Effects of Framing on Petition Signing for a Smoke-Free Campus Laura Admiraal - 10800697
Bachelor Thesis Social Psychology Dhr. dr. G. Dik
University of Amsterdam June 8th 2018
Abstract
This study analyses the influence of framing on signing the petition for a smoke-free campus among students and faculty members of the University of Amsterdam., including smokers and non-smokers. Framing theory states that presenting an informationally invariant message in a gain frame compared to a loss frame influences decision-making. On the one hand, the gain frame presents the positive consequences of a smoke-free campus, while on the other hand, the loss frame emphasises the positive consequences that are missed when not
implementing a smoke-free campus. First of all, results showed that non-smokers had a more positive attitude towards a smoke-free campus than smokers, thereby supporting H1.
Secondly, framing did not influence people’s attitudes toward a smoke-free campus, contrary to H2. Thirdly, for smokers the chance of signing the petition for a smoke-free campus is higher after reading a loss rather than a gain frame. This contradicts H3. Lastly, no
moderating effect was found of the visceral state of nicotine craving among smokers on the relationship between framing and signing the petition for a smoke-free campus. However, more time since someone had last smoked increases the chance of signing the petition. These findings are in conflict with H4. Finally, possible explanations for these unexpected findings and recommendations for future research are discussed.
Effects of Framing on Petition Signatures for a Smoke-Free Campus Smoke-free policies at universities
As is broadly known, smoking comes with serious health risks for the smoker and the non-smokers inhaling the tobacco fumes. Associations between smoking and lung cancer are considered general knowledge, but smoking also increases the risks on many other types of cancer, including bladder, pancreatic and kidney cancer. Causality has also been found between cancer and coronary hart and cardiovascular diseases (US Department of Human Health and Services, 2004). Smoking is the leading cause of preventable deaths and causes 480,000 deaths per year in the United States of America alone and therefore accounts for one out of five deaths. Moreover, smokers die on average ten years earlier than non-smokers (Jamal et al., 2018). Passive smoking is associated with many health risks as well. Of the aforementioned 480,000 annual smoking-related deaths, second-hand smoking accounts for 41,000 of them (US Department of Human Health and Services, 2014). Furthermore, inhaling tobacco smoke is related to multiple negative health outcomes, ranging from food allergies to certain types of cancer (Cao, Yang, Gan, & Lu, 2015). Thus smoking, active and passive, negatively influences the quality of life and is associated with higher mortality risks
(Carbone, Kverndokk, & Røgeberg, 2005). These facts illustrate that smoking requires strict regulation, not only to protect smokers, but non-smokers as well.
In many countries indoor smoking bans are already at force (World Health Organization, 2008). This has led to 5 million people that have quit smoking and approximately 2,5 million less smoking-related deaths worldwide (Callinan, Clarke, Doherty, & Kelleher, 2010). Outdoor smoking is harmful for non-smokers as well, because smoke that is close to an entrance can enter the building and passers-by unintentionally inhale the smoke (López et al., 2012). This is a substantial problem, since people often smoke near entrances (Sureda,
of the students and 73% of the faculty the university campus is the place where they have most contact with second-hand smoke in daily life (Lupton & Townsend, 2015). Despite the bans on indoor smoking at universities, still 83% of students at American universities report contact with second-hand smoke (Wolfson, McCoy, & Sutfin, 2009). Moreover, a study comparing a smoke-free campus with a university that had only implemented outdoor regulations prohibiting smoking near entrances, revealed similar results. The number of smokers at the smoke-free campus university decreased from 16.5% to 12.8%, whereas the number slightly increased at the university with the outdoor regulation (Seo, Macy, Torabi, & Middlestadt, 2011). Other important findings include the steeper decline in daily smoking at the smoke-free campus. Lastly, students at the smoke-free campus showed a 5% increase in their support for smoke-free policies after the implementation, opposed to a 3% decrease among the students at the outdoor regulated university (Seo et al., 2011). This demonstrates that neither indoor smoking bans, nor outdoor restrictions on smoking near entrances, are enough to prevent students from second-hand smoking. Thus calling for smoke-free campuses.
