.18 Additional covariates associated with adverting exposure, alcohol use, or alcohol-related problems included the amount of time watching television27,28; observing friends drinking29; observing well-known adults drinking30; participating in sports31; age, gender, ethnicity, language acculturation32,33; and parents’ occupation and education (see Appendix for assessments).
Data Analyses
Construction of the structural equation models used to test the hypotheses involved 2 steps.34 First, a measurement model established the simple structure of the model, measurement invariance across gender,35 and acceptability of parcels as indicators.36 The second step involved fitting of 4 latent growth-curve models, one for each measure of exposure to alcohol advertising. Goodness-of-fit statistics37 included the χ2 test, Comparative Fit Index, Tucker-Lewis Index, Root Mean Squared Error of Approximation, and the Standardized Root Mean Square Residual. The current analyses used full information maximum likelihood estimation38 to adjust for uncertainty associated with missing data. Mediation effects (ie, specific and total indirect effects) were assessed using the multivariate δ method.39 This method estimates significance for the product of 2 regression coefficients, the coefficient for the mediator regressed on the predictor and the coefficient for the outcome regressed on the mediator adjusted for the predictor and is consistent with criteria recommended by MacKinnon et al.40 Mplus41 was used to fit the measurement and the latent growth models. SEs were adjusted for clustering by school.41
Results
Demographic characteristics for time 1 of the study, as shown in Table 1, indicated that the students in seventh grade were 12.51 (SD = 0.54) years old. Thirteen percent were non-Hispanic whites and 48% were Hispanic. Boys reported significantly more alcohol use than girls for past 30-day use of beer, lifetime binging with beer, and past 30 days binging with beer, and boys reported more negative consequences as a result of alcohol use. Participants more likely to have been lost to follow-up included those in wave 1 who knew peers (odds ratio [OR] = 1.30; 95% confidence interval [CI] = 1.16–1.44) or adults (OR = 1.13; 95% CI = 1.05–1.21) who drank alcohol, were exposed to more alcohol commercials on popular shows (OR = 1.28; 95% CI = 1.01–1.61), or were Asian compared with whites (OR = 2.00; 95% CI = 1.30–3.08). There was no difference for those lost to follow-up based on gender, age acculturation, participation in sports, parents’ education, lifetime or past 30-day alcohol use, alcohol-related problems, TV viewing, self-reported exposure to advertisements, or liking of alcohol advertisements.
TABLE 1.
Demographic Information for Participants in Seventh Grade
| Item | Total | Girls | Boys |
|---|---|---|---|
| Gender, n (%) | 3890 (100) | 1905 (50.14) | 1894 (49.86) |
| Age, mean (SD) | 12.51 (0.54) | 12.51 (0.54) | 12.51 (0.53) |
| Ethnicity, n (%) | |||
| White/non-Hispanic | 520 (13.37) | 261 (13.78) | 259 (13.60) |
| Hispanic | 1862 (47.87) | 937 (49.47) | 923 (48.45) |
| Asian | 662 (17.02) | 324 (17.11) | 338 (17.74) |
| Black/African American | 120 (3.08) | 56 (2.96) | 64 (3.36) |
| Native Hawaiian or Pacific Islander | 30 (0.77) | 15 (0.79) | 15 (0.79) |
| American Indian or American Native | 37 (0.95) | 17 (0.90) | 20 (1.05) |
| Don’t know | 491 (12.62) | 196 (10.35) | 206 (10.81) |
| Mixed | 168 (4.32) | 88 (4.65) | 80 (4.20) |
| Language acculturation, mean (SD) | 4.22 (0.76) | 4.14 (0.79) | 4.28 (0.72) |
| At least 1 drink of beer in lifetime, n (%) | |||
| 0 d | 1595 (56.94) | 842 (59.21) | 753 (54.60) |
| 1 d | 532 (18.99) | 260 (18.28) | 272 (19.72) |
| 2 d | 242 (8.64) | 123 (8.65) | 119 (8.63) |
| 3 to 9 d | 216 (7.71) | 101 (7.10) | 115 (8.34) |
