Why Control Tonight Will Fail to Control Excessive Drinking: A Critique Using Similarity, Optimistic Bias, and Reactance Theories – Richard Ahl
Excessive alcohol consumption is a major public health problem in the United States. In addition to causing short-term health problems (including vomiting and memory loss), heavy drinking can lead to instant death, such as through alcohol poisoning or car accidents (1). It increases the risk of long-term illness, including cancer, liver cirrhosis, stroke, addiction and heart failure (2). In the year 2001 alone, it was responsible for an estimated 75,000 American deaths (3). This number makes it a leading cause of disability and death (2).
Excessive drinking is defined as “per-occasion consumption or average consumption of alcohol that puts individuals at increased risk for alcohol-related health and social problems” (2). It is estimated that 30% of Americans who consume alcohol (or 17% of the entire population) are excessive drinkers (2). About 10 to 15% of excessive drinkers are dependent on, or addicted to, alcohol (2). Excessive drinking can take the form of binge drinking, heavy drinking, or both. A binge drinker is someone who drinks large quantities of alcohol (five or more drinks) on a single occasion at least once per month (4). Binge drinkers are 14 times more likely to drive while intoxicated than non-binge drinkers (4). A heavy drinker is defined as a woman who has more than 30 drinks per month, or a man who has more than 60 (2).
Alcohol consumption often begins at a young age. The minimum legal drinking age of 21 has failed to curb underage drinking; in fact, some have argued that it may increase dangerous drinking practices (5-7). An estimated 62% of twelfth graders currently use alcohol, and the percentage of college students under the age of 21 that regularly consume alcohol is even higher (1). Over 30% of high school students binge drink on a monthly basis (1). Amongst college students, this figure rises to nearly 50% (8). Of all adults aged 18 and older, binge drinking is most common amongst those between 18 and 25 years old. In this age group, about 61% of men and 34% of women are binge drinkers (4). Because binge drinking is especially common amongst young adults, and because young adults may be particularly sensitive to the biological effects of alcohol (including addiction), it is crucial for alcohol-related public health interventions to target this age group. Control Tonight is one recent intervention that attempts to discourage binge drinking in this population.
Control Tonight: The intervention
Control Tonight is a web-based public health intervention sponsored by the Pennsylvania Liquor Control Board (9). It cautions its viewers about the consequences of binge drinking. Overall, the focus is on harm reduction: the intervention does not advocate total abstinence from drinking, but rather keeping alcohol consumption to a reasonable level. (Some of the text states, “it’s not unusual to like to unwind and have a little fun with friends after a stressful week.”) The minimum legal drinking age (MLDA) of 21 is not mentioned in the intervention materials. The materials strongly imply that the characters in the intervention are above the age of 21. (Both storylines in the intervention involve interactions with either police officers or paramedics, but at no point do they state that the characters are in legal trouble for being underage drinkers.) Thus, it seems that the intervention is targeted to those of legal drinking age.
One of the online ads referring the viewer to the main website shows the picture of a woman’s bare legs, with her underwear around her ankles, in front of a tile floor background. The bold, all-caps text on the right side of the picture reads as follows: “2:19 AM. She didn’t want to do it, but she couldn’t say no. When your friends drink, they end up making bad decisions, like going home with someone they don’t know very well. Decisions like that leave them vulnerable to dangers like date rape. Help your friends stay in control and stay safe.” Unsurprisingly, women’s and rape survivor’s advocacy groups have criticized this offensive advertisement because it suggests that victims, rather than perpetrators, are ultimately responsible for date rape (10).
The main webpage invites the viewer to state his/her name and sex, along with the name and sex of two friends. Then, it takes the viewer through one of two storylines. In the first story, targeted to men, “your team’s in the big game tonight. So you and your friends are hitting up your favorite sports bar.” After getting drunk on beer and shots, your team loses. Then, your male friend starts a fight with a fan of the opposing team and winds up getting arrested for assault. In the second story, you are throwing a birthday party for a female friend. The party guests encourage her to drink excessively. One friend decides to purchase more beer. “He just did a keg stand five minutes ago, but he drives after drinking all the time. He’ll be okay, right?” Then, your female friend starts vomiting uncontrollably. An ambulance is called, and “all you can do is stand there and watch as the paramedics wipe puke off her and try to insert the IV. What a birthday present.” The last part of the website, the Field Guide, contains tips for binge drinking prevention. It provides guidelines for the amount of alcohol in specific mixed drinks, recommends eating before drinking, and provides tips for discussing unhealthy drinking habits with friends.
Control Tonight does have some strengths. The language used is informal, simple, and fairly authentic to how people actually speak. The overall focus on harm reduction is a good one: It is easier to convince people to restrain their drinking than to stop it completely. This intervention’s arguments can easily apply to those under 21, as well as those over 21. In contrast, an intervention focused solely on those below 21 could cause older adults to ignore the message entirely. Many tips in the Field Guide are informative, practical, and free of sensationalism. The section on reaching out to binge-drinking friends states, “scolding your friend or sounding preachy may only increase their resistance to hearing the truth and getting help.” Unfortunately, the creators of this intervention did not take their own advice. The next section of this paper will use insights from behavioral science to criticize the weaknesses of this intervention.
Failure to create a strong sense of interpersonal similarity
According to Silvia, “interpersonal similarity can reduce reactance by increasing compliance and by reducing resistance” (11). His experiment tested the effects of similarity on participants’ reception of an opinion-based essay. In the high-similarity condition, participants received an essay written by someone who, they were told, shared their first name, birthday, and college class year. In other conditions, participants received identical essays, but were either told that they did not have these features in common or were not told anything about the author. Participants in the high-similarity condition were significantly more likely to agree with writer’s opinions than those in the low-similarity or control conditions. Comparable results were found when the manipulation focused on personal values. Participants were significantly more likely to agree with the writer’s opinions when told that they shared many important life values (such as a belief in equality) than when told they had different life values. Again, the actual essays were identical in both conditions.
These findings have significant implications for the field of public health. They suggest that individuals are more likely to heed public health messages when they believe they share characteristics with the subjects/characters in the messages. For instance, a male college student might be more receptive to an anti-smoking ad featuring a male college student than one featuring an elderly woman. Individuals are also more likely to heed them when they perceive affinity with the author of the messages (i.e. does the targeted individual believe that the person sponsoring the messages is similar to them?). Andsager and colleagues found that college students were significantly more receptive to a written public health message about tanning when they perceived themselves to be similar to the message’s main character (12). Depending on the experimental conditions, the main character, Chris, either mentions drinking alcohol or does not mention drinking alcohol, and either does or does not mention spending time with friends. There were four conditions total (alcohol/friends, no alcohol/friends, alcohol/no friends, and no alcohol/no friends), and participants received different messages depending on the condition. Participants deemed themselves to be most similar to Chris, and most receptive to the message, in the alcohol/friends condition. The reason is that, in this condition, Chris most resembled the participants, most of whom drank alcohol and enjoyed socializing with friends.