Now that the importance of smoke-free campuses is clear, the feasibility of the policy must be assessed. The first step is to gauge endorsement. Lupton and Townsend (2015) did this at universities in the United States and the United Kingdom. This showed overall support for a smoke-free campus. The support among faculty members (68.4%) is higher than among students (58.9%). Even more so, after implementation the support for the smoke-free campus had increased compared to before the ban (Lupton & Townsend, 2015). This tendency is common and mirrors the post-ban increase in support for indoor smoking bans in restaurants and other public places (Brown, Moodie, & Hastings, 2009). Concluding, the endorsement among students and faculty members for a smoke-free campus seems to be positive and increases after implementation. Past research did not differentiate support among smokers as
opposed to non-smokers. However, since smoking on campus causes many health related risks for both non-smokers and smokers in combination with only smokers that get the benefits of nicotine satisfaction, it is expected that non-smokers have a more positive attitude towards a smoke-free campus.
H1: Non-smokers have a more positive attitude towards a smoke-free campus than smokers.
The United States are currently the leader when it comes to smoke-free campuses. A quarter of the American universities are already smoke-free (American Nonsmoker’s Rights Foundation, 2018). In The Netherlands no such smoke-free campuses exist, yet. This study is commissioned by the University of Amsterdam to assess the endorsement and feasibility. Framing effects
A practical psychological tool that could be useful in increasing support for a smoke-free campus is framing. This derives from the prospect theory by Kahneman and Tversky (1979) that states that people tend to evaluate an objective outcome differently depending on how it is presented. Heuristics are often used in decision-making, these are short-cuts that decrease the cognitive load of the process. This limits the amount of information that is processed, thereby sometimes leading towards a wrong or irrational decision. Important is that the prospect theory demonstrates the lack of invariance of information. Meaning that although the information in two messages is exactly the same, the way it is described can influence people’s decision. Framing states that people react differently towards objective outcomes that are framed as avoiding a loss or as obtaining a gain (Tversky & Kahneman, 1981). People become more risk averse when making decisions based on gain-framed information. This is called the certainty effect (Kahneman & Tversky, 1979). On the other hand, when presented with a loss-frame, people tend to prefer uncertain over certain outcomes. So, people take more risks if they perceive a decision as a potential loss, as opposed to a potential gain.
Overall, people have a bigger aversion for losing than affection with winning, so avoiding losses often gets prioritized (Kahneman & Tversky, 1979).
Framing has mainly been deployed in the health sector. It aims to shape people’s health behaviour positively, by framing health messages in a certain way. It differs per health goal whether a loss-frame or a gain-frame is most effective. This depends on whether the health behaviour is perceived as having a high or low risk. This deserves some explanation, the risk associated with health behaviour refers to the possible outcome of that behaviour. For
example, detection tests are perceived as high risk health behaviour, because a possible outcome is being diagnosed with a life threatening disease. An example is a pap smear test to detect cervical cancer (Unim, Meggiolaro, Semyonov, Maffongelli, & La Torre, 2014) . On the other hand, preventive health behaviours such as quitting smoking are associated with low risk, since all possible outcomes are positive, with the exception of withdrawal symptoms (Rothman & Salovey, 1997). This risk perception determines what type of frame is most effective.
