212
British Journal of Health Psychology (2015), 20, 212–220 © 2014 The British Psychological Society www.wileyonlinelibrary.com
The effects of social and health consequence framing on heavy drinking intentions among college students John H. Kingsbury1*†, Frederick X. Gibbons1‡ and Meg Gerrard2‡ 1
Department of Psychological and Brain Sciences, Dartmouth College, Hanover, New Hampshire, USA 2 Department of Psychiatry, Dartmouth Medical School, Lebanon, New Hampshire, USA Objectives. Many interventions targeting college student drinking have focused on negative health effects of drinking heavily; however, some research suggests that social factors may have a stronger influence on the drinking behaviour of young people. Moreover, few studies have examined message framing effects in the context of alcohol consumption. This study investigated the effects of social and health consequence framing on college students’ intentions to engage in heavy drinking. Design. This study used a 2 9 2 experimental design with an appended control condition. Methods. One hundred and twenty-four college students (74 women; Mage = 18.9) participated in this study for course credit. Participants read vignettes that were ostensibly written by a recent graduate from the university, who described an episode of drinking in which he or she experienced either social or health consequences. These consequences were framed as either a gain (i.e., positive consequences of not drinking heavily) or a loss (i.e., negative consequences of drinking heavily). After reading the vignette, participants completed a measure of heavy drinking intentions. Results. Regression analyses revealed that social consequences were associated with lower heavy drinking intentions when framed as a loss and that health consequences were associated with lower heavy drinking intentions when framed as a gain. These effects were stronger among those who reported higher (vs. lower) levels of previous drinking. Conclusions. Results suggest that interventions that focus on the negative health effects of heavy drinking may be improved by instead emphasizing the negative social consequences of drinking heavily and the positive health consequences of avoiding this behaviour.
Statement of contribution What is already known on this subject? Previous studies have shown that gain frames are more effective than loss frames when highlighting the health consequences of health risk behaviours, such as heavy drinking. The heavy drinking behaviour of young people is influenced by social factors (e.g., perceived social consequences). However, little is known about framing effects for social consequences of heavy drinking.
†
Minnesota Department of Health, St Paul, Minnesota, USA Department of Psychology, University of Connecticut, Storrs, Connecticut, USA
‡
Correspondence should be addressed to John H. Kingsbury, Minnesota Department of Health, 85 7th Place East, 4th Floor, St Paul, MN 55102, USA (email: [email protected]). DOI:10.1111/bjhp.12100
Social and health consequence framing
213
What does this study add? This study builds on previous research by demonstrating that a loss frame is more effective than a gain frame when highlighting the social consequences of health risk behaviour. Framing effects are strongest for those with more previous drinking experience.
Excessive alcohol consumption is a serious public health problem, especially on college campuses. There are over 1,800 alcohol-related deaths and nearly 600,000 alcohol-related injuries each year among U.S. college students aged 18–24 (Hingson, Zha, & Weitzman, 2009). Intervention efforts to reduce drinking among young people commonly present information about, or images of, the negative health consequences of drinking (e.g., pictures of a drunk driving accident, or an individual who has passed out from heavy drinking). However, these messages are often met with resistance or are rejected (Brehm, 1966; Ringold, 2002). In some cases, particularly among high-risk individuals, these messages have had iatrogenic effects, that is, they have produced increases in risk behaviour (Miller, Burgoon, Grandpre, & Alvaro, 2006; Werch & Owen, 2002).
Framing Health consequences The framing literature suggests that, rather than focusing on the negative health consequences of drinking, focusing on the positive consequences of not drinking may be more effective (Rothman & Salovey, 1997). This approach is consistent with prospect theory (Tversky & Kahneman, 1981), which posits that behaviours with more certain outcomes, such as flossing, are more influenced by potential positive consequences – a gain frame (e.g., reduced risk of gum disease), whereas behaviours with less certain outcomes, such as getting a mammogram, are more influenced by potential negative consequences – a loss frame (e.g., a more severe diagnosis). A recent review found some support for this hypothesis. Gallagher and Updegraff (2012) reported that few studies had examined framing effects for health risk behaviours (k = 4), but those that did found that gain frames were associated with less risk behaviour than loss frames (Gerend & Cullen, 2008; Schneider et al., 2001; Steward, Schneider, Pizarro, & Salovey, 2006). Moreover, effects for gain frame messages on smoking and alcohol use were among the largest observed in this meta-analysis. Thus, while research on framing and risk behaviours is scant, evidence suggests that gain frames may be more effective than loss frames. Questions also remain about what information should be included in messages aimed at reducing risk behaviour. Existing studies examining framing effects for risk behaviours have primarily focused on the health consequences of these behaviours (e.g., gain frame: maintaining a healthy weight, having a healthy liver; loss frame: increased likelihood of risky sexual behaviour, getting in a car accident; Gerend & Cullen, 2008). However, evidence suggests that the decision to drink heavily among young people may be more strongly influenced by social factors (Read, Wood, Kahler, Maddock, & Palfai, 2003). College students report that drinking promotes social facilitation and a sense of belonging with fellow students (Gilles, Turk, & Fresco, 2006; Labrie, Hummer, & Pedersen, 2007). Heavy drinking is particularly common for first-year college students, who are trying to make friends and adjust to a new social environment (Borsari, Murphy, & Barnett, 2007).
214
John H. Kingsbury et al.
In short, heavy drinking among college students is frequently socially motivated. That being the case, the current study tried to use this same (social) motivation to reduce the behaviour.