On a surface level, it may seem as though Control Tonight harnesses perceived similarity to increase its appeal. The name and sex of the viewer and his/her friends are incorporated into the storylines. This tactic encourages the viewers to imagine themselves in the situations described in the storylines. However, the actual situations described in the plotlines are significant barriers to inducing perceived similarity. The situations are both very specific and somewhat unlikely, and these factors make it difficult for the viewers to feel like the scenarios could happen to them. The sports bar situation is unlikely to resonate with those who are not avid sports fans. Even amongst those who are sports fans, most have not had the experience of getting arrested for physically assaulting an opposing fan. The party situation is more general than the sports bar situation; lots of people have planned a party for friends. However, how many people have driven a car after doing a keg stand, and how many people have been hospitalized for alcohol poisoning? The viewers will feel (or at least want to feel) unlike the characters in the stories, due to the extreme nature of their mistakes. Consequently, they may ignore the anti-binge drinking message.
The sensationalism of the stories undermines the credibility of the authors, causing the viewers to feel unlike the people who wrote Control Tonight. The insensitive treatment of date rape in the advertisement makes the creators seem immoral: they will use any method at their disposal to make drinking seem dangerous. Few people will feel affinity with, or want to listen to, someone who implies that women who drink alcohol are “getting what they deserve” if they become victims of date rape (10).
Failure to address implications of optimistic bias theory
Human beings, for better or for worse, have a tendency to underestimate the likelihood that bad things will happen to them. They also overestimate the likelihood that good things will happen to them. This concept is known as unrealistic optimism, or the optimistic bias (13). In a pioneering study on this concept, Weinstein surveyed college seniors and asked them to rate their chances of experiencing a variety of negative and positive life events (13). They were specifically instructed to rate their chances relative to their college classmates, rather than the population at large. In general, the participants tended to rate themselves as less likely than their peers to experience negative life events (i.e. “attempting suicide”), and more likely to experience positive events (i.e. “traveling to Europe”). The magnitude of this effect was particularly strong for events that were viewed as being controllable. For negative events that were deemed controllable, participants seemed to form a stereotype of people who experience the hardship in question. Then, they compared themselves to this stereotype and deemed themselves different from it, and thus less likely to experience the negative event (13).
“Having a drinking problem” was one of the 24 negative events for which participants showed a particularly significant optimistic bias. Participants felt themselves to be especially unlikely to develop a drinking problem. Of course, it is mathematically impossible for most participants to be less likely to develop a drinking problem than the other participants in the study. It seems that the participants in this study perceived themselves to be largely immune from problems with alcohol. The participants may have created a stereotype of a problem drinker and than compared themselves favorably to that stereotype. As the reasoning goes, alcoholism is something that happens to other people, who develop it because of character weaknesses and/or poor life choices. (The truth is that it can happen for a variety of reasons that one cannot control, such as genetic predisposition.) The high undesirability of alcoholism is another factor that created an unrealistically low expectation of personal risk.
College-age individuals as a population are particularly likely to have an optimistic bias regarding risky activities (14). Young adults are more likely than their parents to rate risky activities (such as drug and alcohol usage) as less likely to harm them than their parents (14). This is particularly true for experimental, rather than frequent, involvement in risky behaviors (14). They are less likely to believe that engaging in threatening activities will result in personal harm.
Even when individuals acknowledge a given behavior to be harmful in general, it does not necessarily mean that they perceive the behavior to be harmful to themselves. Part of the optimistic bias is the belief that one is personally immune to the effects of behaviors that are harmful to others. Optimistic bias may be both an effect and a cause of unhealthy drinking in young adults. Wild and colleagues studied risk perception of heavy alcohol usage amongst college students (15). The participants were asked, “to what extent do you believe that you would be personally at risk of getting hurt or sick because of your drinking?” and, “to what extent do you believe that some other person your age who drinks the way you do would be at risk of getting hurt or sick?” Overall, the participants rated the likelihood of harm to be significantly higher for the second question than for the first. They believed themselves to be less likely to experience drinking-related harm than a hypothetical peer with identical drinking habits. This effect was especially strong for the heaviest, riskiest drinkers. Another study found that college students who believed themselves to be at a lower risk of having alcohol problems than their peers, when controlling for their actual drinking habits, were significantly more likely to experience drinking-related harm in the future (16). Optimistic bias contributes to unhealthy drinking habits.
Control Tonight seems to be based on the idea that exposing young adults to worst-case scenarios of binge drinking will cause them to feel personally at risk of experiencing similar worst-case scenarios. However, research on optimistic bias and alcohol usage challenges this idea. Believing that binge drinking can be harmful in general is not the same as believing that binge drinking will result in harm to oneself. The gap between these two beliefs is especially wide for the groups the intervention is trying to reach: young adults and those with risky drinking behaviors (14-16). Control Tonight may actually exacerbate feelings of optimistic bias in the target audience. Many young adults binge drink (1,4,8), but few experience the extremely negative events depicted in the storylines. They are likely to feel that their personal experiences are more accurate indications of binge drinking’s effects than the over-the-top claims of a public health message. Someone who drinks heavily but has never assaulted a stranger or been rushed to the ER due to drinking may believe that s/he is smarter, more careful, or more experienced than the Control Tonight characters, and thus the situations described in the intervention do not apply to them. Control Tonight could serve to affirm an individual’s flawed belief in personal exemption from drinking-related harm: “horrible things may happen to others when they drink, but they don’t happen to me.” It is particularly easy to feel optimistic bias and distance oneself from a group of harm-prone people if that group can be easily stereotyped. Some of the Control Tonight characters can be stereotyped as showing blatantly poor judgment. This could cause the viewer to feel superior to them, and thus exempt from the problems that befall them.
Creation of a reactance motivational state
No one likes to be told what they can or cannot do. This concept forms the basis of reactance theory (11). Reactance theory states that people deeply value their sense of autonomy and personal freedom. If people believe their freedom has been unfairly curtailed, they may respond by rebelling (11). When people are prohibited from doing something, they often react by refusing to comply with the prohibition. This “reactance motivational state” is particularly strong amongst young adults, who tend to place a high premium on individuality and rebellion (6). Some scholars refer to reactance theory as the “forbidden fruit” argument (6).