One of the first studies to examine the effects of framing on health behaviour analysed whether framing could be deployed to increase the number of women that perform a breast self-examination, a test to detect breast cancer (Meyerowitz & Chaiken, 1987). Women were asked to read a pamphlet that either emphasized the positive consequences of performing the examination or the negative consequences of not performing it, respectively gain and loss frames. As expected, women who read the loss frame had more positive attitudes, intentions and behaviours towards self-examination, as opposed to those who read the gain frame (Meyerowitz & Chaiken, 1987). This is in line with the theory, since a detection test is perceived as risky and therefore a loss-frame is more effective. With regard to framing and prevention behaviour, many studies were interested in smoking cessation. Listening to an auditory accompanied visual messages on the consequences of smoking had more impact on
smoking-related beliefs, attitudes and behaviour when framed as a gain instead of a loss (Schneider et al., 2001). Thirty days after receiving the message, participants that had seen a gain frame message reported smoking less than those that had seen a loss frame message. Additionally, a later study found similar results using gain or loss framed texts. Thus results demonstrated that texts emphasizing the benefits of not smoking led to greater intention to quit smoking than texts emphasizing the costs of smoking (Steward, Schneider, Pizzaro, & Salovey, 2003). In short, research supports the notion that preventive behaviour is better stimulated with a gain-frame and detection behaviour with a loss frame.
The health behaviour that this research is interested in, supporting a smoke-free campus by signing a petition, is a type of prevention behaviour, because it prevents smokers from health risk associated with smoking. Namely, at a smoke-free campus smokers tend to smoke less cigarettes per day and inhale less second-hand smoke (Seo et al., 2011). Signing a
petition that supports this, is therefore considered to be prevention behaviour and is consequently better stimulated using a gain frame as opposed to a loss frame (Rothman & Salovey, 1997). Therefore it is expected that a message that conveys the consequences of a smoke-free campus is more persuasive for smokers in a gain frame than a loss frame. Thus it is expected:
H2: Smokers will be more positive towards a smoke-free campus after reading a gain framed message rather than a loss framed message.
Logically, this positive attitude is expected to be transferred into corresponding positive behaviour, namely signing the petition for a smoke-free campus. Therefore, it is predicted:
H3: More smokers will sign the petition for a smoke-free campus after reading a gain frame than after reading a loss frame.
A side note that must be made, is that although this evidence seems coherent and clear, preference and effectiveness of loss versus gain frames can differ between people. These
differences can be the result of different prior attitudes towards the health behaviour or the potential patients in question (Rothman & Salovey, 1997). For example, the framing effects are smaller for stigmatized populations. One experiment showed that when a problem was framed as the number of potential lives lost, the option that conveyed a certain number of people dying was more often assigned to AIDS patients than leukemia patients. The inversed effect was found when the problem was framed as number of potential lives saved (Levin & Chapman, 1993). Thus, contextual or even cultural factors may influence the effects of framing on an individual level.
Risk perception of health behaviour is influenced by multiple factors. When it comes to smoking, such a factor could be visceral states. Visceral states refer to the degree to which physical needs, such as hunger, thirst and energy are satisfied. Dissatisfaction of a physical need will impact someone’s behaviour. This impact is structurally underestimated by people, due to the cold-to-hot empathy gap, which describes that it is difficult for people to imagine being hungry or tired and it is even harder to determine what impact that has on their own behaviour (Nordgren, Van Der Pligt, & Van Harreveld, 2006). For example, when people that have just eaten make predictions about the ability to stick to a diet, they overestimate this ability, because they underestimate the impact of hunger. In other words, they systematically underestimate the risk of feeling hungry on sticking to a diet. Smoking forms no exception to this rule. Smokers often claim that they could quit smoking if they wanted to. Making this statement, the risk of nicotine craving on smoking cessation is underestimated (Nordgren et al., 2006). Logically, if a factor influences risk perception, it automatically influences the effect of framing on health behaviour. If people perceive more risk, because they are experiencing the consequences of nicotine deficiency, a loss frame becomes more effective than a gain frame. Vice versa, if people have recently smoked and thus have low nicotine
craving, the risk perception is low and a gain frame is more effective than a gain frame. Therefore, the fourth hypothesis is:
H4: For smokers, the visceral state of nicotine craving moderates the relationship between framing and signing the petition for a smoke-free campus, in such a way that higher nicotine craving is associated with a stronger effect of loss frame as opposed to gain frame on the chance of signing the petition.
This study aims to improve the understanding of the effects of framing health messages on attitudes and behaviour concerning smoking, in this case signing a petition for a smoke-free campus. Figure 1 illustrates the theoretical framework.