Social consequences Few studies have investigated framing effects for the social consequences of risk behaviour. One exception is a study by Kiene, Barta, Zelenski, and Cothran (2005), which examined framing effects for a behaviour that has both interpersonal and health implications, condom use (Rothman & Salovey, 1997). They found that persuasive statements about the potential interpersonal consequences of condom use (e.g., a partner’s reaction to suggested condom use) were rated as more convincing and important when framed as a loss than as a gain. Conversely, messages featuring health consequences of condom use were rated as more convincing and important when framed as a gain. The authors explained the pattern of results for social consequences by citing the emotional risk involved in discussing condom use with a partner, that is, the partner may or may not respond negatively. These effects were stronger for those who were higher in issue involvement, suggesting that those who view the consequences as more personally relevant may be more influenced by message characteristics. These results are consistent with research demonstrating that perceived relevance is a key determinant of persuasive message effectiveness (Petty & Cacioppo, 1986). Results of the Kiene et al.’s (2005) study suggest that framing the social consequences of risk behaviour as a loss may be more effective than framing them as a gain. However, this study had participants rating the perceived effectiveness of the messages rather than assessing effects on personal attitudes or intentions to engage in the behaviour, which limits the applicability of these findings to behaviour change. Furthermore, this study utilized a within-subjects design, which may have affected the independence of the ratings for each type of message. The current study aimed to address these limitations and extend the literature, by using an experimental design to investigate framing effects for social and health consequences on college students’ intentions to drink heavily. It was hypothesized that health consequences would be associated with lower intentions when framed as a gain and that social consequences would be associated with lower intentions when framed as a loss. These effects were expected to be stronger among those for whom the behaviour and the consequences would be more relevant (i.e., those with more drinking experience).
Method Participants One hundred and twenty-four undergraduates (74 women; Mage = 18.9) who were enrolled in introductory psychology classes participated in this study for course credit. This study was approved by the IRB at the participating university.
Experimental design and procedure Participants were randomly assigned to one of five conditions in this study, which employed a 2 (consequences: social, health) 9 2 (frame: gain, loss) design with an appended control condition. Participants were told that they would be evaluating vignettes that were written by similar others (i.e., alumni of their university). Vignettes
Social and health consequence framing
215
were used to convey information about drinking consequences because they are less likely to prompt optimistic bias, which can influence the effectiveness of persuasive messages (Klein et al., 2010). Participants in the control group read a vignette that featured a recent alumnus who had difficulty with study habits and schoolwork, but improved his or her school performance after attending a talk by a successful alumnus from the university (see Appendix S1). The control vignette focused on schoolwork troubles because this is a common issue for college students and is unrelated to alcohol. Participants in the experimental conditions read the control vignette and one that focused on alcohol. Alcohol vignettes featured a recent alumnus who went out with friends and experienced social or health consequences of drinking. The consequences were presented using either a gain frame (i.e., positives of not drinking heavily) or loss frame (i.e., negatives of drinking heavily). The social–gain frame vignette featured a student who was able to more easily converse with others, leading to a future date with an attractive member of the opposite sex. The student in the health–gain frame vignette talked about feeling good the day after going out, and the health benefits of moderate drinking (e.g., lower risk of heart disease). Social–loss vignette featured a student who drank so much that he or she vomited in front of an attractive member of the opposite sex. The student in the health–loss vignette blacked out and was taken to the hospital for fear of alcohol poisoning.
Measures Participants rated the vignettes on different dimensions: how typical the student’s experience was, how easy it was to imagine the consequences happening to the self, and how similar the experience was to a previous personal experience (1 = not at all, 7 = very). Two items were used to assess past drinking: ‘How often have you consumed alcohol in the past four weeks?’ (1 = not at all, 6 = more than a few times per week) and ‘How often have you had four or more drinks in a 2-hr span over the past four weeks?’ (1 = 0 times, 6 = more than 4 times). These two items were combined to form a composite scale of past drinking (a = .80). The primary outcome assessed intentions to drink heavily using a conditional item. Previous studies have demonstrated that conditional items are better predictors of risk behaviour than those that are more global (i.e., not situation specific; Gibbons, Gerrard, & Lane, 2003; Pomery, Gibbons, Reis-Bergan, & Gerrard, 2009). Intentions to drink heavily were assessed with a single face-valid item: ‘Imagine you are at a party where alcohol is being served. . .You have had several drinks over the past 2 hr and you have decided to stop drinking. A friend bumps into you, and you begin a conversation. The friend offers to get you another drink’. Participants were asked, ‘In this situation, would you plan to continue drinking?’ 1 = definitely not, 7 = definitely.
Results Descriptive results Participants’ reports of their previous drinking were comparable to national norms (SAMHSA, 2011). They reported that, on average, they had consumed alcohol 2–4 times, and had drunk heavily 1–2 times, in the past month. Twenty-two participants (18%) reported abstaining from drinking in the past 4 weeks. Past drinking did not differ across condition (p = .77).
216
John H. Kingsbury et al.
Ratings of the alcohol consequences were examined to determine whether the vignettes differed on dimensions other than those that were specifically manipulated. Consequences in the gain frame vignettes were rated as easier to imagine (M = 4.94 vs. 2.98) and more similar to personal experiences (M = 4.09 vs. 2.17) than those in the loss frame vignettes (all ps < .001). There were no differences between the social/health vignettes on these dimensions (ps > .29), nor any Frame 9 Consequences interactions. Past drinking was unrelated to ease of imagination and similarity to personal experience and was positively associated with how typical the experience was (p < .01). Frame, consequences, and past drinking Control comparisons Planned contrasts were used to examine differences in intentions to drink heavily among the four experimental conditions and the control group. Results revealed that, as predicted, those in the social–loss and health–gain conditions (Ms = 2.22, 2.53, respectively) reported significantly lower intentions than did those in the control group, M = 3.33; F(1, 118) = 5.61, p = .02. Conversely, intentions to drink heavily in the social– gain and health–loss conditions (Ms = 3.00, 3.05) did not differ from those in the control group (F < 0.49, p = .49).
Moderation by past drinking Linear regression analysis was used to test whether past drinking moderated the effects of frame and consequences on intentions to drink heavily. Dummy codes were used for the consequences and frame conditions, and the control group was assigned a coefficient of ‘0’, so that responses from these participants were factored into the error term, but did not expend a degree of freedom (Hornbeck, 1973). Past drinking was entered as a continuous variable, and two- and three-way interaction terms were created to estimate interactive effects. Results revealed no significant main effects of frame or consequences (ps > .34), but a significant positive effect of past drinking (p < .001), that is, participants with more drinking experience reported higher intentions to drink heavily. Past drinking did not interact with frame or consequences (ps > .43); however, the Frame 9 Consequences interaction was significant, b = .17, t(116) = 2.16, p = .03. Simple effects analyses revealed that there were no framing effects for health consequences, b = .16, t (47) = 1.15, p = .26, but the loss frame led to lower intentions than did the gain frame for social consequences, b = .25, t(47) = 2.19, p = .03. The Frame 9 Consequences interaction was further qualified by a Past Drinking 9 Frame 9 Consequences interaction, b = .24, t(116) = 2.94, p = .004, which followed the predicted pattern (Figure 1). Simple effects analyses indicated that, for social consequences, the loss frame was associated with lower intentions than the gain frame for those higher in past drinking (i.e., +1 SD), b = .47, t(47) = 2.85, p = .007, but not for those lower in past drinking (i.e., 1 SD), b = .02, t(47) = .12, p = .90. Conversely, for health consequences, the gain frame was associated with lower intentions than the loss frame for those higher in past drinking, b = .60, t(47) = 3.11, p = .003, but not for those lower in past drinking, b = .22, t(47) = 1.23, p = .23. In sum, consistent with hypotheses, results indicate that: (1) the social–loss frame was more effective than the social–gain frame, (2) the health– gain frame was more effective than the health–loss frame, and (3) these effects were stronger for those reporting higher levels of past drinking.