Studies on youth drinking have addressed the role of reactance theory in alcohol consumption. Two studies found a strong increase in drinking amongst those under the age of 21 when 21 became the minimum legal drinking age. College students under the age of 21 became significantly more likely to drink after it became illegal for those in their age group to purchase alcohol (6). Similarly, college students under the age of 21 were found to be more likely to drink more than those over the age of 21, although similar illegal drug consumption levels were found in both age groups (7). Presumably, the age-based prohibition made the younger students desire alcohol more, whereas laws against drugs affected both age groups equally (7). This pattern of behavior is consistent with reactance theory: if someone tells you that you shouldn’t do something, you may want to do it even more. While Control Tonight does not target underage drinking specifically, these studies are important because they show how people respond to others’ attempts to limit their drinking.
Bensley and Wu found that college students perceived dogmatic and pro-abstinence alcohol-prevention messages to be less effective than messages that were less negative (17). When exposed to a harsh, high-threat message, participants showed greater intentions to drink alcohol than those exposed to a low-threat message. Similarly, participants drank significantly more alcohol during a sham alcohol taste test after being exposed to the high-threat message. The authors concluded, “dogmatic alcohol prevention materials may have counterproductive effects for some college students” (17). Similar findings were found in a study conducted on Canadian college students: Negative anti-drinking ads inducing feelings of guilt, shame, and fear seemed to cause more problem drinking (18).
Control Tonight certainly falls under the category of a negative message. It focuses on worst-case examples of binge drinking and suggests that similar results are likely to befall the viewers of the messages. It describes consequences in extreme terms, and it features examples of selfish, dangerous, and reckless behavior. Those who do drink excessively are unlikely to agree with Control Tonight. Instead, they are likely to respond by feeling angry, upset, and personally attacked. Because binge drinking is a common part of American youth culture, many young adults view it as an essential part of themselves. One possible consequence of viewing this campaign could be binge drinking to “prove them wrong.” Viewers might be inclined to binge drink and demonstrate that their drinking is different, regardless of whether it actually is, and to affirm their freedom to make decisions. They may want to prove (if only to themselves) that they can handle their alcohol, have a good time, and not lose control. Control Tonight is essentially a challenge to the viewer: if you binge drink, this is what will happen to you. Viewers will be eager to validate their life choices, rebel against the PA Liquor Board, and show that they are different. Control Tonight’s tone is harsh enough to create reactance, and reactance is likely to exacerbate unhealthy drinking.
A better approach to binge drinking prevention: The intervention
What can be learned from Control Tonight’s mistakes? To create a public health campaign that will curb binge drinking amongst young adults, the following conditions should be met. The viewers should feel affinity with the characters in the campaign, and they should not be made to doubt the integrity of its creators. The campaign should recognize the implications of optimistic bias. It should emphasize personal, real-life stories, and it should use optimistic bias advantageously, by making the viewers feel empowered to perform positive behaviors. Finally, it should avoid a harsh, condemnatory tone so that the viewers do not feel motivated to rebel against the campaign.
The “Good Friends, Good Nights” campaign would share some similarities with Control Tonight. It would have a harm reduction approach and focus on limiting binge drinking, rather than stopping drinking altogether. It would be geared towards young adults, both under and over the age of 21. This would allow it to reach a wide range of viewers and not waste resources on an intervention that only appeals to a restricted age group. It would use clear, simple language, without excessive slang. It would include practical tips on healthy drinking, such as those found in Control Tonight’s Field Guide. However, the content and overall tone of the campaign would be quite different.
“Good Friends, Good Nights” is the title of my proposed anti-binge drinking campaign. Part 1 of the campaign would consist of print ads, billboards, and online ads, all of which would refer viewers to a website. The website, part 2, would constitute the core of the campaign. Each of the part 1 ads would display one of several pictures and a text section. The pictures would show groups of smiling people, either talking in a restaurant/bar, outside a sports stadium, at a concert, at a party, or dancing, depending on the picture. The ads would show a mix of ages, races, and both men and women. Some pictures would seem to depict college students (i.e. an outdoor party featuring people wearing college sweatshirts) , and others would seem to depict post-collegiate adults (i.e. professionally-dressed people at a bar). Some, but not all of the people, would be shown with glasses or cups in their hands. Ideally, most viewers would be able to identify with at least one person depicted in one of the pictures. The text would contain messages such as, “Be a good friend. Relax, have fun, and help your friends stay safe,” “help your friends have a good night, and a good morning,” and “make your nights fun, long, and memorable,” and then state, “visit our website to find out how.” These messages would have connotations with alcohol usage but would not explicitly reference alcohol.
The main component of part 2, the website, would consist of personal, real-life stories from actual young adults, along with their first names, ages, job/student status, home states, and a statement that they are the real first names of actual people. The stories would depict a mixture of experiences. Employees of “Good Friends, Good Nights” would visit colleges, bars, and workplaces. They would find young adults and ask to survey them in exchange for modest financial compensation. The open-ended surveys would include prompts for personal stories about experiences with alcohol, as well as alcohol-free ones. Stories would be chosen and edited to represent the following categories of experiences, to create twelve stories total: stories of fun nights with friends in which no drinking was involved (including stories from non-drinkers), stories of fun nights in which moderate drinking with friends was involved, and stories in which people either stopped their friends from drinking excessively or helped their friends after their friends drank excessively (either by making sure they got home safely, taking them to the emergency room, or helping them seek treatment for alcoholism). The stories will emphasize the value of good friendship and the ways in which friendship enriches our lives. The inclusion of the no-drinking stories is important to indicate that drinking is not necessary in order to have a good time. The website would have a Field Guide section with tips on healthy, safe drinking, much like the guide in Control Tonight. It would also contain resources for helping to end problem drinking habits. Visitors to the site could enter in their zip code and receive information on local substance abuse counselors and centers, to be used either by their friends or themselves.
Why “Good Friends, Good Nights” will work: insights from social science
This campaign would make the viewer feel similar to the people depicted in the materials. This similarity would be created on the levels of external characteristics as well as deep core values. In doing so, it would increase the viewer’s receptiveness to the campaign’s message (11,12). The pictures and stories will depict young adults with a variety of ages, ethnic backgrounds, home states, and occupations. It is likely that most young adult viewers will feel a sense of affinity with them due to a shared characteristic, and because they are real people (11,12). They will also want to feel like them, as the campaign’s characters are being depicted in a positive light.