Figure 1. Theoretical framework
Method
Participants
The survey used in this study is distributed through the intranet of the University of Amsterdam and the social channels of the author. Also, students and faculty members at the campus were approached and asked to fill out the survey. The only exclusion criterion was
that all participants had to study or work at the Roeterseiland campus of the University of Amsterdam. Additionally, a power analysis using Gpower with a significance level of .05 and a desired power of .80 was conducted (Faul, Erdfelder, Lang, & Buchner, 2007). This yielded a minimum number of required participants, namely 150 smokers and 75 non-smokers. Materials
This study uses only a survey, which starts with presenting the participant with either a gain frame or a loss frame. The gain frame stated that a smoke-free campus decreases the risks of various diseases and increases the quality of life. Complementary, the loss frame stated that a campus where it was allowed to smoke the risk of various diseases increases and quality of life decreases. For the moderator visceral state the following question was included in the questionnaire: “How many minutes ago have you smoked your last cigarette?” Attitude towards a smoke-free campus was measured by the question “To what extent are you against or in favour of making the Roeterseiland campus smoke-free?” on a scale from 0 (strongly against) to 10 (strongly in favour). The behavioural outcome measure was digitally signing the petition for a smoke-free campus at the end of the survey.
Control variables were included to check whether the assumption about the underlying mechanism were correct. Risk perception was measured through the question “How positively or negatively will a smoke-free campus impact you?” and answered by placing the cursor along the line of 0 (negative) to 100 (positive). A self-report measure of craving was included that provided information about the operationalization of nicotine craving, namely “How badly do you crave a cigarette at this moment?” on a scale of 0 (not at all) to 100 (extremely), again by placing a cursor along this line. Because this self-reported craving is more subjective than asking someone when they smoked last, visceral state analysis were conducted with the latter. Yet, the two variables are expected to show some kind of positive correlation, since more time since someone’s last smoked cigarette is
expected to be associated with higher nicotine craving. Finally, people reported the average amount of cigarettes they smoked per day.
Analyses
First of all, standardisation checks were executed to account for confounding effects of gender and age. Concerning gender, a Chi Squared Test was used to analyse the proportion of men and women in the different conditions of framing. To check if people in one condition differ in age from people in the other condition, an independent samples t-test was used.
Thereafter, the following hypotheses have been tested. To assess whether
non-smokers have a more positive attitude about a smoke-free campus compared to non-smokers (H1) an independent samples t-test was used. To analyse if smokers are more positive about a smoke-free campus after reading a gain frame as opposed to a loss frame (H2) a logistic regression was performed, so that smokers could also be compared to non-smokers. Testing whether the influence of reading a gain or loss frame on signing the petition is moderated by smoking (H3) was also executed with a logistic regression. Lastly, the moderating effect of visceral states on the relationship between framing and signing the petition (H4) was analysed with a logistic regression as well. All logistic regression were carried out with Hayes’ (2017) process tool.
Results
The survey was completed by 273 people, of whom 46 were excluded for not meeting the inclusion criteria. For example, some had never been at the Roeterseiland campus, did not except the terms and conditions of this research or stopped the survey too quickly. Therefore, analyses are carried out over 277 participants, of whom 152 (67%) were smokers and 75 were non-smokers (33%).
The standardisation checks demonstrated equal distribution of gender and age across the conditions of gain and loss frame. First of all, a Chi Squared test showed that the
distribution of sex was equal across both conditions, χ2(1) = .74, p = .39. This means that the number of men and woman does not differ significantly across the two conditions of loss and gain frame. So, results cannot be attributed to gender differences across conditions. In
addition, an independent samples t-test showed that people in the loss frame condition (M = 22.49, SD = 2.24) are not significantly younger than those in the gain frame condition (M = 22.55, SD = 2.80). So, results also cannot be attributed to age differences across conditions. Next, the results of hypotheses testing are discussed.