Social and health consequence framing High Past Drinking (+1 SD) ** **
Intentions to Drink Heavily
5 4.5
Loss
4
Gain
3.5 3 2.5 2 1.5 1 Social
Health
Low in Past Drinking (-1 SD)
5 Intentions to Drink Heavily
217
4.5
Loss
4
Gain
3.5 3 2.5 2 1.5 1 Social
Health
Figure 1. Interaction between consequences and frame predicting intentions to drink heavily by past drinking experience. Note. **p < .01. High and low in past drinking are defined as plus/minus one standard deviation, respectively.
Discussion Findings from this study are consistent with prospect theory and previous studies demonstrating that gain frames are more effective than loss frames when highlighting the health consequences of risk behaviours (Gerend & Cullen, 2008; Schneider et al., 2001). The finding that social consequences were more effective when framed as a loss is consistent with a study by Kiene et al. (2005) that found that loss-framed messages about interpersonal discussions of condom use were rated as more convincing and important than gain-framed messages. The current study extends these findings by demonstrating framing effects for social consequences on heavy drinking intentions and by showing that the social–loss- and health–gain-framed messages were associated with lower drinking intentions than those reported by a control group.
Previous drinking experience Framing effects were stronger for those who reported higher levels of previous drinking. The perceived relevance of the behaviour in the vignettes may have contributed to relatively strong effects for those reporting more previous drinking. For those who were higher in previous drinking and read about health consequences, the gain frame was associated with much lower intentions to drink than the loss frame. These results lend support to previous findings that loss-framed messages featuring health consequences
218
John H. Kingsbury et al.
may be met with resistance, and thus, are less successful, when presented to high-risk individuals (Detweiler, Bedell, Salovey, Rothman, & Pronin, 1999; Stock, Gibbons, & Gerrard, 2012). Collectively, the vignettes appeared to have weaker effects on those with less drinking experience. It is possible that the consequences in the vignettes, which talked about engaging (or not engaging) in heavy drinking, were viewed as less relevant among this group. Another plausible explanation for these findings is that the drinking intentions of college students who choose to drink less heavily are less malleable than those who drink heavily. More work is needed to test these hypotheses. The advantage of the social–loss frame over the social–gain frame for those higher in previous drinking is consistent with one of the basic tenets of prospect theory: individuals are more influenced by potential positive consequences when considering behaviours with more certain outcomes and more influenced by potential negative consequences when considering behaviours with less certain outcomes. The health consequences of not drinking heavily are relatively certain, that is, not drinking heavily is clearly associated with better health than is drinking heavily. Thus, prospect theory would predict an advantage for gain frames for health consequences (Rothman & Salovey, 1997). Conversely, social consequences of not drinking heavily may be, in some sense, uncertain or ‘risky’ for heavy drinkers. A heavy drinker who chooses not to drink may be unsure of how his or her friends, who may be heavy drinkers themselves, will react. Thus, social–loss-framed messages may effectively take a behaviour that is frequently performed to facilitate social relationships, and make it seem undesirable by highlighting ways in which it can harm social standing. In essence, this approach ‘uses’ the same motivation that promotes the behaviour to inhibit it (Gerrard et al., 2006; Gibbons, Gerrard, Lane, Mahler, & Kulik, 2005; Mahler, Kulik, Gibbons, Gerrard, & Harrell, 2003). Further research is needed to test this hypothesis.
Limitations An important limitation of this study is that it did not assess behaviour change. Previous studies using framing manipulations have detected changes in behaviour at 1-month follow-ups (Gerend & Cullen, 2008), so we expect that the messages used in this study might have some effect on behaviour change, but this assumption should be tested in future studies. This study would have been strengthened by assessing the perceived severity of consequences, as some of the consequences may have been viewed as more extreme than others, which may have influenced intentions to drink. Research examining the content of framed messages has found that the framing component of a message (whether it is gain or loss) is more important than the action or outcome described (Detweiler et al., 1999), so we expect that perceived severity would not influence. More generally, the current study did not identify mediating mechanisms that explain why differently framed messages had an effect on intentions. More work is needed to address these issues. Finally, the drinking intentions construct consisted of only a single item. Although the item is face valid, additional measures of drinking intentions and drinking willingness (the latter being particularly effective for risky behaviours, such as heavy drinking; Gibbons et al., 2003; Spijkerman, Larsen, Gibbons, & Engels, 2010) would have improved the construct and should be included in future studies.
Conclusion This study examined the effects of framed messages featuring social and health consequences on college students’ intentions to drink heavily. Social consequences
Social and health consequence framing
219
were more effective when framed as a loss, and health consequences were more effective when framed as a gain. These effects were stronger among those who reported higher levels of previous drinking. Findings suggest that interventions that focus on the negative health effects of heavy drinking may be improved by instead emphasizing negative social consequences of drinking heavily and positive health consequences of avoiding this behaviour. These messages may be more apt to promote behaviour change among young people engaging in high-risk drinking.
Acknowledgements We would like to thank the National Cancer Institute for funding part of the time put into this project (R25 CA057711).