On a deeper level, viewers will feel like the characters because of the values they seem to espouse. Having shared life values can be a strong basis for a sense of interpersonal similarity (11). The main value emphasized by this campaign is friendship: being a good friend and helping others. Most people like to think of themselves in positive terms, and thus most people like to think of themselves as being good to their friends (13). This means that the viewer will feel like the characters in the campaign, who are depicted as enjoying time spent with friends and demonstrating the values of caring and compassion. The personal stories on the website will strike a balance between specific and general: specific enough to be vivid, but not so specific that they are narrow in scope or are only relatable to a select few. The fact that the stories are based on real people, rather than fictional characters, would also help create a sense of similarity. Finally, the viewers would be unlikely to question their affinity with the campaign’s creators, as the campaign appeals to them on positive terms and does not use sensationalized fictional stories. By creating a sense of similarity, the campaign will increase the likelihood that the viewers will heed its message (11,12).
Optimistic bias is a vexing problem for many public health practitioners. Even when people recognize that a given behavior is dangerous in general, they won’t necessarily feel that the behavior is dangerous to themselves personally (13-16). However, optimistic bias cuts both ways: It also means that people have unrealistically positive expectations of their character strengths and their ability to perform positive behaviors (13). “Good Friends, Good Nights” has an optimistic tone. Its message will encourage the viewers to “act like good friends” and help give them the confidence to do so. Because the viewers will want to think of themselves positively, they will want to emulate the responsible behaviors modeled by the characters in the website. The intervention is focused on how one acts towards others, but it should also serve to reduce binge drinking in the viewers as well. Ideally, the campaign would help set a new norm of positive, responsible drinking. The goal is to present many good examples of drinking behaviors. The viewers will want to feel on par with, or superior to, these examples, and will begin to modify their behaviors accordingly.
“Good Friends, Good Nights” has a light-handed approach. Rather than challenging the viewers and telling them that their lifestyle is wrong, it gives examples of positive behaviors and then lets the viewers connect the dots on their own. The campaign essentially identifies a solution without excessively pointing fingers at the problem. Thus, it avoids the main pitfalls of critical anti-binge drinking interventions, which often cause the viewers to react angrily and then ignore their advice (16-18).
This campaign avoids creating a reactance state due to its positive tone, its harm reduction approach, its empowering message, its appeal to the value of friendship, its use of real-life first-person narratives, and its lack of direct criticism. A certain amount of negativity is necessary when discussing the binge drinking of the characters’ friends. However, the reactance caused by this negativity will be minimized because the viewers are being addressed as the friends of binge drinkers, rather than the binge drinkers themselves. The characters in the website’s stories are describing how they behaved in positive drinking situations and how they helped friends who were binge drinking, not how they changed their own drinking habits. In other words, the viewers are being treated as responsible drinkers who should help keep their friends safe, rather than irresponsible drinkers who need to change their own habits and who require help from others. The intervention has two goals: to help the viewers encourage their friends to drink responsibly, and to help the viewers drink responsibly themselves. The second goal of this intervention, however, is somewhat hidden. Asking viewers to change their drinking directly might be interpreted as a challenge to their freedom. However, asking viewers to change their friends’ drinking will allow them to internalize the message of responsible drinking without causing them to feel personally attacked. The use of real-life stories will reduce reactance because the viewers will be less likely to feel that an authority figure is trying to change them. Instead, the messages seem to be coming from their peers. The approach described in this paper would not stop binge drinking altogether, but it should prove more effective at reducing its prevalence than the negative campaigns that have targeted young adults in recent years.
References
1.The National Center on Addiction and Substance Abuse at Columbia University. Teen Tipplers: America’s Underage Drinking Epidemic. New York, NY: The National Center on Addiction and Substance Abuse, 2003.
2. Town M, Naimi TS, Mokdad AH, Brewer RD. Heath care access among U.S. adults who drink alcohol excessively: missed opportunities for prevention. Preventing Chronic Disease 2006; 3: 1-8.
3. Centers for Disease Control and Prevention. Alcohol-attributable deaths and years of potential life lost -- United States, 2001. Morbidity and Mortality Weekly Report 2004; 53: 866-870.
4. Naimi TS, Brewer RD, Mokdad A, Denny C, Serdula MK, Marks JS. Binge drinking among US adults. Journal of the American Medical Association 2003; 289: 70-75.
5. Lanza-Kaduce L, Richards P. Raising the minimum drinking age: some unintended consequences of good intentions. Justice Quarterly 1989; 6: 247-262.
6. Engs RC, Hanson DJ. Reactance theory: a test with collegiate drinking. Psychological Reports 1989; 64: 1083-1086.
7. Allen DN, Sprenkel DG, Vitale PA. Reactance theory and alcohol consumption laws: Further confirmation among collegiate alcohol consumers. Journal of Studies on Alcohol 1994; 55: 34-40.
8. U.S. Department of Health and Human Services. Binge Drinking in Adolescents and College Students. Washington, D.C: U.S. Department of Health and Human Services, 2000.
9. Pennsylvania Liquor Control Board. Control Tonight. PA: Pennsylvania Liquor Control Board, 2011. www.controltonight.com .
10. The Huffington Post. PA Liquor Board Ad Accused of Blaming Rape on Victims. The Huffington Post, 2011. http://www.huffingtonpost.com/2011/12/08/liquor-control-board-ad_n_1137083.html
11. Silvia PJ. Deflecting reactance: the role of similarity in increasing compliance and reducing resistance. Basic and Applied Social Psychology 2005; 27: 277-284.
12. Andsager JL, Bemker V, Choi H, Torwel V. Perceived similarity of exemplar traits and behavior: effects on message evaluation. Communication Research 2006; 33: 3-18.
13. Weinstein ND. Unrealistic optimism about future life events. Journal of Personality and Social Psychology 1980; 39: 806-820.
14. Cohn LD, Macfarlane S, Yanez C, Imai WK. Risk-perception: differences between adolescents and adults. Health Psychology 1995; 14: 217-222.
15. Wild TC, Hinson R, Cunningham J, Bacchiochi J. Perceived vulnerability to alcohol-related harm in young adults: independent effects of risky alcohol use and drinking motives. Experimental and Clinical Psychopharmacology 2001; 9: 117-125.
16. Dillard AJ, Midboe AM, Klein WMP. The dark side of optimism: unrealistic optimism about problems with alcohol predicts subsequent negative event experiences. Personality and Social Psychology Bulletin 2009; 35: 1540-1550.
17. Bensley LS, Wu R. The role of psychological reactance in drinking following alcohol prevention messages. Journal of Applied Social Psychology 2006; 21: 1111-1124.