Firstly, an independent samples t-test shows that non-smokers (M = 7.31, SD = 2.02) have a more positive attitude about a smoke-free campus than smokers (M = 4.55, SD = 2.53), F(180.27) = 7.24, p < .001, d = 1.20. Thus, the first hypothesis (H1) is supported.
Secondly, a logistic regression showed that there is no main effect of framing on attitudes about a smoke-free campus, b = -.25, p = .79 (see Table 1). So, there was no
difference in attitude about a smoke-free campus between smokers that read a loss-frame (M = 4.53, SD = 2.61) in comparison to a gain-frame (M = 4.58, SD = 2.46). Therefore the second hypothesis (H2) is rejected. It did result in a main effect for smoking, which further supports H1, b = 2.61, p <.001. No interaction between smoking and framing on attitude was found, b = .30, p = .65.
Table 1. Results of the logistic regression of the moderating influence of smoking on the influence of framing on attitude towards a smoke-free campus.
Thirdly, the possible moderating effect of smoking on the influence of framing on signing the petition is analysed through a logistic regression. Table 2 shows the results. First of all, there is a main effect for smoking on petition signing, in such a way that non-smokers have a higher chance of signing the petition than non-smokers, b = 1.17, p = .01 (see Figure 2). However, for framing no main effect was found, b = -.11, p = .72. Thus reading either a gain or a loss frame has no direct influence on signing the petition. Furthermore, results yielded an interaction of smoking and framing on signing of the petition, b = 1.57, p = .03. Figure 2 shows the direction of the moderation. Evidently, for smokers the chance of signing the petition for a smoke-free campus is higher after reading a loss frame than after reading a gain frame. This is contrary to expectations, thus H3 is rejected. For non-smokers, the opposite is the case, such that reading a gain frame has a higher chance of being followed by signing the petition, than reading a loss frame.
b SE t p Constant 1.92 [.60, 3.24] .67 2.87 .00 Smoking 2.61 [1.67, 3.54] .47 5.50 <.001 Framing -.25 [-2.12, 1.62] .95 -.26 .79 Interaction .30 [-1.02, 1.63] .67 .45 .65
Table 2. Results of logistic regression on the moderating effect of smoking on the influence of framing on petition signing.
Figure 2. Moderating effect of smoking on the influence of framing on the chance to sign the petition. b SE z p Constant .33 [.03, .64] .15 2.15 .03 Smoking 1.95 [1.23, 2.67] .37 5.33 .00 Framing -.11 [-.72, .50] .31 -.36 .72 Interaction 1.57 [.14, 3.01] 0.73 2.15 .03
Lastly, the moderating effect of visceral states, operationalized as the time since participants’ last smoked cigarette, on the influence of framing on petition signing was tested with a logistic regression. This analysis only includes smokers. Results are shown in Table 3 and comprise one main effect and no interaction effect. First of all, there is a main effect for visceral states on signing of the petition, b = .003, p = .03. Looking at Figure 3, this main effect is interpreted as follows: the longer someone has not smoked, the greater the chance of signing the petition. Based on the theory of visceral states and corresponding with H4 the opposite effect would be expected. No main effect for framing was found, although Figure 3 suggest that overall reading a loss frame results in a bigger chance of signing the petition than a gain-frame. Yet, this effect is not significant, b = -.47, p = .18. Finally, results also showed that there is no significant moderating effect of visceral states, b = -.003, p = .28. This means that visceral states do not influence the effectiveness of the two types of frames on petition signing behaviour for smokers. Therefore, the fourth hypothesis (H4) is rejected.
Table 3. Results of binary logistic regression on the moderating effect of visceral states on the influence of framing on petition signing among smokers.
b SE z p Constant -.27 [-.61, .07] .17 -1.54 .12 Visceral states .003 [.00, .01] .00 2.21 .03 Framing -.47 [-1.15, .21] .35 -1.34 .18 Interaction -.003 [-.01, .00] .00 -1.09 .28
Figure 3. Moderating effect of visceral states on the influence of framing on the chance to sign the petition among smokers.