References Borsari, B., Murphy, J. G., & Barnett, N. P. (2007). Predictors of alcohol use during the first year of college: Implications for prevention. Addictive Behaviors, 32, 2062–2086. doi:10.1016/j. addbeh.2007.01.017 Brehm, J. W. (1966). A theory of psychological reactance. New York, NY: Academic Press. Detweiler, J., Bedell, B., Salovey, P., Rothman, A., & Pronin,, (1999). Message framing & sunscreen use: Gain-framed messages motivate beach-goers. Health Psychology, 18 (2), 189–196. doi:10. 1037/0278-6133.18.2.189 Gallagher, K., & Updegraff, J. (2012). Health message framing effects on attitudes, intentions, and behavior: A meta-analytic review. Annals of Behavioral Medicine, 43 (1), 101–116. doi:10. 1007/s12160-011-9308-7 Gerend, M. A., & Cullen, M. (2008). Effects of message framing and temporal context on college student drinking behavior. Journal of Experimental Social Psychology, 44, 1167–1173. doi:10. 1016/j.jesp.2008.02.007 Gerrard, M., Gibbons, F. X., Brody, G. H., Murry, V. M., Cleveland, M. J., & Wills, T. A. (2006). A theory-based dual focus alcohol intervention for pre-adolescents: The Strong African American Families Program. Psychology of Addictive Behavior, 20, 185–195. doi:10.1037/0893-164X.20.2.185 Gibbons, F. X., Gerrard, M., & Lane, D. J. (2003). A social-reaction model of adolescent health risk. In J. J. Suls & K. A. Wallston (Eds.), Social psychological foundations of health and illness (pp. 107–136). Oxford, UK: Blackwell. Gibbons, F. X., Gerrard, M., Lane, D. J., Mahler, H. I. M., & Kulik, J. A. (2005). Using UV photography to reduce use of tanning booths: A test of cognitive mediation. Health Psychology, 24, 358–363. doi:10.1037/0278-6133.24.4.358 Gilles, D., Turk, C., & Fresco, D. (2006). Social anxiety, alcohol expectancies, and self-efficacy as predictors of heavy drinking in college students. Addictive Behaviors, 31, 388–398. doi:10. 1016/j.addbeh.2005.05.020 Hingson, R. W., Zha, W., & Weitzman, E. R. (2009). Magnitude of and trends in alcohol-related mortality and morbidity among U.S. college students ages 18–24, 1998–2005. Journal of Studies on Alcohol and Drugs, 70 (4), S12–20. doi:10.1542/peds.2008-2176 Hornbeck, F. W. (1973). Factorial analyses of variance with appended control groups. Systems Research and Behavioral Science, 18, 213–220. doi:10.1002/bs.3830180310 Kiene, S. M., Barta, W. D., Zelenski, J. M., & Cothran, D. L. (2005). Why are you bringing up condoms now? The effect of message content on framing effects of condom use messages. Health Psychology, 24, 321–326. doi:10.1037/0278-6133.24.3.321 Klein, W. M. P., Lipkus, I. M., Scholl, S. M., McQueen, A., Cerully, J. L., & Harris, P. R. (2010). Self-affirmation moderates effects of unrealistic optimism and pessimism on reactions to tailored risk feedback. Psychology & Health, 25, 1195–1208. doi:10.1080/08870440903261970
220
John H. Kingsbury et al.
Labrie, J. W., Hummer, J. F., & Pedersen, E. R. (2007). Reasons for drinking in the college student context: The differential role and risk of the social motivator. Journal of Studies on Alcohol and Drugs, 68, 393–398. doi:10.1016/j.addbeh.2009.06.001 Mahler, H. I. M., Kulik, J. A., Gibbons, F. X., Gerrard, M., & Harrell, J. (2003). Effects of appearance-based intervention on sun protection intentions and self-reported behaviors. Health Psychology, 22 (2), 199–209. doi:10.1037/0278-6133.22.2.199 Miller, C. H., Burgoon, M., Grandpre, J. R., & Alvaro, E. M. (2006). Identifying principal risk factors for the initiation of adolescent smoking behaviors: The significance of psychological reactance. Health Communication, 19, 241–252. doi:10.1207/s15327027hc1903_6 Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. doi:10.1007/978-1-4612-4964-1_1 Pomery, E. A., Gibbons, F. X., Reis-Bergan, M., & Gerrard, M. (2009). From willingness to intention: Experience moderates the shift from reactive to reasoned behavior. Personality and Social Psychology Bulletin, 35, 894–908. doi:10.1177/0146167209335166 Read, J. P., Wood, M. D., Kahler, C. W., Maddock, J. E., & Palfai, T. P. (2003). Examining the role of drinking motives in college student alcohol use and problems. Psychology of Addictive Behaviors, 17 (1), 13–23. doi:10.1037/0893-164X.17.1.13 Ringold, D. J. (2002). Boomerang effects in response to public health interventions: Some unintended consequences in the alcoholic beverage market. Journal of Consumer Policy, 25 (1), 27–63. doi:10.1023/A:1014588126336 Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121 (1), 3–19. doi:10.1037/0033-2909.121.1.3 SAMHSA (2011). Results from the 2010 National Survey on Drug Use and Health: Summary of national findings. Rockville, MD: Substance Abuse & Mental Health Services Admin. Schneider, T. R., Salovey, P., Pallonen, U., Mundorf, N., Smith, N. F., & Steward, W. T. (2001). Visual and auditory message framing effects on tobacco smoking. Journal of Applied Social Psychology, 31, 667–682. doi:10.1111/j.1559-1816.2001.tb01407.x Spijkerman, R., Larsen, H., Gibbons, F. X., & Engels, R. C. (2010). Students’ drinker prototypes and alcohol use in a naturalistic setting. Alcoholism: Clinical and Experimental Research, 34 (1), 64–71. doi:10.1111/j.1530-0277.2009.01067.x Steward, W. T., Schneider, T. R., Pizarro, J., & Salovey, P. (2006). Need for cognition moderates responses to framed smoking-cessation messages. Journal of Applied Social Psychology, 33, 2439–2464. doi:10.1111/j.1559-1816.2003.tb02775.x Stock, M., Gibbons, F.X., & Gerrard, M. (2012). Psychological reactance and the rejection of persuasive health communications: Is it a heuristic process? Manuscript in preparation. Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211, 453–458. doi:10.1007/978-1-4613-2391-4_2 Werch, C. E., & Owen, D. M. (2002). Iatrogenic effects of alcohol and drug prevention programs. Journal of Studies on Alcohol, 63, 581–590. Received 30 May 2013; revised version received 23 January 2014
Supporting Information The following supporting information may be found in the online edition of the article: Appendix S1. Vignettes (Male Versions).