18. Mullman J. Why binge drinking PSAs may leave some reaching for another drink. Advertising Age, 2010. http://adage.com/article/news/study-binge-drinking-psas-exacerbate-problem/142459/
Excessive drinking is defined as “per-occasion consumption or average consumption of alcohol that puts individuals at increased risk for alcohol-related health and social problems” (2). It is estimated that 30% of Americans who consume alcohol (or 17% of the entire population) are excessive drinkers (2). About 10 to 15% of excessive drinkers are dependent on, or addicted to, alcohol (2). Excessive drinking can take the form of binge drinking, heavy drinking, or both. A binge drinker is someone who drinks large quantities of alcohol (five or more drinks) on a single occasion at least once per month (4). Binge drinkers are 14 times more likely to drive while intoxicated than non-binge drinkers (4). A heavy drinker is defined as a woman who has more than 30 drinks per month, or a man who has more than 60 (2).
Alcohol consumption often begins at a young age. The minimum legal drinking age of 21 has failed to curb underage drinking; in fact, some have argued that it may increase dangerous drinking practices (5-7). An estimated 62% of twelfth graders currently use alcohol, and the percentage of college students under the age of 21 that regularly consume alcohol is even higher (1). Over 30% of high school students binge drink on a monthly basis (1). Amongst college students, this figure rises to nearly 50% (8). Of all adults aged 18 and older, binge drinking is most common amongst those between 18 and 25 years old. In this age group, about 61% of men and 34% of women are binge drinkers (4). Because binge drinking is especially common amongst young adults, and because young adults may be particularly sensitive to the biological effects of alcohol (including addiction), it is crucial for alcohol-related public health interventions to target this age group. Control Tonight is one recent intervention that attempts to discourage binge drinking in this population.
Control Tonight: The intervention
Control Tonight is a web-based public health intervention sponsored by the Pennsylvania Liquor Control Board (9). It cautions its viewers about the consequences of binge drinking. Overall, the focus is on harm reduction: the intervention does not advocate total abstinence from drinking, but rather keeping alcohol consumption to a reasonable level. (Some of the text states, “it’s not unusual to like to unwind and have a little fun with friends after a stressful week.”) The minimum legal drinking age (MLDA) of 21 is not mentioned in the intervention materials. The materials strongly imply that the characters in the intervention are above the age of 21. (Both storylines in the intervention involve interactions with either police officers or paramedics, but at no point do they state that the characters are in legal trouble for being underage drinkers.) Thus, it seems that the intervention is targeted to those of legal drinking age.
One of the online ads referring the viewer to the main website shows the picture of a woman’s bare legs, with her underwear around her ankles, in front of a tile floor background. The bold, all-caps text on the right side of the picture reads as follows: “2:19 AM. She didn’t want to do it, but she couldn’t say no. When your friends drink, they end up making bad decisions, like going home with someone they don’t know very well. Decisions like that leave them vulnerable to dangers like date rape. Help your friends stay in control and stay safe.” Unsurprisingly, women’s and rape survivor’s advocacy groups have criticized this offensive advertisement because it suggests that victims, rather than perpetrators, are ultimately responsible for date rape (10).
The main webpage invites the viewer to state his/her name and sex, along with the name and sex of two friends. Then, it takes the viewer through one of two storylines. In the first story, targeted to men, “your team’s in the big game tonight. So you and your friends are hitting up your favorite sports bar.” After getting drunk on beer and shots, your team loses. Then, your male friend starts a fight with a fan of the opposing team and winds up getting arrested for assault. In the second story, you are throwing a birthday party for a female friend. The party guests encourage her to drink excessively. One friend decides to purchase more beer. “He just did a keg stand five minutes ago, but he drives after drinking all the time. He’ll be okay, right?” Then, your female friend starts vomiting uncontrollably. An ambulance is called, and “all you can do is stand there and watch as the paramedics wipe puke off her and try to insert the IV. What a birthday present.” The last part of the website, the Field Guide, contains tips for binge drinking prevention. It provides guidelines for the amount of alcohol in specific mixed drinks, recommends eating before drinking, and provides tips for discussing unhealthy drinking habits with friends.
Control Tonight does have some strengths. The language used is informal, simple, and fairly authentic to how people actually speak. The overall focus on harm reduction is a good one: It is easier to convince people to restrain their drinking than to stop it completely. This intervention’s arguments can easily apply to those under 21, as well as those over 21. In contrast, an intervention focused solely on those below 21 could cause older adults to ignore the message entirely. Many tips in the Field Guide are informative, practical, and free of sensationalism. The section on reaching out to binge-drinking friends states, “scolding your friend or sounding preachy may only increase their resistance to hearing the truth and getting help.” Unfortunately, the creators of this intervention did not take their own advice. The next section of this paper will use insights from behavioral science to criticize the weaknesses of this intervention.
Failure to create a strong sense of interpersonal similarity
According to Silvia, “interpersonal similarity can reduce reactance by increasing compliance and by reducing resistance” (11). His experiment tested the effects of similarity on participants’ reception of an opinion-based essay. In the high-similarity condition, participants received an essay written by someone who, they were told, shared their first name, birthday, and college class year. In other conditions, participants received identical essays, but were either told that they did not have these features in common or were not told anything about the author. Participants in the high-similarity condition were significantly more likely to agree with writer’s opinions than those in the low-similarity or control conditions. Comparable results were found when the manipulation focused on personal values. Participants were significantly more likely to agree with the writer’s opinions when told that they shared many important life values (such as a belief in equality) than when told they had different life values. Again, the actual essays were identical in both conditions.
These findings have significant implications for the field of public health. They suggest that individuals are more likely to heed public health messages when they believe they share characteristics with the subjects/characters in the messages. For instance, a male college student might be more receptive to an anti-smoking ad featuring a male college student than one featuring an elderly woman. Individuals are also more likely to heed them when they perceive affinity with the author of the messages (i.e. does the targeted individual believe that the person sponsoring the messages is similar to them?). Andsager and colleagues found that college students were significantly more receptive to a written public health message about tanning when they perceived themselves to be similar to the message’s main character (12). Depending on the experimental conditions, the main character, Chris, either mentions drinking alcohol or does not mention drinking alcohol, and either does or does not mention spending time with friends. There were four conditions total (alcohol/friends, no alcohol/friends, alcohol/no friends, and no alcohol/no friends), and participants received different messages depending on the condition. Participants deemed themselves to be most similar to Chris, and most receptive to the message, in the alcohol/friends condition. The reason is that, in this condition, Chris most resembled the participants, most of whom drank alcohol and enjoyed socializing with friends.