Analysis of the control variables were executed to explain unexpected findings of H2, H3 and H4. Concerning H4, this revealed that the time since someone’s last smoked cigarette is not significantly correlated with the self-report measure of cigarette craving, r = -.138, p = .10. Even though, this was supposed to be the operationalization of nicotine craving and accordingly, these two variables were expected to correlate positively. However, cigarette craving does positively correlate with the amount of cigarettes that people smoke per day, r = .28, p = .001, suggesting that people who smoke more have higher cigarette craving in
general. Also, cigarette craving correlates negatively with signing the petition, r = -.28, p < .001. Supporting this finding, a post-hoc logistic regression showed a main effect of self-reported nicotine craving on signing the petition, such that higher nicotine craving decreased the chance of signing the petition, b = -.02, p <.001.
Moreover, the measure of risk perception did not correlate with the time since someone’s last smoked cigarette, r = .13, p = .11. Risk perception did correlate with
self-reported nicotine craving, r = -.22, p = .001, and with the number of cigarettes that people on average smoke per day, r = -.33, p <.001. These findings are interpreted in the next section, the discussion.
Discussion
In line with the expectations this study found that non-smokers have a more positive attitude towards a smoke-free campus than smokers and sign the petition for a smoke-free campus more often (H1). Yet, framing had no direct effect on attitudes, nor did it interact with smoking (H2). Nevertheless, there is a moderating effect of smoking on the
effectiveness of loss and gain frames on the chance to sign a petition for a smoke-free campus. For smokers, reading a loss gain has a higher chance of signing the petition than reading a gain frame. This is in contrast with H3. For non-smokers, the opposite effect was found, namely that reading a gain frame results in a higher chance of signing the petition than reading a loss frame. Finally, H4 predicted a moderating effect of visceral states, the time since the last smoked cigarette in minutes, on the influence of framing on signing the petition. It was predicted that having high nicotine craving increases the effectiveness of a loss frame as opposed to a gain from on the chance of signing the petition. No such interaction has been found, therefore H4 is rejected. In contrast to the theory of visceral states, there is a main effect for visceral states, namely that higher nicotine craving leads to higher chance of signing the petition. Next, possible explanations for these unexpected findings are offered.
First of all, a peculiar finding is that on the one hand the interaction of H2 was rejected, while the interaction of H3 was supported. This means that smoking does interact with framing on behaviour, like signing the petition in this case, but not with attitude.
Literature offers an explanation for this finding, because attitudes are not a necessarily a good predictor of behaviour (Ajzen, 1991). Research on the effect of framing on attitudes and behaviour shows contradictory findings. On the one hand, many studies showed framing
effects both prevention attitudes and behaviour (Meyerowitz & Chaiken, 1987; Chang, 2007). Moreover, effects of framing on attitudes and behaviour regarding smoking have been found (Goodall & Appiah, 2008; Moorman & Van Den Putte, 2008). These studies support H2. However a meta-analysis found that framing does have an effect on prevention behaviour, but not on corresponding attitude (Gallagher & Updegraff, 2011). This explains the unexpected findings of H2.
Explanations for the unexpected finding that a loss frame is more effective for smokers than a gain frame, might be found in the definition of risk. This study defined risk perception merely as perception of health risks associated with signing the petition for a smoke-free campus. Even though no health risks are associated with this action, other types of risk might be. One example are social risks. Smoking is viewed as a social activity, especially among students (Moran, Wechsler, & Rigotti, 2004). Prohibitng smoking on campus can therefore be perceived as a threat to social contact. Another example is the risk of withdrawal symptoms. If students and faculty members are not allowed to smoke on campus, it is possible that they are not able to smoke for multiple hours, when they have a dense schedule. Research showed that after twelve hours of not smoking, smokers report more negative affect, distress and the urge to smoke (Weinberger, Platt, Shuter, & Goodwin, 2016). Thus social and withdrawal risks can increase the perceived risk associated with signing the petition and therefore affect the influence of framing. Framing theory states that behaviour that is associated with risks is more effectively influenced through gain than loss frames. Thus, if smokers perceive signing a petition for a smoke-free campus as risky behaviour, because of social or withdrawal risks for example, a gain frame should yield more positive results than a loss frame. This could explain the unexpected findings of H3.