| 10 to 19 d | 86 (3.07) | 39 (2.74) | 47 (3.41) |
| 20 to 39 d | 50 (1.79) | 24 (1.69) | 26 (1.89) |
| 40 to 99 d | 30 (1.07) | 15 (1.05) | 15 (1.09) |
| 100 or more days | 50 (1.79) | 18 (1.27) | 32 (2.32) |
| At least 1 drink of beer in past 30 days, n (%)a | |||
| 0 d | 2414 (83.18) | 1243 (84.44) | 1171 (81.89) |
| 1 d | 281 (9.68) | 140 (9.51) | 141 (9.86) |
| 2 d | 90 (3.10) | 40 (2.72) | 50 (3.50) |
| 3 to 5 d | 55 (1.90) | 20 (1.36) | 35 (2.45) |
| 6 to 9 d | 27 (0.93) | 16 (1.09) | 11 (0.77) |
| 10 to 19 d | 9 (0.31) | 6 (0.41) | 3 (0.21) |
| 20 to 29 d | 6 (0.21) | 3 (0.20) | 3 (0.21) |
| All 30 d | 20 (0.69) | 4 (0.27) | 16 (1.12) |
| At least 1 drink of wine or liquor in lifetime, n (%) | |||
| 0 d | 1799 (64.67) | 934 (66.15) | 865 (63.14) |
| 1 d | 455 (16.36) | 215 (15.23) | 240 (17.52) |
| 2 d | 210 (7.55) | 113 (8.00) | 97 (7.08) |
| 3 to 9 d | 153 (5.50) | 78 (5.52) | 75 (5.47) |
| 10 to 19 d | 69 (2.48) | 33 (2.34) | 36 (2.63) |
| 20 to 39 d | 40 (1.44) | 17 (1.20) | 23 (1.68) |
| 40 to 99 d | 23 (0.83) | 0 (0.64) | 14 (1.02) |
| 100 or more days | 33 (1.19) | 13 (0.92) | 20 (1.46) |
| At least 1 drink of wine or liquor in past 30 days, n (%) | |||
| 0 d | 2422 (83.81) | 1246 (85.05) | 1176 (82.53) |
| 1 d | 272 (9.41) | 124 (8.46) | 148 (10.39) |
| 2 d | 105 (3.63) | 54 (3.69) | 51 (3.58) |
| 3 to 5 d | 34 (1.18) | 17 (1.16) | 17 (1.19) |
| 6 to 9 d | 23 (0.80) | 14 (0.96) | 9 (0.63) |
| 10 to 19 d | 10 (0.35) | 5 (0.34) | 5 (0.35) |
| 20 to 29 d | 6 (0.21) | 2 (0.14) | 4 (0.28) |
| All 30 d | 18 (0.62) | 3 (0.20) | 15 (1.05) |
| 3 or more drinks of beer in a row in lifetime, n (%)a | |||
| 0 d | 2432 (88.12) | 1258 (89.92) | 1174 (86.26) |
| 1 d | 134 (4.86) | 61 (4.36) | 73 (5.36) |
| 2 d | 70 (2.54) | 33 (2.36) | 37 (2.74) |
| 3 to 9 d | 45 (1.63) | 13 (0.93) | 32 (2.35) |
| 10 to 19 d | 26 (0.94) | 13 (0.93) | 13 (0.96) |
| 20 to 39 d | 25 (0.91) | 14 (1.00) | 11 (0.81) |
| 40 to 99 d | 8 (0.29) | 2 (0.14) | 6 (0.44) |
| 100 or more days | 20 (0.72) | 5 (0.36) | 15 (1.10) |
| 3 or more drinks of beer in a row in past 30 days, n (%)a | |||
| 0 d | 2688 (92.91) | 1383 (94.40) | 1305 (91.39) |
| 1 d | 105 (3.63) | 47 (3.21) | 58 (4.06) |
| 2 d | 34 (1.18) | 14 (0.96) | 20 (1.40) |
| 3 to 5 d | 25 (0.86) | 9 (0.61) | 16 (1.12) |
| 6 to 9 d | 11 (0.38) | 5 (0.34) | 6 (0.42) |
| 10 to 19 d | 7 (0.24) | 3 (0.20) | 4 (0.28) |
| 20 to 29 d | 6 (0.21) | 2 (0.14) | 4 (0.28) |
| All 30 d | 17 (0.59) | 2 (0.14) | 15 (1.05) |
| 3 or more drinks of wine or liquor in lifetime, n (%) | |||
| 0 d | 2448 (89.15) | 1263 (90.67) | 1185 (87.58) |
| 1 d | 135 (4.92) | 55 (3.95) | 80 (5.91) |
| 2 d | 58 (2.11) | 31 (2.23) | 27 (2.00) |
| 3 to 9 d | 43 (1.57) | 20 (1.44) | 23 (1.70) |
| 10 to 19 d | 20 (0.73) | 9 (0.65) | 11 (0.81) |
| 20 to 39 d | 17 (0.62) | 7 (0.50) | 10 (0.74) |
| 40 to 99 d | 6 (0.22) | 2 (0.14) | 4 (0.30) |
| 100 or more days | 19 (0.69) | 6 (0.43) | 13 (0.96) |
| 3 or more drinks of wine or liquor in past 30 days, n (%) | |||
| 0 d | 2707 (93.73) | 1384 (94.60) | 1323 (92.84) |
| 1 d | 92 (3.19) | 43 (2.94) | 49 (3.44) |
| 2 d | 30 (1.04) | 16 (1.09) | 14 (0.98) |
| 3 to 5 d | 18 (0.62) | 10 (0.68) | 8 (0.56) |
| 6 to 9 d | 13 (0.45) | 4 (0.27) | 9 (0.63) |
| 10 to 19 d | 7 (0.24) | 2 (0.14) | 5 (0.35) |
| 20 to 29 d | 6 (0.21) | 2 (0.14) | 4 (0.28) |
| All 30 d | 15 (0.52) | 2 (0.14) | 13 (0.91) |
| Consequences of alcohol use, mean (SD)b | 0.09 (0.41) | 0.08 (0.38) | 0.11 (0.44) |
Alcohol use by student gender was significant for past 30-days use of beer, lifetime binging with beer, and past 30- days binging with beer (all χ2(7) > 14.07, P < .05), but all other comparisons of alcohol use by student gender were nonsignificant (all P > .05).
Consequences of alcohol use differed by gender (t[2648] = –2.15, P < .05); P = proportion.