On a surface level, it may seem as though Control Tonight harnesses perceived similarity to increase its appeal. The name and sex of the viewer and his/her friends are incorporated into the storylines. This tactic encourages the viewers to imagine themselves in the situations described in the storylines. However, the actual situations described in the plotlines are significant barriers to inducing perceived similarity. The situations are both very specific and somewhat unlikely, and these factors make it difficult for the viewers to feel like the scenarios could happen to them. The sports bar situation is unlikely to resonate with those who are not avid sports fans. Even amongst those who are sports fans, most have not had the experience of getting arrested for physically assaulting an opposing fan. The party situation is more general than the sports bar situation; lots of people have planned a party for friends. However, how many people have driven a car after doing a keg stand, and how many people have been hospitalized for alcohol poisoning? The viewers will feel (or at least want to feel) unlike the characters in the stories, due to the extreme nature of their mistakes. Consequently, they may ignore the anti-binge drinking message.
The sensationalism of the stories undermines the credibility of the authors, causing the viewers to feel unlike the people who wrote Control Tonight. The insensitive treatment of date rape in the advertisement makes the creators seem immoral: they will use any method at their disposal to make drinking seem dangerous. Few people will feel affinity with, or want to listen to, someone who implies that women who drink alcohol are “getting what they deserve” if they become victims of date rape (10).
Failure to address implications of optimistic bias theory
Human beings, for better or for worse, have a tendency to underestimate the likelihood that bad things will happen to them. They also overestimate the likelihood that good things will happen to them. This concept is known as unrealistic optimism, or the optimistic bias (13). In a pioneering study on this concept, Weinstein surveyed college seniors and asked them to rate their chances of experiencing a variety of negative and positive life events (13). They were specifically instructed to rate their chances relative to their college classmates, rather than the population at large. In general, the participants tended to rate themselves as less likely than their peers to experience negative life events (i.e. “attempting suicide”), and more likely to experience positive events (i.e. “traveling to Europe”). The magnitude of this effect was particularly strong for events that were viewed as being controllable. For negative events that were deemed controllable, participants seemed to form a stereotype of people who experience the hardship in question. Then, they compared themselves to this stereotype and deemed themselves different from it, and thus less likely to experience the negative event (13).
“Having a drinking problem” was one of the 24 negative events for which participants showed a particularly significant optimistic bias. Participants felt themselves to be especially unlikely to develop a drinking problem. Of course, it is mathematically impossible for most participants to be less likely to develop a drinking problem than the other participants in the study. It seems that the participants in this study perceived themselves to be largely immune from problems with alcohol. The participants may have created a stereotype of a problem drinker and than compared themselves favorably to that stereotype. As the reasoning goes, alcoholism is something that happens to other people, who develop it because of character weaknesses and/or poor life choices. (The truth is that it can happen for a variety of reasons that one cannot control, such as genetic predisposition.) The high undesirability of alcoholism is another factor that created an unrealistically low expectation of personal risk.
College-age individuals as a population are particularly likely to have an optimistic bias regarding risky activities (14). Young adults are more likely than their parents to rate risky activities (such as drug and alcohol usage) as less likely to harm them than their parents (14). This is particularly true for experimental, rather than frequent, involvement in risky behaviors (14). They are less likely to believe that engaging in threatening activities will result in personal harm.
Even when individuals acknowledge a given behavior to be harmful in general, it does not necessarily mean that they perceive the behavior to be harmful to themselves. Part of the optimistic bias is the belief that one is personally immune to the effects of behaviors that are harmful to others. Optimistic bias may be both an effect and a cause of unhealthy drinking in young adults. Wild and colleagues studied risk perception of heavy alcohol usage amongst college students (15). The participants were asked, “to what extent do you believe that you would be personally at risk of getting hurt or sick because of your drinking?” and, “to what extent do you believe that some other person your age who drinks the way you do would be at risk of getting hurt or sick?” Overall, the participants rated the likelihood of harm to be significantly higher for the second question than for the first. They believed themselves to be less likely to experience drinking-related harm than a hypothetical peer with identical drinking habits. This effect was especially strong for the heaviest, riskiest drinkers. Another study found that college students who believed themselves to be at a lower risk of having alcohol problems than their peers, when controlling for their actual drinking habits, were significantly more likely to experience drinking-related harm in the future (16). Optimistic bias contributes to unhealthy drinking habits.
Control Tonight seems to be based on the idea that exposing young adults to worst-case scenarios of binge drinking will cause them to feel personally at risk of experiencing similar worst-case scenarios. However, research on optimistic bias and alcohol usage challenges this idea. Believing that binge drinking can be harmful in general is not the same as believing that binge drinking will result in harm to oneself. The gap between these two beliefs is especially wide for the groups the intervention is trying to reach: young adults and those with risky drinking behaviors (14-16). Control Tonight may actually exacerbate feelings of optimistic bias in the target audience. Many young adults binge drink (1,4,8), but few experience the extremely negative events depicted in the storylines. They are likely to feel that their personal experiences are more accurate indications of binge drinking’s effects than the over-the-top claims of a public health message. Someone who drinks heavily but has never assaulted a stranger or been rushed to the ER due to drinking may believe that s/he is smarter, more careful, or more experienced than the Control Tonight characters, and thus the situations described in the intervention do not apply to them. Control Tonight could serve to affirm an individual’s flawed belief in personal exemption from drinking-related harm: “horrible things may happen to others when they drink, but they don’t happen to me.” It is particularly easy to feel optimistic bias and distance oneself from a group of harm-prone people if that group can be easily stereotyped. Some of the Control Tonight characters can be stereotyped as showing blatantly poor judgment. This could cause the viewer to feel superior to them, and thus exempt from the problems that befall them.
Creation of a reactance motivational state
No one likes to be told what they can or cannot do. This concept forms the basis of reactance theory (11). Reactance theory states that people deeply value their sense of autonomy and personal freedom. If people believe their freedom has been unfairly curtailed, they may respond by rebelling (11). When people are prohibited from doing something, they often react by refusing to comply with the prohibition. This “reactance motivational state” is particularly strong amongst young adults, who tend to place a high premium on individuality and rebellion (6). Some scholars refer to reactance theory as the “forbidden fruit” argument (6).