The finding that higher nicotine craving leads to a higher chance of signing the petition goes against the theory of visceral states. A methodological point of critique could
explain this unexpected finding. Namely, the visceral state of interest in this research was nicotine craving. This was operationalized by measuring the time since someone’s last smoked cigarette. Thus making the assumption that nicotine craving is determined by the time that people have not smoked. Against expectations, no relation was found between reported nicotine craving and time since last cigarette. Moreover, while the time since last cigarette increased the chance of signing the petition, self-reported nicotine craving is positively associated with signing the petition. Another finding that further supports self-reported nicotine craving over the used measure of time, is correlation with risk perception. Namely, the visceral states theory suggest that loss frames will work better for smokers with high nicotine craving, because they overestimate the risk and consequences associated with not smoking, or in this case signing the petition for a smoke-free campus. Logically, the measure of visceral states should correlate with the underlying mechanism of risk perception, such that higher nicotine craving is associated with higher risk perception. Risk perception did not correlate with the time measure of visceral states, but it did with self-reported nicotine craving. In hindsight, self-reported nicotine craving, while being subjective, seems to be a better operationalization of nicotine craving than time since last smoked cigarette. However, the validity of that measure is not known. Thus, future research should use standardized questionnaires for nicotine craving of which validity and reliability are confirmed, like the Tobacco Craving Questionnaire (TCQ-12) and the single craving item of the Minnesota Nicotine Withdrawal Scale (MNC) (Berlin, Singleton, & Heishman, 2013).
A point of critique can be made concerning the sampling procedure. After distributing the survey online, the sample size was inadequate. This required the researchers to ask people on campus to fill out the survey. Because mainly smokers were needed, people who were smoking were approached. This had consequences for the analyses on visceral states because many of the respondents either had had their last cigarette less than fifteen minutes earlier
(45.1%), while another large group had not smoked for at least five hours (38.2%). Data is therefore not evenly distributed and it is almost as if there are two groups in the experiment. In order to make claims about all phases of nicotine craving, a more diverse sample should be collected (Field, 2009). Another flaw concerning sampling is that a smoke-free campus policy affects both students and faculty staff, therefore opinions from both groups are valuable. Even though this survey was open for students as well as faculty staff, of all 227 participants, only 5 were faculty members (2.2%). Hence, results and conclusion of this study cannot be generalized to faculty members, as they were severely underrepresented in the sample (Field, 2009). Future research should thus strive to obtain a more inclusive sample.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179-211.
American Nonsmokers’ Rights Foundation (2018, January). U.S. colleges and universities with smokefree and tobacco-free policies January 2018. Retrieved on February 22, 2018, from:
https://no-smoke.org/wp-content/uploads/pdf/smokefreecollegesuniversities.pdf Berlin, I., Singleton, E. G., & Heishman, S. J. (2013). Predicting smoking relapse with a multidimensional versus a single-item tobacco craving measure. Drug & Alcohol Dependence, 132(3), 513-520.
Brown, A., Moodie, C., & Hastings, G. (2009). A longitudinal study of policy effect (smoke-free legislation) on smoking norms: ITC Scotland/United Kingdom. Nicotine & Tobacco Research, 11(8), 924-932.
Callinan, J. E., Clarke, A., Doherty, K., & Kelleher, C. (2010). Legislative smoking bans for reducing secondhand smoke exposure, smoking prevalence and tobacco
consumption. Cochrane Database Syst Rev, 4(4).
Cao, S., Yang, C., Gan, Y., & Lu, Z. (2015). The health effects of passive smoking: An overview of systematic reviews based on observational epidemiological
evidence. PloS one, 10(10), e0139907.
Carbone, J. C., Kverndokk, S., & Røgeberg, O. J. (2005). Smoking, health, risk, and perception. Journal of Health Economics, 24(4), 631-653.