Copyright of British Journal of Health Psychology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
British Journal of Health Psychology (2015), 20, 212–220 © 2014 The British Psychological Society www.wileyonlinelibrary.com
The effects of social and health consequence framing on heavy drinking intentions among college students John H. Kingsbury1*†, Frederick X. Gibbons1‡ and Meg Gerrard2‡ 1
Department of Psychological and Brain Sciences, Dartmouth College, Hanover, New Hampshire, USA 2 Department of Psychiatry, Dartmouth Medical School, Lebanon, New Hampshire, USA Objectives. Many interventions targeting college student drinking have focused on negative health effects of drinking heavily; however, some research suggests that social factors may have a stronger influence on the drinking behaviour of young people. Moreover, few studies have examined message framing effects in the context of alcohol consumption. This study investigated the effects of social and health consequence framing on college students’ intentions to engage in heavy drinking. Design. This study used a 2 9 2 experimental design with an appended control condition. Methods. One hundred and twenty-four college students (74 women; Mage = 18.9) participated in this study for course credit. Participants read vignettes that were ostensibly written by a recent graduate from the university, who described an episode of drinking in which he or she experienced either social or health consequences. These consequences were framed as either a gain (i.e., positive consequences of not drinking heavily) or a loss (i.e., negative consequences of drinking heavily). After reading the vignette, participants completed a measure of heavy drinking intentions. Results. Regression analyses revealed that social consequences were associated with lower heavy drinking intentions when framed as a loss and that health consequences were associated with lower heavy drinking intentions when framed as a gain. These effects were stronger among those who reported higher (vs. lower) levels of previous drinking. Conclusions. Results suggest that interventions that focus on the negative health effects of heavy drinking may be improved by instead emphasizing the negative social consequences of drinking heavily and the positive health consequences of avoiding this behaviour.
Statement of contribution What is already known on this subject? Previous studies have shown that gain frames are more effective than loss frames when highlighting the health consequences of health risk behaviours, such as heavy drinking. The heavy drinking behaviour of young people is influenced by social factors (e.g., perceived social consequences). However, little is known about framing effects for social consequences of heavy drinking.
†
Minnesota Department of Health, St Paul, Minnesota, USA Department of Psychology, University of Connecticut, Storrs, Connecticut, USA
‡
Correspondence should be addressed to John H. Kingsbury, Minnesota Department of Health, 85 7th Place East, 4th Floor, St Paul, MN 55102, USA (email: [email protected]). DOI:10.1111/bjhp.12100
Social and health consequence framing
213
What does this study add? This study builds on previous research by demonstrating that a loss frame is more effective than a gain frame when highlighting the social consequences of health risk behaviour. Framing effects are strongest for those with more previous drinking experience.
Excessive alcohol consumption is a serious public health problem, especially on college campuses. There are over 1,800 alcohol-related deaths and nearly 600,000 alcohol-related injuries each year among U.S. college students aged 18–24 (Hingson, Zha, & Weitzman, 2009). Intervention efforts to reduce drinking among young people commonly present information about, or images of, the negative health consequences of drinking (e.g., pictures of a drunk driving accident, or an individual who has passed out from heavy drinking). However, these messages are often met with resistance or are rejected (Brehm, 1966; Ringold, 2002). In some cases, particularly among high-risk individuals, these messages have had iatrogenic effects, that is, they have produced increases in risk behaviour (Miller, Burgoon, Grandpre, & Alvaro, 2006; Werch & Owen, 2002).
Framing Health consequences The framing literature suggests that, rather than focusing on the negative health consequences of drinking, focusing on the positive consequences of not drinking may be more effective (Rothman & Salovey, 1997). This approach is consistent with prospect theory (Tversky & Kahneman, 1981), which posits that behaviours with more certain outcomes, such as flossing, are more influenced by potential positive consequences – a gain frame (e.g., reduced risk of gum disease), whereas behaviours with less certain outcomes, such as getting a mammogram, are more influenced by potential negative consequences – a loss frame (e.g., a more severe diagnosis). A recent review found some support for this hypothesis. Gallagher and Updegraff (2012) reported that few studies had examined framing effects for health risk behaviours (k = 4), but those that did found that gain frames were associated with less risk behaviour than loss frames (Gerend & Cullen, 2008; Schneider et al., 2001; Steward, Schneider, Pizarro, & Salovey, 2006). Moreover, effects for gain frame messages on smoking and alcohol use were among the largest observed in this meta-analysis. Thus, while research on framing and risk behaviours is scant, evidence suggests that gain frames may be more effective than loss frames. Questions also remain about what information should be included in messages aimed at reducing risk behaviour. Existing studies examining framing effects for risk behaviours have primarily focused on the health consequences of these behaviours (e.g., gain frame: maintaining a healthy weight, having a healthy liver; loss frame: increased likelihood of risky sexual behaviour, getting in a car accident; Gerend & Cullen, 2008). However, evidence suggests that the decision to drink heavily among young people may be more strongly influenced by social factors (Read, Wood, Kahler, Maddock, & Palfai, 2003). College students report that drinking promotes social facilitation and a sense of belonging with fellow students (Gilles, Turk, & Fresco, 2006; Labrie, Hummer, & Pedersen, 2007). Heavy drinking is particularly common for first-year college students, who are trying to make friends and adjust to a new social environment (Borsari, Murphy, & Barnett, 2007).
214
John H. Kingsbury et al.
In short, heavy drinking among college students is frequently socially motivated. That being the case, the current study tried to use this same (social) motivation to reduce the behaviour.