Measurement Model
The measurement model examined the factor loading, simple structure, and measurement invariance of the latent variables proposed for the models. Indicators loaded well on their hypothesized latent variables in separate models for girls and boys. Examination of a priori hypothesized modification indices for cross-loadings among the alcohol use, alcohol-related problems, ad exposure, and liking of ads target latent factors provided support for a simple structure among the factors. The measurement model findings for the alcohol-related problems factor warranted the use of parcels of indicators in the structural model to provide more stable model estimation.36,42 Tests for invariance of loadings and thresholds in a multigroup model by gender was adequate to compare structural models across gender.43 Similar tests for invariance of loadings and thresholds in a multigroup model by grade provided evidence for invariance across time for items measuring alcohol use in the growth curves.
Latent Growth Models
The latent growth factors for alcohol use over times 1 through 3 and the latent factor for alcohol-related problems were regressed on each of the 4 alcohol ad exposure measures in 4 separate series of model evaluations. The hypothesized moderator, liking of alcohol ads, was included in each of the 4 models. In addition, the growth factors were simultaneously regressed on covariates measured at time 1, including age, observing peers drink, observing adults drink, playing sports, general TV watching, language acculturation, and socioeconomic status (occupation and education of each participant’s parents). All structural growth models differed by gender, so only those results for multigroup models by gender are presented here.
As shown in Table 2 and Fig 2, the coefficient for the intercept regressed on the interaction term was significant for boys and for girls. Figure 3 depicts this interaction illustrating that the level of exposure to ads was more predictive of alcohol use in seventh grade for those students who reported a greater liking of alcohol ads. There was no interaction in the prediction of the slope for the latent growth for alcohol use.
TABLE 2.
Standardized Parameter Estimates for the Mediation Model
| Girls | Boys | |||
|---|---|---|---|---|
| Parameter Estimate | SE | Parameter Estimate | SE | |
| Intercept on | ||||
| T1 alcohol use | 0.759*** | 0.046 | 0.821*** | 0.038 |
| T2 alcohol use | 0.590*** | 0.060 | 0.643*** | 0.047 |
| T3 alcohol use | 0.466*** | 0.056 | 0.506*** | 0.030 |
| Slope on | ||||
| T1 alcohol use | 0.000 | 0.000 | 0.000 | 0.000 |
| T2 alcohol use | 0.404*** | 0.036 | 0.349*** | 0.057 |
| T3 alcohol use | 0.640*** | 0.056 | 0.549*** | 0.101 |
| T4 alcohol-related problems on | ||||
| T4 problems 1 | 0.707*** | 0.029 | 0.720*** | 0.035 |
| T4 problems 2 | 0.692*** | 0.039 | 0.721*** | 0.056 |
| T4 problems 3 | 0.705*** | 0.038 | 0.736*** | 0.048 |
| T4 problems 4 | 0.734*** | 0.050 | 0.780*** | 0.037 |
| Intercept on T1 predictors | ||||
| Popular shows | –0.052 | 0.034 | –0.027 | 0.031 |
| Liking of ads | 0.267*** | 0.047 | 0.171*** | 0.028 |
| Shows x Liking | 0.091* | 0.042 | 0.093* | 0.046 |
| T1 problems | 0.297* | 0.123 | 0.264** | 0.084 |
| Age | 0.030 | 0.031 | 0.040 | 0.030 |
| Peer drinking | 0.426*** | 0.060 | 0.539*** | 0.052 |
| Playing sports | 0.006 | 0.043 | –0.009 | 0.024 |
| Adult drinking | 0.155*** | 0.036 | 0.138** | 0.053 |
| General TV viewing | 0.012 | 0.034 | 0.012 | 0.037 |
| Language acculturation | 0.050 | 0.042 | –0.098* | 0.040 |
| Parents’ jobs | 0.000 | 0.041 | 0.112* | 0.046 |
| Parents’ education | –0.041 | 0.045 | –0.002 | 0.030 |
| Slope on T1 predictors | ||||
| Popular shows | 0.190** | 0.058 | 0.113 | 0.063 |
| Liking of ads | –0.021 | 0.078 | 0.129* | 0.060 |
| Shows x Liking | –0.083 | 0.068 | –0.112 | 0.081 |
| T1 problems | –0.125 | 0.135 | 0.076 | 0.156 |
| Age | 0.031 | 0.039 | –0.075 | 0.068 |
| Peer drinking | 0.057 | 0.075 | –0.483*** | 0.128 |
| Playing sports | –0.137 | 0.073 | –0.015 | 0.074 |
| Adult drinking | –0.029 | 0.067 | –0.103 | 0.119 |
| General TV viewing | –0.021 | 0.064 | –0.059 | 0.062 |
| Language acculturation | 0.029 | 0.073 | 0.227* | 0.097 |
| Parents’ jobs | 0.130 | 0.075 | –0.135 | 0.109 |
| Parents’ education | –0.085 | 0.064 | 0.009 | 0.090 |
| T4 alcohol-related problems on | ||||
| Intercept | 0.393* | 0.166 | 0.177 | 0.303 |
| Slope | 0.478*** | 0.106 | 0.179 | 0.214 |
| Popular shows | –0.054 | 0.065 | –0.007 | 0.058 |