Chang, C. T. (2007). Health‐care product advertising: The influences of message framing and perceived product characteristics. Psychology & Marketing, 24(2), 143-169.
Goodall, C., & Appiah, O. (2008). Adolescents' perceptions of Canadian cigarette package warning labels: Investigating the effects of message framing. Health
Communication, 23(2), 117-127.
Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191.
Field, A. (2009). Discovering statistics using SPSS. Thousand Oaks, California: Sage Publications.
Gallagher, K. M., & Updegraff, J. A. (2011). Health message framing effects on attitudes, intentions, and behavior: A meta-analytic review. Annals of Behavioral Medicine, 43(1), 101-116.
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. New York, New York: Guilford Publications. Jamal, A., Phillips, E., Gentzke, A. S., Homa, D. M., Babb, S. D., King, B. A., & Neff, L. J.
(2018). Current cigarette smoking among adults—United States, 2016. Morbidity and Mortality Weekly Report, 67(2), 53.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the Econometric Society, 47(2), 263-291.
Levin, I. P., & Chapman, D. P. (1993). Risky decision making and allocation of resources for leukemia and AIDS programs. Health Psychology, 12(2), 110.
Lupton, J. R., & Townsend, J. L. (2015). A systematic review and meta-analysis of the acceptability and effectiveness of university smoke-free policies. Journal of American College Health, 63(4), 238-247.
Moorman, M., & van den Putte, B. (2008). The influence of message framing, intention to quit smoking, and nicotine dependence on the persuasiveness of smoking cessation messages. Addictive Behaviors, 33(10), 1267-1275.
Moran, S., Wechsler, H., & Rigotti, N. A. (2004). Social smoking among US college students. Pediatrics, 114(4), 1028-1034.
Nordgren, L. F., Van Der Pligt, J., & Van Harreveld, F. (2006). Visceral drives in retrospect: Explanations about the inaccessible past. Psychological Science, 17(7), 635-640. O'Keefe, D. J., & Nan, X. (2012). The relative persuasiveness of gain-and loss-framed
messages for promoting vaccination: A meta-analytic review. Health Communication, 27(8), 776-783.
Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121(1), 3-19.
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(4), 667-682.
Seo, D. C., Macy, J. T., Torabi, M. R., & Middlestadt, S. E. (2011). The effect of a smoke-free campus policy on college students' smoking behaviors and attitudes. Preventive Medicine, 53(4-5), 347-352.
Steward, W. T., Schneider, T. R., Pizarro, J., & Salovey, P. (2003). Need for cognition moderates responses to framed smoking‐cessation messages. Journal of Applied Social Psychology, 33(12), 2439-2464.
Sureda, X., Fernández, E., López, M. J., & Nebot, M. (2013). Secondhand tobacco smoke exposure in open and semi-open settings: A systematic review. Environmental Health Perspectives, 8121(7), 766
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 453-458.
Unim, B., Meggiolaro, A., Semyonov, L., Maffongelli, E., & La Torre, G. (2014). Role of pap-test in cervical cancer prevention: A systematic review and
meta-analysis. European Journal of Public Health, 24(2), 124-139.
US Department of Health and Human Services. (2004). The health consequences of smoking: A report of the Surgeon General.
US Department of Health and Human Services. The health consequences of smoking—50 years of progress: A report of the Surgeon General. Atlanta, GA:
US Department of Health and Human Services, CDC, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 2014.
http://www.surgeongeneral.gov/library/reports/50-years-of-progress/full-report.pdf Weinberger, A. H., Platt, J. M., Shuter, J., & Goodwin, R. D. (2016). Gender differences in
self-reported withdrawal symptoms and reducing or quitting smoking three years later: A prospective, longitudinal examination of US adults. Drug & Alcohol Dependence, 165, 253-259.
Wolfson, M., McCoy, T. P., & Sutfin, E. L. (2009). College students' exposure to secondhand smoke. Nicotine & Tobacco Research, 11(8), 977-984.