Social consequences Few studies have investigated framing effects for the social consequences of risk behaviour. One exception is a study by Kiene, Barta, Zelenski, and Cothran (2005), which examined framing effects for a behaviour that has both interpersonal and health implications, condom use (Rothman & Salovey, 1997). They found that persuasive statements about the potential interpersonal consequences of condom use (e.g., a partner’s reaction to suggested condom use) were rated as more convincing and important when framed as a loss than as a gain. Conversely, messages featuring health consequences of condom use were rated as more convincing and important when framed as a gain. The authors explained the pattern of results for social consequences by citing the emotional risk involved in discussing condom use with a partner, that is, the partner may or may not respond negatively. These effects were stronger for those who were higher in issue involvement, suggesting that those who view the consequences as more personally relevant may be more influenced by message characteristics. These results are consistent with research demonstrating that perceived relevance is a key determinant of persuasive message effectiveness (Petty & Cacioppo, 1986). Results of the Kiene et al.’s (2005) study suggest that framing the social consequences of risk behaviour as a loss may be more effective than framing them as a gain. However, this study had participants rating the perceived effectiveness of the messages rather than assessing effects on personal attitudes or intentions to engage in the behaviour, which limits the applicability of these findings to behaviour change. Furthermore, this study utilized a within-subjects design, which may have affected the independence of the ratings for each type of message. The current study aimed to address these limitations and extend the literature, by using an experimental design to investigate framing effects for social and health consequences on college students’ intentions to drink heavily. It was hypothesized that health consequences would be associated with lower intentions when framed as a gain and that social consequences would be associated with lower intentions when framed as a loss. These effects were expected to be stronger among those for whom the behaviour and the consequences would be more relevant (i.e., those with more drinking experience).
Method Participants One hundred and twenty-four undergraduates (74 women; Mage = 18.9) who were enrolled in introductory psychology classes participated in this study for course credit. This study was approved by the IRB at the participating university.
Experimental design and procedure Participants were randomly assigned to one of five conditions in this study, which employed a 2 (consequences: social, health) 9 2 (frame: gain, loss) design with an appended control condition. Participants were told that they would be evaluating vignettes that were written by similar others (i.e., alumni of their university). Vignettes
Social and health consequence framing
215
were used to convey information about drinking consequences because they are less likely to prompt optimistic bias, which can influence the effectiveness of persuasive messages (Klein et al., 2010). Participants in the control group read a vignette that featured a recent alumnus who had difficulty with study habits and schoolwork, but improved his or her school performance after attending a talk by a successful alumnus from the university (see Appendix S1). The control vignette focused on schoolwork troubles because this is a common issue for college students and is unrelated to alcohol. Participants in the experimental conditions read the control vignette and one that focused on alcohol. Alcohol vignettes featured a recent alumnus who went out with friends and experienced social or health consequences of drinking. The consequences were presented using either a gain frame (i.e., positives of not drinking heavily) or loss frame (i.e., negatives of drinking heavily). The social–gain frame vignette featured a student who was able to more easily converse with others, leading to a future date with an attractive member of the opposite sex. The student in the health–gain frame vignette talked about feeling good the day after going out, and the health benefits of moderate drinking (e.g., lower risk of heart disease). Social–loss vignette featured a student who drank so much that he or she vomited in front of an attractive member of the opposite sex. The student in the health–loss vignette blacked out and was taken to the hospital for fear of alcohol poisoning.
Measures Participants rated the vignettes on different dimensions: how typical the student’s experience was, how easy it was to imagine the consequences happening to the self, and how similar the experience was to a previous personal experience (1 = not at all, 7 = very). Two items were used to assess past drinking: ‘How often have you consumed alcohol in the past four weeks?’ (1 = not at all, 6 = more than a few times per week) and ‘How often have you had four or more drinks in a 2-hr span over the past four weeks?’ (1 = 0 times, 6 = more than 4 times). These two items were combined to form a composite scale of past drinking (a = .80). The primary outcome assessed intentions to drink heavily using a conditional item. Previous studies have demonstrated that conditional items are better predictors of risk behaviour than those that are more global (i.e., not situation specific; Gibbons, Gerrard, & Lane, 2003; Pomery, Gibbons, Reis-Bergan, & Gerrard, 2009). Intentions to drink heavily were assessed with a single face-valid item: ‘Imagine you are at a party where alcohol is being served. . .You have had several drinks over the past 2 hr and you have decided to stop drinking. A friend bumps into you, and you begin a conversation. The friend offers to get you another drink’. Participants were asked, ‘In this situation, would you plan to continue drinking?’ 1 = definitely not, 7 = definitely.
Results Descriptive results Participants’ reports of their previous drinking were comparable to national norms (SAMHSA, 2011). They reported that, on average, they had consumed alcohol 2–4 times, and had drunk heavily 1–2 times, in the past month. Twenty-two participants (18%) reported abstaining from drinking in the past 4 weeks. Past drinking did not differ across condition (p = .77).
216
John H. Kingsbury et al.
Ratings of the alcohol consequences were examined to determine whether the vignettes differed on dimensions other than those that were specifically manipulated. Consequences in the gain frame vignettes were rated as easier to imagine (M = 4.94 vs. 2.98) and more similar to personal experiences (M = 4.09 vs. 2.17) than those in the loss frame vignettes (all ps < .001). There were no differences between the social/health vignettes on these dimensions (ps > .29), nor any Frame 9 Consequences interactions. Past drinking was unrelated to ease of imagination and similarity to personal experience and was positively associated with how typical the experience was (p < .01). Frame, consequences, and past drinking Control comparisons Planned contrasts were used to examine differences in intentions to drink heavily among the four experimental conditions and the control group. Results revealed that, as predicted, those in the social–loss and health–gain conditions (Ms = 2.22, 2.53, respectively) reported significantly lower intentions than did those in the control group, M = 3.33; F(1, 118) = 5.61, p = .02. Conversely, intentions to drink heavily in the social– gain and health–loss conditions (Ms = 3.00, 3.05) did not differ from those in the control group (F < 0.49, p = .49).
Moderation by past drinking Linear regression analysis was used to test whether past drinking moderated the effects of frame and consequences on intentions to drink heavily. Dummy codes were used for the consequences and frame conditions, and the control group was assigned a coefficient of ‘0’, so that responses from these participants were factored into the error term, but did not expend a degree of freedom (Hornbeck, 1973). Past drinking was entered as a continuous variable, and two- and three-way interaction terms were created to estimate interactive effects. Results revealed no significant main effects of frame or consequences (ps > .34), but a significant positive effect of past drinking (p < .001), that is, participants with more drinking experience reported higher intentions to drink heavily. Past drinking did not interact with frame or consequences (ps > .43); however, the Frame 9 Consequences interaction was significant, b = .17, t(116) = 2.16, p = .03. Simple effects analyses revealed that there were no framing effects for health consequences, b = .16, t (47) = 1.15, p = .26, but the loss frame led to lower intentions than did the gain frame for social consequences, b = .25, t(47) = 2.19, p = .03. The Frame 9 Consequences interaction was further qualified by a Past Drinking 9 Frame 9 Consequences interaction, b = .24, t(116) = 2.94, p = .004, which followed the predicted pattern (Figure 1). Simple effects analyses indicated that, for social consequences, the loss frame was associated with lower intentions than the gain frame for those higher in past drinking (i.e., +1 SD), b = .47, t(47) = 2.85, p = .007, but not for those lower in past drinking (i.e., 1 SD), b = .02, t(47) = .12, p = .90. Conversely, for health consequences, the gain frame was associated with lower intentions than the loss frame for those higher in past drinking, b = .60, t(47) = 3.11, p = .003, but not for those lower in past drinking, b = .22, t(47) = 1.23, p = .23. In sum, consistent with hypotheses, results indicate that: (1) the social–loss frame was more effective than the social–gain frame, (2) the health– gain frame was more effective than the health–loss frame, and (3) these effects were stronger for those reporting higher levels of past drinking.