| Liking of ads | –0.102 | 0.064 | –0.095 | 0.062 |
| Shows x Liking | 0.040 | 0.072 | 0.167 | 0.094 |
| T1 problems | 0.050 | 0.070 | 0.014 | 0.090 |
| Age | 0.036 | 0.049 | –0.004 | 0.034 |
| Peer drinking | –0.022 | 0.085 | 0.234 | 0.214 |
| Playing sports | 0.050 | 0.059 | 0.027 | 0.044 |
| Adult drinking | –0.027 | 0.041 | 0.021 | 0.074 |
| General TV viewing | 0.022 | 0.062 | –0.021 | 0.063 |
| Language acculturation | 0.013 | 0.063 | –0.048 | 0.086 |
| Parents’ jobs | –0.003 | 0.103 | 0.061 | 0.092 |
| Parents’ education | 0.006 | 0.064 | –0.018 | 0.100 |
| Intercepts for latent factors | ||||
| Problems with alcohol at T4 | 0.000 | 0.000 | 0.232 | 0.184 |
| Growth curve intercept | 0.526*** | 0.032 | 0.496*** | 0.032 |
| Growth curve slope | 0.495*** | 0.059 | 0.441*** | 0.104 |
| Residual variances | ||||
| T4 problems 1 | 0.500*** | 0.041 | 0.482*** | 0.050 |
| T4 problems 2 | 0.522*** | 0.054 | 0.481*** | 0.080 |
| T4 problems 3 | 0.502*** | 0.054 | 0.458*** | 0.071 |
| T4 problems 4 | 0.462*** | 0.074 | 0.392*** | 0.058 |
| T1 alcohol use | .0.424*** | 0.070 | 0.326*** | 0.063 |
| T2 alcohol use | 0.513*** | 0.049 | 0.631*** | 0.046 |
| T3 alcohol use | 0.404*** | 0.075 | 0.648*** | 0.067 |
| Intercept | 0.386*** | 0.096 | .0.253** | 0.077 |
| Slope | 0.921*** | 0.055 | 0.686*** | 0.087 |
| T4 alcohol-related problems | 0.661*** | 0.061 | 0.849*** | 0.054 |
| Effects from Shows to Problems | ||||
| Total | 0.017 | 0.051 | 0.008 | 0.046 |
| Total indirect | 0.070 | 0.042 | 0.015 | 0.028 |
| Indirect Shows – I – Problems | –0.021 | 0.018 | –0.005 | 0.009 |
| Indirect Shows – S – Problems | 0.091* | 0.040 | 0.02 | 0.028 |
| Direct Shows – Problems | –0.054 | 0.065 | –0.007 | 0.058 |
| Effects from Liking to Problems | ||||
| Total | –0.007 | 0.063 | –0.042 | 0.041 |
| Total indirect | 0.095 | 0.057 | 0.053 | 0.058 |
| Indirect Liking – I – Problems | 0.105* | 0.048 | 0.030 | 0.052 |
| Indirect Liking – S – Problems | –0.010 | 0.038 | 0.023 | 0.030 |
| Direct Liking – Problems | –0.102 | 0.064 | –0.095 | 0.062 |
| Effects from Interaction SxL to Problems | ||||
| Total | 0.036 | 0.066 | 0.164* | 0.069 |
| Total indirect | –0.004 | 0.040 | –0.004 | 0.045 |
| Indirect from SxL – I – Problems | 0.036 | 0.026 | 0.016 | 0.032 |
| Indirect from SxL – S – Problems | –0.039 | 0.031 | –0.020 | 0.029 |
| Direct from SxL – Problems | 0.040 | 0.072 | 0.167 | 0.094 |
I, intercept factor for growth curve; na, not available, slope variance fixed at 0; S, slope factor for growth curve; SxL, interaction term for popular shows and liking of alcohol ads; T1, time 1; T2, time 2; T3, time 3.
P < .05.
P < .01.
< .001.
FIGURE 2.
Mediation model for alcohol-related problems. Alcohol use = past 30 days + past 6 months. I, growth curve intercepts; S, growth curve slopes. Standardized parameter estimates: boys/girls (P < .05). Paths that were nonsignificant for both boys and girls are not included in the figure for clarity (eg, the direct effect of popular shows on wave 4 problems was not significant and is not shown). Adjusted for wave 1 problems, age, drinking peers, drinking adults, playing sports, general TV watching, acculturation, parents’ jobs, parents’ education, and clustering by school. Fit indices: χ2(130) = 182.66, P = .002; Comparative Fit Index = 0.98; Tucker-Lewis Index = 0.97; Root Mean Squared Error of Approximation = .015; Standardized Root Mean Square Residual = .026. ns = non-significant.
FIGURE 3.
Interaction of exposure to ads with liking of ads. Liking of ads plotted at the mean, the mean plus 1 SD, and the mean minus 1 SD.
Significant mediation effects or indirect effects were observed among girls for the path from exposure to ads on popular shows at time 1 through the growth curve slopes to problems at time 4 (δ method indirect effect: ab = 0.091, P = .02) and for the path from liking of ads at time 1 through the growth curve intercepts to problems at time 4 (ab = 0.105, P = .03). Among boys, there was a significant total effect of the interaction term for popular shows and liking of ads at time 1 on problems at time 4, which included the direct effect on time 4 problems and indirect effects through the intercept and slope (δ method total effect: b = 0.164, P = .02). These effects among girls and boys were significant even after adjustment for time 1 problems, age, friends drinking, adults drinking, playing sports, general TV watching, acculturation, parents’ jobs, parents’ education, and clustering by school.
