Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Feb 1.
Published in final edited form as: Health Psychol. 2017 Nov 20;37(2):188–193. doi: 10.1037/hea0000558

How Do Tobacco Power Walls Influence Adolescents? A Study of Mediating Mechanisms

Claude M Setodji 1, Steven C Martino 1, Min Gong 1, Michael S Dunbar 1, Daniela Kusuke 1, Angela Sicker 1, William G Shadel 1
PMCID: PMC5794530  NIHMSID: NIHMS920956  PMID: 29154602

Abstract

Objectives

Tobacco power walls display hundreds of tobacco products and are known to be key part of the impact of point-of-sale (POS) tobacco advertising on risk for smoking in adolescents. The current study examined factors that are hypothesized to mediate the effect of tobacco power wall exposure on adolescents’ susceptibility for smoking in the future.

Methods

Participants (N = 148) aged 11–17 years were invited to shop in the RAND StoreLab (RSL), a life-sized replica of a convenience store. They were randomized to one of two conditions: one in which the power wall was located in its typical position behind the cashier and the other in which it was hidden behind an opaque wall. Participants shopped in the RSL and then completed measures of susceptibility for smoking in the future, perceived smoking norms, and perceived accessibility of cigarettes. Participants’ movements in the store were electronically tracked.

Results

Having the tobacco power wall behind the cashier increased adolescents’ susceptibility for smoking in the future by 14.3% (total effect) compared to when the power wall was hidden (p = .01), and 14% of this effect was mediated by participants’ perceived smoking norms. Time spent in front of the cashier and perceived accessibility of cigarettes did not play a role in the association between study condition and susceptibility for smoking in the future.

Conclusions

The tobacco power wall increases adolescents’ smoking risk, and this effect is partly explained by the effect of the power wall on adolescents’ perceptions of how normative smoking is.

Keywords: Point-of-sale cigarette advertising, power walls, risk of smoking, mediators

Adolescents have been a favorite target of the tobacco industry for many years (Coughlin & Janecek, 1998). At present, the vast majority of the tobacco industry’s advertising and marketing dollars go toward point-of-sale (POS) promotional efforts (Cohen et al., 2008; Federal Trade Commission, 2015). A key feature of POS environment advertising is the tobacco “power wall” (Dewhirst, 2004). Tobacco power walls display hundreds of different brands of cigarettes and other tobacco products; feature branded posters, product slogans, and price placards; and are prominently placed in retail locations, typically behind the cashier. Exposure to the tobacco-marketing-rich POS retail environment is known to contribute to cigarette smoking in adolescents (Robertson et al., 2015) and susceptibility to smoking has been reported to be a useful construct to identify adolescents at risk of taking up smoking (Choi et al., 2001; Pierce et al., 1996). However, the mechanisms through which exposure to the tobacco power wall increases smoking susceptibility in adolescents are not well understood. The purpose of this study is to evaluate several such mechanisms.

Three classes of variables have been proposed as possible mediators of the tobacco power wall effect on smoking susceptibility: 1) time spent in front of the tobacco power wall (an index of exposure), 2) norms regarding smoking, and 3) the perceived accessibility of cigarettes. First, as noted previously, the tobacco industry pays substantial sums of money to retailers for the prominent placement of their products in stores (Feighery et al., 2003). Locating tobacco power walls strategically behind the cash register presumably increases the likelihood of repeated exposures to the brands of cigarettes and other tobacco products displayed on the power wall. Such repeated and prolonged exposure by itself could have consequences for adolescent tobacco use. The advertising and persuasion literature has documented that mere exposure to a product can increase its appeal, thereby increasing the likelihood of purchasing the product (Baker, 1999; Crano & Prislin, 2006; Zajonc, 1968). Thus, mere exposure is one possible mechanism through which the tobacco power wall could have an effect on adolescents’ smoking risk.

Second, exposure to tobacco power walls at POS locations could act to normalize tobacco use (McNeill et al., 2011), thus causing adolescents to believe that smoking is more prevalent than it actually is (Sargent, 2005). Previous research has shown that adolescents’ perceived descriptive norms (i.e., perceptions about typical patterns of behavior among one’s peers) regarding smoking predict smoking uptake (Olds et al., 2005; Primack et al., 2007; Sussman et al., 1988) and that the association between exposure to tobacco marketing in general and increased smoking risk is partly mediated by perceived smoking norms (Brown & Moodie, 2009). Thus, the tobacco power wall could have its effect on smoking risk by normalizing -or, increasing the perceived prevalence of- smoking.