Social and health consequence framing High Past Drinking (+1 SD) ** **
Intentions to Drink Heavily
5 4.5
Loss
4
Gain
3.5 3 2.5 2 1.5 1 Social
Health
Low in Past Drinking (-1 SD)
5 Intentions to Drink Heavily
217
4.5
Loss
4
Gain
3.5 3 2.5 2 1.5 1 Social
Health
Figure 1. Interaction between consequences and frame predicting intentions to drink heavily by past drinking experience. Note. **p < .01. High and low in past drinking are defined as plus/minus one standard deviation, respectively.
Discussion Findings from this study are consistent with prospect theory and previous studies demonstrating that gain frames are more effective than loss frames when highlighting the health consequences of risk behaviours (Gerend & Cullen, 2008; Schneider et al., 2001). The finding that social consequences were more effective when framed as a loss is consistent with a study by Kiene et al. (2005) that found that loss-framed messages about interpersonal discussions of condom use were rated as more convincing and important than gain-framed messages. The current study extends these findings by demonstrating framing effects for social consequences on heavy drinking intentions and by showing that the social–loss- and health–gain-framed messages were associated with lower drinking intentions than those reported by a control group.
Previous drinking experience Framing effects were stronger for those who reported higher levels of previous drinking. The perceived relevance of the behaviour in the vignettes may have contributed to relatively strong effects for those reporting more previous drinking. For those who were higher in previous drinking and read about health consequences, the gain frame was associated with much lower intentions to drink than the loss frame. These results lend support to previous findings that loss-framed messages featuring health consequences
218
John H. Kingsbury et al.
may be met with resistance, and thus, are less successful, when presented to high-risk individuals (Detweiler, Bedell, Salovey, Rothman, & Pronin, 1999; Stock, Gibbons, & Gerrard, 2012). Collectively, the vignettes appeared to have weaker effects on those with less drinking experience. It is possible that the consequences in the vignettes, which talked about engaging (or not engaging) in heavy drinking, were viewed as less relevant among this group. Another plausible explanation for these findings is that the drinking intentions of college students who choose to drink less heavily are less malleable than those who drink heavily. More work is needed to test these hypotheses. The advantage of the social–loss frame over the social–gain frame for those higher in previous drinking is consistent with one of the basic tenets of prospect theory: individuals are more influenced by potential positive consequences when considering behaviours with more certain outcomes and more influenced by potential negative consequences when considering behaviours with less certain outcomes. The health consequences of not drinking heavily are relatively certain, that is, not drinking heavily is clearly associated with better health than is drinking heavily. Thus, prospect theory would predict an advantage for gain frames for health consequences (Rothman & Salovey, 1997). Conversely, social consequences of not drinking heavily may be, in some sense, uncertain or ‘risky’ for heavy drinkers. A heavy drinker who chooses not to drink may be unsure of how his or her friends, who may be heavy drinkers themselves, will react. Thus, social–loss-framed messages may effectively take a behaviour that is frequently performed to facilitate social relationships, and make it seem undesirable by highlighting ways in which it can harm social standing. In essence, this approach ‘uses’ the same motivation that promotes the behaviour to inhibit it (Gerrard et al., 2006; Gibbons, Gerrard, Lane, Mahler, & Kulik, 2005; Mahler, Kulik, Gibbons, Gerrard, & Harrell, 2003). Further research is needed to test this hypothesis.
Limitations An important limitation of this study is that it did not assess behaviour change. Previous studies using framing manipulations have detected changes in behaviour at 1-month follow-ups (Gerend & Cullen, 2008), so we expect that the messages used in this study might have some effect on behaviour change, but this assumption should be tested in future studies. This study would have been strengthened by assessing the perceived severity of consequences, as some of the consequences may have been viewed as more extreme than others, which may have influenced intentions to drink. Research examining the content of framed messages has found that the framing component of a message (whether it is gain or loss) is more important than the action or outcome described (Detweiler et al., 1999), so we expect that perceived severity would not influence. More generally, the current study did not identify mediating mechanisms that explain why differently framed messages had an effect on intentions. More work is needed to address these issues. Finally, the drinking intentions construct consisted of only a single item. Although the item is face valid, additional measures of drinking intentions and drinking willingness (the latter being particularly effective for risky behaviours, such as heavy drinking; Gibbons et al., 2003; Spijkerman, Larsen, Gibbons, & Engels, 2010) would have improved the construct and should be included in future studies.
Conclusion This study examined the effects of framed messages featuring social and health consequences on college students’ intentions to drink heavily. Social consequences
Social and health consequence framing
219
were more effective when framed as a loss, and health consequences were more effective when framed as a gain. These effects were stronger among those who reported higher levels of previous drinking. Findings suggest that interventions that focus on the negative health effects of heavy drinking may be improved by instead emphasizing negative social consequences of drinking heavily and positive health consequences of avoiding this behaviour. These messages may be more apt to promote behaviour change among young people engaging in high-risk drinking.
Acknowledgements We would like to thank the National Cancer Institute for funding part of the time put into this project (R25 CA057711).