The covariates, alcohol-related problems at time 1 and friends and close adult drinking at time 1, were significant predictors of the intercept for girls. The same covariates plus language acculturation and parent jobs were significant predictors of the intercept for boys. For boys, drinking by friends and language acculturation were significant predictors of the slope, and the sign of the coefficients for these predictors changed between the intercept and the slope, suggesting that those higher in alcohol use at time 1 might have had lower growth rates than those lower in use at time 1. None of the time 1 variables were significant direct-effect (unmediated) predictors of alcohol-related problems at time 4 for boys or girls.
Mediation models for the other 3 exposure measures (frequency of watching sports show, cued recall of ads, and self-reported frequency of seeing alcohol ads) fit the data very well (results not shown). In all 3 models for girls, the intercept for the growth of alcohol use mediated the influence of liking of alcohol ads at time 1 on alcohol-related problems at time 4. No other indirect effects were significant for girls or boys. In these 3 mediation models for girls, both the intercept and slope for the growth of alcohol use were positive predictors of the level of alcohol-related problems at time 4, whereas this was not the case for boys.
Discussion
This study provides evidence supporting the hypothesis that exposure to alcohol advertising and affective reactions to those advertisements on television influence underage drinking and the development of alcohol-related problems. The growth of alcohol use from the seventh through the ninth grades is predicted by the frequency of watching popular shows and self-reports on the liking of alcohol ads. In partial support of hypothesis 1, there is a significant interaction between exposure to ads and liking of ads in the prediction of the intercept (but not the slope) for a growth curve modeled across these grade levels for both male and female students. The interaction shows that the level of exposure to ads is more predictive of a higher level of alcohol use in seventh grade for those students who report a greater liking of alcohol ads. In addition to this interaction observed at time 1, the frequency of watching popular shows at time 1 predicts the slope for the growth of alcohol use for girls, and the liking of alcohol ads at time 1 predicts the slope for boys.
In support of hypothesis 2, the mediation model shows that the influence of alcohol ads at time 1 on the occurrence of alcohol-related problems at time 4 is mediated by the growth of alcohol use. Among girls, there was a significant indirect effect of exposure to ads on popular shows in time 1 on problems in time 4 through the growth of alcohol use, and among boys, there was a significant total effect from the shows and liking interaction term in time 1 to problems in time 4. These relationships are significant even after adjusting for a range of other covariates measured at time 1 that are known to be associated with alcohol use. The other 3 measures of exposure to alcohol advertising show similar findings, although these measures are somewhat less predictive of the growth in alcohol use and alcohol-related problems.
Although causality cannot be verified in 1 observational study, the relevant theories and empirical evidence from the current prospective study and previous research are consistent with possible causal effects linking alcohol advertising to underage alcohol use and alcohol-related problems. In the current study, measures of exposure at time 1 are associated with the increasing use of alcohol over time and the development of alcohol-related problems at time 4, demonstrating a temporal ordering of predictors and outcomes. In addition, the models for this study control for a range of potentially confounding variables, including strong predictors, such as previous alcohol-related problems and peer influences. In previous studies, the indirect measure of exposure to alcohol ads on popular shows is predictive of alcohol use22,26 and measures for liking of alcohol ads are predictive of alcohol use.5,6,10,15
The findings here are also consistent with well-established theories on vicarious learning, such as Social Learning Theory,44 theories on persuasive messages in the media, such as the Elaboration Likelihood Model,16 and with the more recent Message Interpretation Process model by Austin and colleagues.5 Austin and colleagues5 provide evidence for the influence of alcohol advertising on alcohol use through a number of affective mediators, including liking of advertisements.5,45 Liking or desirability of alcohol advertisements predicts identification with portrayals of alcohol use in advertisements, which, in turn, predicts liking of brands of beer and positive expectancies for alcohol use. The overall influence of liking of advertisements on alcohol use might be somewhat larger in the current model if these mediating pathways were taken into account. In another study of advertising, Austin et al46 found that a media-literacy intervention increased skepticism (reduced liking) for advertising, as expected, but also increased recall of advertisement. This is consistent with the current study where memory and liking of advertisements interact. That is, a greater memory for alcohol advertisement does not necessarily mean an increase in alcohol use; it also depends on liking of the advertisements. This combination of theory and empirical evidence across research teams provides reasonably good support for the influence of exposure to alcohol advertisements on alcohol use and alcohol-related problems among adolescents.