Third, exposure to POS marketing and the tobacco power wall in particular may foster the perception that tobacco products are easily accessible (Pollay, 2007). Although adolescents are much less likely to actually obtain cigarettes from retail outlets today than at any time in the last several decades (SAMSHA, 2004), nearly 13% of teen smokers say that they usually obtain cigarettes at POS (CDC, 2016). Moreover, retailer density is known to be associated with increased perceptions of cigarette accessibility among adolescents (Lipperman-Kreda, Grube, Friend, & Mair, 2016), as is exposure to tobacco products at POS (Wakefield et al., 2006). As perceived accessibility of cigarettes has been shown to increase the risk of tobacco use among adolescents (Unger et al., 2003), perceived accessibility is another plausible mechanism by which the tobacco power wall may increase smoking risk.

The current study was designed to investigate these three potential mediators of the tobacco power wall effect on adolescents’ susceptibility to smoke. The setting for the study was the RAND StoreLab, a life-sized replica of a convenience store that was designed to evaluate how to best regulate tobacco product advertising at POS during simulated shopping experiences (Shadel et al. 2016). In the current study, we experimentally manipulated the presence or absence of the tobacco power wall to evaluate the effect of this manipulation on middle and high school students’ susceptibility to smoke in the future and test whether exposure, descriptive norms and accessibility of cigarettes act as mediators of any such effect. We hypothesized that the presence or absence of power wall will be associated with the amount of time that adolescents spend at tobacco POS (mere exposure), increase their perceptions of the number of their peers who smoke (descriptive norms) and increase the perceived ease with which they can access cigarettes (perceived accessibility). Furthermore, we hypothesized that power-wall-induced changes in these mediators would, in turn, increase students’ susceptibility to future smoking.

Methods

Participants

Participants in this study were 148 middle and high school students, ages 11–17 years (M=13.7, SD= 2.0), with no physical or psychiatric problem that would interfere with completing the study (based on parent report). Participants were recruited using print, internet, and radio advertisements around the city of Pittsburgh as described elsewhere (Shadel et al. 2016). Recruitment advertisements contained no information about smoking or tobacco in order to reduce potential sample biases. Advertisements indicated that the study was about teens’ purchasing habits at convenience stores. Parents of interested participants telephoned the study center to complete a brief eligibility screening and provided written consent. Adolescents assented to their own participation. Adolescents were enrolled irrespective of their experiences with smoking (8.9% reported smoking at least a puff of a cigarette in the past). Forty-three percent of the sample was male, 55% was Caucasian, 30% African American and 15% some other race. More information about participant characteristics is presented in Table 1.

Table 1.

Participant Baseline Characteristics By Condition

Experimental condition
Hidden Cashier p-value
Age (M, SD) 13.58 (1.97) 13.81 (2.00) 0.48
Gender: % Male 45.95 40.54
Race
 % African American 28.38 31.08 0.65
 % Other 17.57 12.16
 % Caucasian 54.05 56.76
% susceptible to smoking at baseline 16.22 14.86 0.82
Open in a new tab

Procedure

The study was approved by RAND’s Human Subjects Protection Committee. Data collection took place at the RAND StoreLab (RSL), a life-sized replica of a convenience store that occupies 1,500 square feet inside of an office building (the RSL is only open to study participants). Details about the development of the RSL and photographs of its interior appear elsewhere (Shadel et al., 2016). In brief, the RSL stocks over 650 unique products (e.g., dairy products, baked goods, produce, snack foods, beverages, health and beauty aids, confectionery, magazines/newspapers, as well as tobacco products). Industry guidelines dictated the stocking and arrangement of products (National Association of Convenience Stores, 2015). Product prices are consistent with those in the state of Pennsylvania, where the research was conducted. Product posters appear on the walls, shelves, and windows of the store. Posters for tobacco products appear in the windows and doors of the RSL, as well as on the tobacco power wall (the same tobacco posters appeared on the RSL windows and doors, regardless of experimental condition). The RSL was also equipped with an overhead tracking device that records the location and movement of a participant 10 times every second allowing for the assessment of the amount of time spent at specific defined locations in the store.

During the consent process, participants and their parents were told about the general parameters of the study (e.g., that the study involved assessing adolescents’ convenience store shopping habits and involved minimal risk), and that some parts of the study could not be revealed at the outset because that knowledge could affect study outcomes. Participants and their parents were informed that they would be provided with all information about the study at the end. Their consent/assent indicated agreement to participate in the study without full knowledge of the details of the study.