Studies on youth drinking have addressed the role of reactance theory in alcohol consumption. Two studies found a strong increase in drinking amongst those under the age of 21 when 21 became the minimum legal drinking age. College students under the age of 21 became significantly more likely to drink after it became illegal for those in their age group to purchase alcohol (6). Similarly, college students under the age of 21 were found to be more likely to drink more than those over the age of 21, although similar illegal drug consumption levels were found in both age groups (7). Presumably, the age-based prohibition made the younger students desire alcohol more, whereas laws against drugs affected both age groups equally (7). This pattern of behavior is consistent with reactance theory: if someone tells you that you shouldn’t do something, you may want to do it even more. While Control Tonight does not target underage drinking specifically, these studies are important because they show how people respond to others’ attempts to limit their drinking.
Bensley and Wu found that college students perceived dogmatic and pro-abstinence alcohol-prevention messages to be less effective than messages that were less negative (17). When exposed to a harsh, high-threat message, participants showed greater intentions to drink alcohol than those exposed to a low-threat message. Similarly, participants drank significantly more alcohol during a sham alcohol taste test after being exposed to the high-threat message. The authors concluded, “dogmatic alcohol prevention materials may have counterproductive effects for some college students” (17). Similar findings were found in a study conducted on Canadian college students: Negative anti-drinking ads inducing feelings of guilt, shame, and fear seemed to cause more problem drinking (18).
Control Tonight certainly falls under the category of a negative message. It focuses on worst-case examples of binge drinking and suggests that similar results are likely to befall the viewers of the messages. It describes consequences in extreme terms, and it features examples of selfish, dangerous, and reckless behavior. Those who do drink excessively are unlikely to agree with Control Tonight. Instead, they are likely to respond by feeling angry, upset, and personally attacked. Because binge drinking is a common part of American youth culture, many young adults view it as an essential part of themselves. One possible consequence of viewing this campaign could be binge drinking to “prove them wrong.” Viewers might be inclined to binge drink and demonstrate that their drinking is different, regardless of whether it actually is, and to affirm their freedom to make decisions. They may want to prove (if only to themselves) that they can handle their alcohol, have a good time, and not lose control. Control Tonight is essentially a challenge to the viewer: if you binge drink, this is what will happen to you. Viewers will be eager to validate their life choices, rebel against the PA Liquor Board, and show that they are different. Control Tonight’s tone is harsh enough to create reactance, and reactance is likely to exacerbate unhealthy drinking.
A better approach to binge drinking prevention: The intervention
What can be learned from Control Tonight’s mistakes? To create a public health campaign that will curb binge drinking amongst young adults, the following conditions should be met. The viewers should feel affinity with the characters in the campaign, and they should not be made to doubt the integrity of its creators. The campaign should recognize the implications of optimistic bias. It should emphasize personal, real-life stories, and it should use optimistic bias advantageously, by making the viewers feel empowered to perform positive behaviors. Finally, it should avoid a harsh, condemnatory tone so that the viewers do not feel motivated to rebel against the campaign.
The “Good Friends, Good Nights” campaign would share some similarities with Control Tonight. It would have a harm reduction approach and focus on limiting binge drinking, rather than stopping drinking altogether. It would be geared towards young adults, both under and over the age of 21. This would allow it to reach a wide range of viewers and not waste resources on an intervention that only appeals to a restricted age group. It would use clear, simple language, without excessive slang. It would include practical tips on healthy drinking, such as those found in Control Tonight’s Field Guide. However, the content and overall tone of the campaign would be quite different.
“Good Friends, Good Nights” is the title of my proposed anti-binge drinking campaign. Part 1 of the campaign would consist of print ads, billboards, and online ads, all of which would refer viewers to a website. The website, part 2, would constitute the core of the campaign. Each of the part 1 ads would display one of several pictures and a text section. The pictures would show groups of smiling people, either talking in a restaurant/bar, outside a sports stadium, at a concert, at a party, or dancing, depending on the picture. The ads would show a mix of ages, races, and both men and women. Some pictures would seem to depict college students (i.e. an outdoor party featuring people wearing college sweatshirts) , and others would seem to depict post-collegiate adults (i.e. professionally-dressed people at a bar). Some, but not all of the people, would be shown with glasses or cups in their hands. Ideally, most viewers would be able to identify with at least one person depicted in one of the pictures. The text would contain messages such as, “Be a good friend. Relax, have fun, and help your friends stay safe,” “help your friends have a good night, and a good morning,” and “make your nights fun, long, and memorable,” and then state, “visit our website to find out how.” These messages would have connotations with alcohol usage but would not explicitly reference alcohol.
The main component of part 2, the website, would consist of personal, real-life stories from actual young adults, along with their first names, ages, job/student status, home states, and a statement that they are the real first names of actual people. The stories would depict a mixture of experiences. Employees of “Good Friends, Good Nights” would visit colleges, bars, and workplaces. They would find young adults and ask to survey them in exchange for modest financial compensation. The open-ended surveys would include prompts for personal stories about experiences with alcohol, as well as alcohol-free ones. Stories would be chosen and edited to represent the following categories of experiences, to create twelve stories total: stories of fun nights with friends in which no drinking was involved (including stories from non-drinkers), stories of fun nights in which moderate drinking with friends was involved, and stories in which people either stopped their friends from drinking excessively or helped their friends after their friends drank excessively (either by making sure they got home safely, taking them to the emergency room, or helping them seek treatment for alcoholism). The stories will emphasize the value of good friendship and the ways in which friendship enriches our lives. The inclusion of the no-drinking stories is important to indicate that drinking is not necessary in order to have a good time. The website would have a Field Guide section with tips on healthy, safe drinking, much like the guide in Control Tonight. It would also contain resources for helping to end problem drinking habits. Visitors to the site could enter in their zip code and receive information on local substance abuse counselors and centers, to be used either by their friends or themselves.
Why “Good Friends, Good Nights” will work: insights from social science
This campaign would make the viewer feel similar to the people depicted in the materials. This similarity would be created on the levels of external characteristics as well as deep core values. In doing so, it would increase the viewer’s receptiveness to the campaign’s message (11,12). The pictures and stories will depict young adults with a variety of ages, ethnic backgrounds, home states, and occupations. It is likely that most young adult viewers will feel a sense of affinity with them due to a shared characteristic, and because they are real people (11,12). They will also want to feel like them, as the campaign’s characters are being depicted in a positive light.