References Borsari, B., Murphy, J. G., & Barnett, N. P. (2007). Predictors of alcohol use during the first year of college: Implications for prevention. Addictive Behaviors, 32, 2062–2086. doi:10.1016/j. addbeh.2007.01.017 Brehm, J. W. (1966). A theory of psychological reactance. New York, NY: Academic Press. Detweiler, J., Bedell, B., Salovey, P., Rothman, A., & Pronin,, (1999). Message framing & sunscreen use: Gain-framed messages motivate beach-goers. Health Psychology, 18 (2), 189–196. doi:10. 1037/0278-6133.18.2.189 Gallagher, K., & Updegraff, J. (2012). Health message framing effects on attitudes, intentions, and behavior: A meta-analytic review. Annals of Behavioral Medicine, 43 (1), 101–116. doi:10. 1007/s12160-011-9308-7 Gerend, M. A., & Cullen, M. (2008). Effects of message framing and temporal context on college student drinking behavior. Journal of Experimental Social Psychology, 44, 1167–1173. doi:10. 1016/j.jesp.2008.02.007 Gerrard, M., Gibbons, F. X., Brody, G. H., Murry, V. M., Cleveland, M. J., & Wills, T. A. (2006). A theory-based dual focus alcohol intervention for pre-adolescents: The Strong African American Families Program. Psychology of Addictive Behavior, 20, 185–195. doi:10.1037/0893-164X.20.2.185 Gibbons, F. X., Gerrard, M., & Lane, D. J. (2003). A social-reaction model of adolescent health risk. In J. J. Suls & K. A. Wallston (Eds.), Social psychological foundations of health and illness (pp. 107–136). Oxford, UK: Blackwell. Gibbons, F. X., Gerrard, M., Lane, D. J., Mahler, H. I. M., & Kulik, J. A. (2005). Using UV photography to reduce use of tanning booths: A test of cognitive mediation. Health Psychology, 24, 358–363. doi:10.1037/0278-6133.24.4.358 Gilles, D., Turk, C., & Fresco, D. (2006). Social anxiety, alcohol expectancies, and self-efficacy as predictors of heavy drinking in college students. Addictive Behaviors, 31, 388–398. doi:10. 1016/j.addbeh.2005.05.020 Hingson, R. W., Zha, W., & Weitzman, E. R. (2009). Magnitude of and trends in alcohol-related mortality and morbidity among U.S. college students ages 18–24, 1998–2005. Journal of Studies on Alcohol and Drugs, 70 (4), S12–20. doi:10.1542/peds.2008-2176 Hornbeck, F. W. (1973). Factorial analyses of variance with appended control groups. Systems Research and Behavioral Science, 18, 213–220. doi:10.1002/bs.3830180310 Kiene, S. M., Barta, W. D., Zelenski, J. M., & Cothran, D. L. (2005). Why are you bringing up condoms now? The effect of message content on framing effects of condom use messages. Health Psychology, 24, 321–326. doi:10.1037/0278-6133.24.3.321 Klein, W. M. P., Lipkus, I. M., Scholl, S. M., McQueen, A., Cerully, J. L., & Harris, P. R. (2010). Self-affirmation moderates effects of unrealistic optimism and pessimism on reactions to tailored risk feedback. Psychology & Health, 25, 1195–1208. doi:10.1080/08870440903261970
220
John H. Kingsbury et al.
Labrie, J. W., Hummer, J. F., & Pedersen, E. R. (2007). Reasons for drinking in the college student context: The differential role and risk of the social motivator. Journal of Studies on Alcohol and Drugs, 68, 393–398. doi:10.1016/j.addbeh.2009.06.001 Mahler, H. I. M., Kulik, J. A., Gibbons, F. X., Gerrard, M., & Harrell, J. (2003). Effects of appearance-based intervention on sun protection intentions and self-reported behaviors. Health Psychology, 22 (2), 199–209. doi:10.1037/0278-6133.22.2.199 Miller, C. H., Burgoon, M., Grandpre, J. R., & Alvaro, E. M. (2006). Identifying principal risk factors for the initiation of adolescent smoking behaviors: The significance of psychological reactance. Health Communication, 19, 241–252. doi:10.1207/s15327027hc1903_6 Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. doi:10.1007/978-1-4612-4964-1_1 Pomery, E. A., Gibbons, F. X., Reis-Bergan, M., & Gerrard, M. (2009). From willingness to intention: Experience moderates the shift from reactive to reasoned behavior. Personality and Social Psychology Bulletin, 35, 894–908. doi:10.1177/0146167209335166 Read, J. P., Wood, M. D., Kahler, C. W., Maddock, J. E., & Palfai, T. P. (2003). Examining the role of drinking motives in college student alcohol use and problems. Psychology of Addictive Behaviors, 17 (1), 13–23. doi:10.1037/0893-164X.17.1.13 Ringold, D. J. (2002). Boomerang effects in response to public health interventions: Some unintended consequences in the alcoholic beverage market. Journal of Consumer Policy, 25 (1), 27–63. doi:10.1023/A:1014588126336 Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121 (1), 3–19. doi:10.1037/0033-2909.121.1.3 SAMHSA (2011). Results from the 2010 National Survey on Drug Use and Health: Summary of national findings. Rockville, MD: Substance Abuse & Mental Health Services Admin. Schneider, T. R., Salovey, P., Pallonen, U., Mundorf, N., Smith, N. F., & Steward, W. T. (2001). Visual and auditory message framing effects on tobacco smoking. Journal of Applied Social Psychology, 31, 667–682. doi:10.1111/j.1559-1816.2001.tb01407.x Spijkerman, R., Larsen, H., Gibbons, F. X., & Engels, R. C. (2010). Students’ drinker prototypes and alcohol use in a naturalistic setting. Alcoholism: Clinical and Experimental Research, 34 (1), 64–71. doi:10.1111/j.1530-0277.2009.01067.x Steward, W. T., Schneider, T. R., Pizarro, J., & Salovey, P. (2006). Need for cognition moderates responses to framed smoking-cessation messages. Journal of Applied Social Psychology, 33, 2439–2464. doi:10.1111/j.1559-1816.2003.tb02775.x Stock, M., Gibbons, F.X., & Gerrard, M. (2012). Psychological reactance and the rejection of persuasive health communications: Is it a heuristic process? Manuscript in preparation. Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211, 453–458. doi:10.1007/978-1-4613-2391-4_2 Werch, C. E., & Owen, D. M. (2002). Iatrogenic effects of alcohol and drug prevention programs. Journal of Studies on Alcohol, 63, 581–590. Received 30 May 2013; revised version received 23 January 2014
Supporting Information The following supporting information may be found in the online edition of the article: Appendix S1. Vignettes (Male Versions).
Copyright of British Journal of Health Psychology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
