Examining How Presumed Media Influence Affects Social Norms and Adolescents' Attitudes and Drinking Behavior Intentions in Rural Thailand

Abstract

This study uses the influence of presumed media influence model as the theoretical framework to examine how perceived social norms (i.e., descriptive, subjective, and injunctive norms) will mediate the influence of pro- and antidrinking media messages on adolescents' intention to consume alcohol in rural Thailand. Data collected from 1,028 high school students indicate that different mechanisms underlie drinking intentions between nondrinkers and those who have consumed alcohol or currently drink. Among nondrinkers, perceived peer attention to prodrinking messages indirectly influenced adolescents' prodrinking attitudes and intentions to consume alcohol through all three types of perceived social norms. Among drinkers, perceived peer attention to pro- and antidrinking messages indirectly influenced adolescents' prodrinking attitudes and intentions to drink alcohol through perceived subjective norm. The findings provide support for the extended influence of presumed media influence model and have practical implications for how antidrinking campaigns targeted at teenagers in Thailand might be designed.

Acknowledgments

This work was supported by the Start-Up Grant (grant number: M4080204) from the Wee Kim Wee School of Communication and Information at Nanyang Technological University.

Notes

¹The sale of alcohol to consumers is allowed from 11 am to 2 pm and from 5 pm to 12 am daily.

²For example, one particular beer brand showed how Thai office workers should stay optimistic. The ad portrayed a friendly environment where everyone was willing to help one another in a work place. Beer never appeared in this ad; only its trademark appeared very briefly at the end (see the BenMore Office website at http://www.adintrend.com/show_ad.php?id=4239). The goal of alcoholic beverage ads is generally to promote and associate some kind of positive values or perceptions with the brand (Centre for Alcohol Studies, 2008; Kannika, 2009).

³Most campaign themes were relevant to adults, especially low-income workers. Before New Year celebrations, ads with themes related to the harms of drinking and driving are more frequently shown. There have also been ongoing efforts to stop people from giving alcoholic beverages as gifts in these ads (Thai Health Promotion Foundation, 2006).

⁴Another ad relates to the new Alcohol Control Act which bans drinking in educational institutions, conveying the key message that if anyone drinks within the school premises, they would be jailed (Thai Health Promotion Foundation, 2006).

⁵Most students who attended the small schools in Uttaradit tend to reside in rural areas, as these schools were generally located in the remote areas. Likewise, most students who attended the larger schools tend to live in urban areas, as these schools were located in town. Therefore, high school students from small schools were chosen for our study because our focus is on the rural population.

⁶Derived from the total number of students who attended school on the day the questionnaire was administered/total number of students in these schools.

⁷In Thailand, parental consent is not needed for surveys conducted in and approved by the schools.

⁸In smaller schools, the teachers assembled all the students into one big hall. In bigger schools, the teachers asked the research assistants to go to several classes to administer the paper-and-pencil survey at the same time. For all schools, everyone participated at the same time. Students were specifically asked not to look over each other's responses. In cases where the survey was conducted in a hall, students were asked to sit further apart from one another. Once the questionnaires were completed by the students, they were directly collected by our research assistants. The questionnaire was translated from English to Thai. The translation of all the questions was then checked with another translator in Thailand.

⁹Data from 34 participants were excluded from the analysis because of missing values and inconsistent responses.

¹⁰If the model fits the data well, χ² should not be significant (Kline, 2004). The χ² is divided by its degrees of freedom to adjust for its sensitivity to large sample size (Bentler, 1989). Relative χ² values that fall in between 1.0 and 5.0 are considered as acceptable fit (Bollen, 1989). For CFI and NFI, values that fall in between .95 and 1.00 are considered as good fit (Hu & Bentler, 1999). For RMSEA, values of .05 or less are considered as good fit (Browne & Cudeck, 1993).

*p < .05. **p < .01. **p < .001.

*p < .05. **p < .01. **p < .001.

¹¹For example, one popular campaign attempted to persuade male adults to quit drinking in order to get them out of poverty (Thai Health Promotion Foundation, 2008).

¹²One potential rival model may be that adolescents' perceived social norms would impact their perception of media effects on their peers. However, when we tested this rival explanation by switching the causal arrows between perceived peer attention to drinking messages and the various perceived social norms, the models produced poorer fit in comparison with our original models (for nondrinkers: χ² = 282.7, df = 19, p < .001, χ²/df = 14.9, RMSEA =.18, CFI =.60, NFI =.62; for drinkers: χ² = 367.1, df = 19, p < .001, χ²/df = 19.3, RMSEA =.18, CFI =.62, NFI =.63). Another rival explanation may be that adolescents' own attitude toward drinking may drive the amount of attention they pay to media messages on drinking. We tested this rival explanation by switching the directions of the factors, but the alternative models yielded poorer fit (for nondrinkers: χ² = 130.0, df = 21, p < .001, χ²/df = 6.19, RMSEA =.11, CFI =.83, NFI =.82; for drinkers: χ² = 183.1, df = 21, p < .001, χ²/df = 8.72, RMSEA =.12, CFI =.82, NFI =.81). It is also likely that adolescents use their own attitudes and behaviors as a basis to infer about norms among their peers (i.e., the projection effect). We tested this alternative explanation by swapping the causal arrows between those factors, but the rival models produced poorer fit in comparison (for nondrinkers: χ² = 291.5, df = 23, p < .001, χ²/df = 12.67, RMSEA =.17, CFI =.59, NFI =.60; for drinkers: χ² = 376.73, df = 23, p < .001, χ²/df = 16.38, RMSEA =.17, CFI =.61, NFI =.62). Therefore, the evidence we presented do not support the rival causal paths.