On a deeper level, viewers will feel like the characters because of the values they seem to espouse. Having shared life values can be a strong basis for a sense of interpersonal similarity (11). The main value emphasized by this campaign is friendship: being a good friend and helping others. Most people like to think of themselves in positive terms, and thus most people like to think of themselves as being good to their friends (13). This means that the viewer will feel like the characters in the campaign, who are depicted as enjoying time spent with friends and demonstrating the values of caring and compassion. The personal stories on the website will strike a balance between specific and general: specific enough to be vivid, but not so specific that they are narrow in scope or are only relatable to a select few. The fact that the stories are based on real people, rather than fictional characters, would also help create a sense of similarity. Finally, the viewers would be unlikely to question their affinity with the campaign’s creators, as the campaign appeals to them on positive terms and does not use sensationalized fictional stories. By creating a sense of similarity, the campaign will increase the likelihood that the viewers will heed its message (11,12).
Optimistic bias is a vexing problem for many public health practitioners. Even when people recognize that a given behavior is dangerous in general, they won’t necessarily feel that the behavior is dangerous to themselves personally (13-16). However, optimistic bias cuts both ways: It also means that people have unrealistically positive expectations of their character strengths and their ability to perform positive behaviors (13). “Good Friends, Good Nights” has an optimistic tone. Its message will encourage the viewers to “act like good friends” and help give them the confidence to do so. Because the viewers will want to think of themselves positively, they will want to emulate the responsible behaviors modeled by the characters in the website. The intervention is focused on how one acts towards others, but it should also serve to reduce binge drinking in the viewers as well. Ideally, the campaign would help set a new norm of positive, responsible drinking. The goal is to present many good examples of drinking behaviors. The viewers will want to feel on par with, or superior to, these examples, and will begin to modify their behaviors accordingly.
“Good Friends, Good Nights” has a light-handed approach. Rather than challenging the viewers and telling them that their lifestyle is wrong, it gives examples of positive behaviors and then lets the viewers connect the dots on their own. The campaign essentially identifies a solution without excessively pointing fingers at the problem. Thus, it avoids the main pitfalls of critical anti-binge drinking interventions, which often cause the viewers to react angrily and then ignore their advice (16-18).
This campaign avoids creating a reactance state due to its positive tone, its harm reduction approach, its empowering message, its appeal to the value of friendship, its use of real-life first-person narratives, and its lack of direct criticism. A certain amount of negativity is necessary when discussing the binge drinking of the characters’ friends. However, the reactance caused by this negativity will be minimized because the viewers are being addressed as the friends of binge drinkers, rather than the binge drinkers themselves. The characters in the website’s stories are describing how they behaved in positive drinking situations and how they helped friends who were binge drinking, not how they changed their own drinking habits. In other words, the viewers are being treated as responsible drinkers who should help keep their friends safe, rather than irresponsible drinkers who need to change their own habits and who require help from others. The intervention has two goals: to help the viewers encourage their friends to drink responsibly, and to help the viewers drink responsibly themselves. The second goal of this intervention, however, is somewhat hidden. Asking viewers to change their drinking directly might be interpreted as a challenge to their freedom. However, asking viewers to change their friends’ drinking will allow them to internalize the message of responsible drinking without causing them to feel personally attacked. The use of real-life stories will reduce reactance because the viewers will be less likely to feel that an authority figure is trying to change them. Instead, the messages seem to be coming from their peers. The approach described in this paper would not stop binge drinking altogether, but it should prove more effective at reducing its prevalence than the negative campaigns that have targeted young adults in recent years.
References
1.The National Center on Addiction and Substance Abuse at Columbia University. Teen Tipplers: America’s Underage Drinking Epidemic. New York, NY: The National Center on Addiction and Substance Abuse, 2003.
2. Town M, Naimi TS, Mokdad AH, Brewer RD. Heath care access among U.S. adults who drink alcohol excessively: missed opportunities for prevention. Preventing Chronic Disease 2006; 3: 1-8.
3. Centers for Disease Control and Prevention. Alcohol-attributable deaths and years of potential life lost -- United States, 2001. Morbidity and Mortality Weekly Report 2004; 53: 866-870.
4. Naimi TS, Brewer RD, Mokdad A, Denny C, Serdula MK, Marks JS. Binge drinking among US adults. Journal of the American Medical Association 2003; 289: 70-75.
5. Lanza-Kaduce L, Richards P. Raising the minimum drinking age: some unintended consequences of good intentions. Justice Quarterly 1989; 6: 247-262.
6. Engs RC, Hanson DJ. Reactance theory: a test with collegiate drinking. Psychological Reports 1989; 64: 1083-1086.
7. Allen DN, Sprenkel DG, Vitale PA. Reactance theory and alcohol consumption laws: Further confirmation among collegiate alcohol consumers. Journal of Studies on Alcohol 1994; 55: 34-40.
8. U.S. Department of Health and Human Services. Binge Drinking in Adolescents and College Students. Washington, D.C: U.S. Department of Health and Human Services, 2000.
9. Pennsylvania Liquor Control Board. Control Tonight. PA: Pennsylvania Liquor Control Board, 2011. www.controltonight.com .
10. The Huffington Post. PA Liquor Board Ad Accused of Blaming Rape on Victims. The Huffington Post, 2011. http://www.huffingtonpost.com/2011/12/08/liquor-control-board-ad_n_1137083.html
11. Silvia PJ. Deflecting reactance: the role of similarity in increasing compliance and reducing resistance. Basic and Applied Social Psychology 2005; 27: 277-284.
12. Andsager JL, Bemker V, Choi H, Torwel V. Perceived similarity of exemplar traits and behavior: effects on message evaluation. Communication Research 2006; 33: 3-18.
13. Weinstein ND. Unrealistic optimism about future life events. Journal of Personality and Social Psychology 1980; 39: 806-820.
14. Cohn LD, Macfarlane S, Yanez C, Imai WK. Risk-perception: differences between adolescents and adults. Health Psychology 1995; 14: 217-222.
15. Wild TC, Hinson R, Cunningham J, Bacchiochi J. Perceived vulnerability to alcohol-related harm in young adults: independent effects of risky alcohol use and drinking motives. Experimental and Clinical Psychopharmacology 2001; 9: 117-125.
16. Dillard AJ, Midboe AM, Klein WMP. The dark side of optimism: unrealistic optimism about problems with alcohol predicts subsequent negative event experiences. Personality and Social Psychology Bulletin 2009; 35: 1540-1550.
17. Bensley LS, Wu R. The role of psychological reactance in drinking following alcohol prevention messages. Journal of Applied Social Psychology 2006; 21: 1111-1124.
18. Mullman J. Why binge drinking PSAs may leave some reaching for another drink. Advertising Age, 2010. http://adage.com/article/news/study-binge-drinking-psas-exacerbate-problem/142459/
posted by Melanie Pennison @ 10:43 AM
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home