A few limitations warrant discussion. First, the current results may be generalized only to public school students in the Los Angeles area. Second, alcohol use measures among young adolescents are often skewed toward 0, and this is true in the current sample. Seventh graders were actually recruited because of their low levels of alcohol use to examine the early development of alcohol use, but, unfortunately, these skewed measures may have contributed, in part, to some of the null findings in this study. Finally, not all results converge across multiple measures of exposure to advertising, but there is little literature available that indicates which exposure measures are optimal. However, it may not be surprising that cued recall of advertisements was not predictive of alcohol use. In the communication theory of Lang,47 cued recall is thought to be a less effective measure of retrieval/accessibility of information than it is a measure of encoding/availability of information.47 In encoding specificity48 and transfer-appropriate processing49 views, cued recall would reflect good accessibility and predictability at the time of drinking decisions only if the retrieval cues at test overlap well with retrieval cues during these later decisions; such overlap is unlikely, as the test cues were still pictures of commercials. However, the use of the indirect measure of exposure on popular shows and liking of ads are used successfully across a range of studies, and, in particular, liking of ads, although not strictly a measure of exposure, is used across product categories to predict the success of individual ads or ad campaigns.18
Conclusions
The accumulation of evidence for the influence of televised alcohol advertisements on underage drinking has important implications for prevention. First, children can be taught about the design of persuasive messages in the media early to help them avoid undue influence by the media on their behaviors.45,50 Second, it is important to have a comprehensive policy to limit the exposure of children to alcohol ads on television and on other media, such as the Internet, print media, and display ads. Although there are other influences on underage drinking, including those of peers and adults, prevention strategies should address the influence of alcohol ads as part of an overall strategy to prevent early initiation of alcohol use and the development of problems related to consumption.
Acknowledgments
The authors thank James Pike for his support on this project. We also thank Nielsen Media Research for providing information on alcohol commercials shown during specific television programs and the viewing ratings of those programs.
Glossary
- CI
confidence interval
- OR
odds ratio
APPENDIX.
Assessments
| Assessment | Items | α | Example Item | Response Option Anchors |
|---|---|---|---|---|
| Current frequency and quantity of alcohol use20 | 9 | 0.91 | During the last 30 d, on how many days did you…have at least 1 drink of beer? | 0 = 0 d |
| 7 = all 30 d | ||||
| Problems associated with alcohol use21 | 8 | 0.93 | How many times have you ever…gone to school drunk? | 1 = never |
| 4 = more than 10 times | ||||
| Exposure to alcohol advertisinga on popular shows22 | 20 | 0.79 | How frequently do you watch MTV? | 1 = never |
| 6 = every day | ||||
| Exposure to alcohol advertisinga on sports shows22,23 | 6 | 0.80 | How often to you watch professional football? | 1 = never |
| 6 = every day | ||||
| Cued recall memory for alcohol advertisements24 | 15 | 0.74 | What product is being advertised in the photo? | Open-ended |
| Self-reported observation of alcohol advertisements25 | 4 | 0.72 | In the past week, how many commercials have you seen for alcohol drinks like beer, wine, or liquor? | 0 = none |
| 6 = 6 or more | ||||
| Liking of alcohol advertisements26 | 3 | 0.78 | Of all the commercials you see on TV, how much do you like the TV commercials for alcohol? | 1 = I like alcohol commercials the most |
| 4 = I like the alcohol commercials the least | ||||
| Propensity to watch TV27,28 | 7 | 0.79 | On a typical weekday, how many hours a day do you watch TV…after school before dinner? | 1 = I do not watch TV |
| 5 = 5 h or more | ||||
| Observed drinking by peers and friends29 | 4 | 0.86 | About how often did you do the following things in the last 6 mo…saw someone your age drink beer or other alcohol? | 0 = never |
| 6 = every day | ||||
| Observed drinking by known adults30 | 3 | 0.84 | About how often did you do the following things in the last 6 mo…saw an adult you know well drink alcohol? | 0 = never |
| 6 = every day | ||||
| Participation in sports31 | 5 | 0.73 | About how often did you do the following things in the last 6 mo… played soccer? | 0 = never |
| 6 = every day | ||||
| Language acculturation32,33 | 3 | 0.67 | What language(s) do you usually speak at home? | 1 = only English |
| 5 = only another language | ||||
| Socioeconomic status51 | 2 | na | What is the highest grade completed by your mother? | 1 = not completed elementary school |
| 6 = Completed graduate school | ||||
| Socioeconomic status51 | 2 | na | What type of work does your father do? | Open-ended (coded) |
na, not applicable.
The frequency of watching popular shows or sports programs was weighted by the frequency of alcohol advertisements broadcast on those shows in the previous 10 mo, as reported by Nielsen Media Research (see text).
Footnotes
Dr Grenard contributed to the conception of the statistical model, analyzed the data, and prepared the manuscript; Dr Dent contributed to the acquisition of data and analysis of the data, revised the methods and analysis sections of the document, and provided final approval of the manuscript; and Dr Stacy contributed to the conception and design of the study, revised the introduction and discussion sections for intellectual content, and approved the final version of the manuscript.
FINANCIAL DISCLOSURES: The authors have indicated that they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by grants from the National Institute on Alcohol Abuse and Alcoholism (AA12128) and the National Institute on Drug Abuse (DA16094, DA023368, and DA024659). Funded by the National Institutes of Health (NIH).
References
- 1.SAMHSA. Results from the 2003 national survey on drug use and health: National Findings. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2004. Report No. DSDUH Series H-25, DHHS Publication No. SMA 04-3964. [Google Scholar]
- 2.Lewinsohn PM, Rohde P, Seeley JR. Alcohol consumption in high school adolescents: frequency of use and dimensional structure of associated problems. Addiction. 1996;91(3):375–390 [DOI] [PubMed] [Google Scholar]
- 3.Atkin CK, Hocking J, Block M. Teenage drinking: Does advertising make a difference? Journal of Communication. 1984;34(2):157–167 [Google Scholar]
- 4.Adlaf EM, Kohn PM. Alcohol advertising, consumption and abuse: a covariance-structural modelling look at Strickland’s data. Br J Addict. 1989;84(7):749–757 [DOI] [PubMed] [Google Scholar]
- 5.Austin EW, Chen M-J, Grube JW. How does alcohol advertising influence underage drinking? The role of desirability, identification and skepticism. J Adolesc Health. 2006;38(4):376–384 [DOI] [PubMed] [Google Scholar]
- 6.Wyllie A, Zhang JF, Casswell S. Responses to televised alcohol advertisements associated with drinking behaviour of 10-17-year-olds. Addiction. 1998;93(3):361–371 [DOI] [PubMed] [Google Scholar]
- 7.Connolly GM, Casswell S, Zhang J-F, Silva PA. Alcohol in the mass media and drinking by adolescents: a longitudinal study. Addiction. 1994;89(10):1255–1263 [DOI] [PubMed] [Google Scholar]
- 8.Ellickson PL, Collins RL, Hambarsoomians K, McCaffrey DF. Does alcohol advertising promote adolescent drinking? Results from a longitudinal assessment. Addiction. 2005;100(2):235–246 [DOI] [PubMed] [Google Scholar]
- 9.Snyder LB, Milici FF, Slater M, Sun H, Strizhakova Y. Effects of alcohol advertising exposure on drinking among youth. Arch Pediatr Adolesc Med. 2006;160(1):18–24 [DOI] [PubMed] [Google Scholar]
- 10.Casswell S, Zhang J-F. Impact of liking for advertising and brand allegiance on drinking and alcohol-related aggression: a longitudinal study. Addiction. 1998;93(8):1209–1217 [DOI] [PubMed] [Google Scholar]
- 11.Hanewinkel R, Sargent JD. Longitudinal study of exposure to entertainment media and alcohol use among German adolescents. Pediatrics. 2009;123(3):989–995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zogg JB. Adolescent Exposure to Alcohol Advertising: A Prospective Extension of Strickland's Model [doctoral dissertation]. Los Angeles, CA: University of Southern California; 2004 [Google Scholar]
- 13.Stacy AW, Zogg JB, Unger JB, Dent CW. Exposure to televised alcohol ads and subsequent adolescent alcohol use. Am J Health Behav. 2004;28(6):498–509 [DOI] [PubMed] [Google Scholar]
- 14.Anderson P, de Bruijn A, Angus K, Gordon R, Hastings G. Impact of alcohol advertising and media exposure on adolescent alcohol use: a systematic review of longitudinal studies. Alcohol Alcohol. 2009;44(3):229–243 [DOI] [PubMed] [Google Scholar]
- 15.Chen MJ, Grube JW, Bersamin M, Waiters E, Keefe DB. Alcohol advertising: what makes it attractive to youth? J Health Commun. 2005;10(6):553–565 [DOI] [PubMed] [Google Scholar]
- 16.Petty RE, Wegener DT. The elaboration likelihood model: current status and controversies. In: Chaiken S, Trope Y, eds. Dual-Process Theories in Social Psychology. New York, NY: Guilford Press; 1999:37–72 [Google Scholar]
- 17.Henriksen L, Feighery EC, Schleicher NC, Fortmann SP. Receptivity to alcohol marketing predicts initiation of alcohol use. J Adolesc Health. 2008;42(1):28–35 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Haley RI, Baldinger AL. The ARF Copy Research Validity Project. J Advert Res. 1991;31(2):11–32 [Google Scholar]
- 19.Wyllie A, Zhang JF, Casswell S. Positive responses to televised beer advertisements associated with drinking and problems reported by 18 to 29-year-olds. Addiction. 1998;93(5):749–760 [DOI] [PubMed] [Google Scholar]
- 20.Kann L. The Youth Risk Behavior Surveillance System: measuring health-risk behaviors. Am J Health Behav. 2001;25(3):272–277 [DOI] [PubMed] [Google Scholar]
- 21.Winters KC, Stinchfield RD, Henly GA. Further validation of new scales measuring adolescent alcohol and other drug abuse. J Stud Alcohol. 1993;54(5):534–541 [DOI] [PubMed] [Google Scholar]
- 22.Strickland DE. Advertising exposure, alcohol consumption and misuse of alcohol. In: Grant M, Plant M, Williams A, eds. Economics and Alcohol: Consumption and Controls. New York, NY: Gardner Press; 1983:201–222 [Google Scholar]
- 23.Madden PA, Grube JW. The frequency and nature of alcohol and tobacco advertising in televised sports, 1990 through 1992. Am J Public Health. 1994;84(2):297–299 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Unger JB, Johnson CA, Rohrbach LA. Recognition and liking of tobacco and alcohol advertisements among adolescents: relationships with susceptibility to substance use. Prev Med. 1995;24(5):461–466 [DOI] [PubMed] [Google Scholar]
- 25.Schooler C, Feighery E, Flora JA. Seventh graders’ self-reported exposure to cigarette marketing and its relationship to their smoking behavior. Am J Public Health. 1996;86(9):1216–1221 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Unger JB, Schuster D, Zogg JB, Dent CW, Stacy AW. Alcohol advertising exposure and adolescent alcohol use: a comparison of exposure measures. Addict Res Theory. 2003;11(3):177–193 [Google Scholar]
- 27.Robinson TN, Chen HL, Killen JD. Television and music video exposure and risk of adolescent alcohol use. Pediatrics. 1998;102(5). Available at: www.pediatrics.org/cgi/content/full/102/5/E54 [DOI] [PubMed] [Google Scholar]
- 28.Grube JW. Television alcohol portrayals, alcohol advertising, and alcohol expectancies among children and adolescents. In: Martin SE, ed. The Effects of the Mass Media on the Use and Abuse of Alcohol. Bethesda, MD: National Institutes of Health; 1995. Research Monograph No. 28:105–121 [Google Scholar]
- 29.Feldman LA, Harvey B, Holowaty P, Shortt L. Alcohol use beliefs and behaviors among high school students. J Adolesc Health. 1999;24(1):48–58 [DOI] [PubMed] [Google Scholar]
- 30.Wood MD, Read JP, Mitchell RE, Brand NH. Do parents still matter? Parent and peer influences on alcohol involvement among recent high school graduates. Psychol Addict Behav. 2004;18(1):19–30 [DOI] [PubMed] [Google Scholar]
- 31.Thorlindsson T, Vilhjalmsson R, Valgeirsson G. Sport participation and perceived health status: a study of adolescents. Soc Sci Med. 1990;31(5):551–556 [DOI] [PubMed] [Google Scholar]
- 32.Marin G, Sabogal F, Marin BV, Otero-Sabogal R, Perez-Stable EJ. Development of a short acculturation scale for Hispanics. Hisp J Behav Sci. 1987;9(2):183–205 [Google Scholar]
- 33.Stacy AW. Memory association and ambiguous cues in models of alcohol and marijuana use. Exp Clin Psychopharmacol. 1995;3(2):183–194 [Google Scholar]
- 34.Anderson JC, Gerbing DW. Structural equation modeling in practice: a review and recommended two-step approach. Psychol Bull. 1988;103(3):411–423 [Google Scholar]
- 35.Vandenberg RJ, Lance CE. A review and synthesis of the measurement invariance literature: suggestions, practices, and recommendations for organizational research. Organ Res Methods. 2000;3(1):4–69 [Google Scholar]
- 36.Little TD, Cunningham WA, Shahar G, Widaman KF. To parcel or not to parcel: exploring the question, weighing the merits. Struct Equ Modeling. 2002;9(2):151–173 [Google Scholar]
- 37.Marsh HW, Hau K-T, Grayson D. Goodness of fit in structural equation models. In: Maydeu-Olivares A, McArdle JJ, eds. Contemporary Psychometrics: A Festschrift for Roderick P McDonald Multivariate Applications Book Series. Mahwah, NJ: Lawrence Erlbaum Associates Publishers; 2005:275–340 [Google Scholar]
- 38.Little RJA, Rubin DB. Statistical Analysis with Missing Data. 2nd ed. Hoboken, NJ: Wiley; 2002 [Google Scholar]
- 39.Bollen KA. Structural Equations with Latent Variables. Wiley Series in Probability and Mathematical Statistics. Applied Probability and Statistics section. Oxford, UK: John Wiley & Sons; 1989:514 [Google Scholar]
- 40.MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects. Psychol Methods. 2002;7(1):83–104 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Muthen LK, Muthen BO. Mplus user's guide. 5th ed. Los Angeles, CA: Muthen & Muthen; 1998-2007.
- 42.Bandalos DL, Finney SJ. Item parceling issues in structural equation modeling. In: Marcoulides GA, Schumacker RE, eds. New Developments and Techniques in Structural Equation Modeling. Mahwah, NJ: Lawrence Erlbaum Associates Publishers; 2001:269–296 [Google Scholar]
- 43.Gregorich SE. Do self-report instruments allow meaningful comparisons across diverse population groups? Testing measurement invariance using the confirmatory factor analysis framework. Med Care. 2006;44(11 suppl 3):S78–S94 [DOI] [PMC free article] [PubMed]
- 44.Bandura A. Social Learning Theory. Upper Saddle River, NJ: Prentice Hall; 1977 [Google Scholar]
- 45.Kupersmidt JB, Scull TM, Austin EW. Media literacy education for elementary school substance use prevention: study of media detective. Pediatrics. 2010;126(3):525–531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Austin EW, Chen YC, Pinkleton BE, Quintero Johnson J. Benefits and costs of Channel One in a middle school setting and the role of media-literacy training. Pediatrics. 2006;117(3). Available at: www.pediatrics.org/cgi/content/full/117/3/e423 [DOI] [PubMed] [Google Scholar]
- 47.Lang A. Defining audio video redundancy from a limited-capacity information-processing perspective. Communic Res. 1995;22(1):86–115 [Google Scholar]
- 48.Tulving E. Elements of Episodic Memory. New York, NY: Oxford University Press; 1983 [Google Scholar]
- 49.Roediger HL, III, Gallo DA, Geraci L. Processing approaches to cognition: the impetus from the levels-of-processing framework. Memory. 2002;10(5–6):319–332 [DOI] [PubMed] [Google Scholar]
- 50.Austin EW, Johnson KK. Effects of general and alcohol-specific media literacy training on children’s decision making about alcohol. J Health Commun. 1997;2(1):17–42 [DOI] [PubMed] [Google Scholar]
- 51.Hollingshead AB, Redlich FC. Social Class and Mental Illness: A Community Study. New York, NY: Wiley; 1958 [DOI] [PMC free article] [PubMed] [Google Scholar]