Prior to entering the store to shop, participants completed a questionnaire in which they reported demographic information, their history of smoking and other forms of tobacco use, and their typical convenience store shopping habits. The questionnaire contained a number of filler items that were similar in structure to the smoking/tobacco measures but assessed other unrelated behaviors (e.g., consumption of fruits, soft drinks, and “junk” food). The purpose of these filler items was to disguise the true focus of the study. After completing the questionnaire, participants were randomized to one of two experimental conditions, the cashier condition in which the tobacco power wall appeared in its typical location behind the cashier, and the hidden condition in which the tobacco power wall was hidden from customers’ view. In the hidden condition, tobacco products were situated behind an opaque wall behind the cashier; a sign on the wall read “Tobacco Sold Here.” A third condition, in which the power wall was relocated to a side wall of the RSL away from the cashier, was included to test hypotheses that are not germane to the current study and thus is not discussed further in this article (for more information about that condition and those hypotheses, see Shadel et al, 2016). All participants were provided with $10 and instructed to shop for whatever items they wanted and to take as long as they liked to make a purchase. Participants were told to check-out and pay for the items as they would in any convenience store. Participants’ movements in the RSL were tracked and automatically recorded using an overhead monitoring system. A research assistant acted as the RSL cashier, checking participants out once they finished shopping, providing change, and bagging items that were purchased. After exiting the RSL, participants completed a questionnaire that assessed two of the three hypothesized mediating variables (perceived smoking norms and perceived cigarette accessibility) as well as the dependent measure, susceptibility to future cigarette smoking. Filler items were also included in this post-shopping questionnaire. Upon completing the questionnaire, participants were debriefed, shown a 20-minute video about cigarette advertising and media literacy, and given a $50 gift card for completing the study.

Measures

Amount of time spent in front of the cashier (i.e., mere exposure)

The RSL was equipped with overhead, fixed-dome cameras (IP cameras with PoE, 3-megapixel resolution and a viewing angle of 134°) that recorded the study participant’s location in the store every tenth of a second. The cameras were installed by RetailNext, Inc. (http://retailnext.net) a company that developed a technology platform to track shoppers’ movements in retail stores. Nine cameras were installed throughout the ceiling of the RSL to fully capture the movements of study participants as they shopped. Using a map of the store, we defined a 200 square-foot (10 feet by 20 feet) zone in front of the cashier as an area from which a participant could clearly observe the power wall if it were present. For all participants, we used the tracking data to compute the time (in seconds) spent in that area and used that value as an index of the amount of time a participant could potentially have been exposed to the power wall (if present).

Perceived descriptive smoking norms

After shopping in the RSL, participants indicated their descriptive smoking norms by answering the following question: “What percentage of kids in your grade smoke cigarettes? (see Unger et al., 2001). ).

Perceived accessibility of cigarettes

After shopping in the RSL, participants indicated their perceptions about the accessibility of cigarettes by completing an item adapted from one contained in the Monitoring the Future (MTF) survey: “How difficult do you think it would be for you to get cigarettes, if you wanted some (1=Probably impossible, 2=Very difficult, 3=Fairly difficult, 4=Fairly easy, 5=Very easy)?”

Susceptibility to future smoking

In both the pre- and post-shopping questionnaire, participants indicated their susceptibility to future cigarette smoking by completing a 3-item scale that was adapted from one developed by Choi et al (2001) and has been shown to predict smoking: “Do you think you will try a cigarette anytime soon?”, “Do you think you will smoke a cigarette anytime in the next year?”; and “If one of your best friends offered you a cigarette, would you smoke it?” Responses to these three items were made on a 1 (Definitely Not) to 10 (Definitely Yes) scale and summed (α = 0.94) to produce a susceptibility score, with higher scores indicative of greater susceptibility. Because the distribution of this measure was skewed (over 75% of participants had the lowest score possible on the scale, i.e., a “3”), scores on this scale were dichotomized: scores of “3” were recoded as “0” (no susceptibility) and scores greater than “3” were coded as “1” (some susceptibility). This scoring approach is a conventional one (Choi et al., 2001; Shadel et al., 2016).

Statistical Analysis

We conducted a series of mediation analyses (Imai et al, 2010; see also Baron and Kenny, 1986) to examine the role of perceived smoking norms, perceived ease of accessibility of cigarettes, and the amount of time spent in front of a power wall in mediating the impact of power wall exposure on adolescent smoking susceptibility (see Figure 1).

Figure 1.

Figure 1

Open in a new tab

Mediation analysis model. The letters λ, α and β represent linkages among variables

Note: * significant at 0.05 significance level

For each of the three hypothesized mediator variables (M), we used regression models to test whether the mediators were affected by the presence or absence of the power wall (T) and also influenced the post-shopping smoking susceptibility outcome (Y) in a way that explained in whole or in part the effect of study condition. First, we regressed the power wall condition and each mediator on the outcome, Y, controlling for covariates summarized in X (age, gender, race, and baseline smoking susceptibility):

g(Y)=μ+λCondition+βMediator+γ1X+εy (1)

where g() is the non-linear logistic regression transformation used because of the binary outcome. Then, we regressed study condition on each mediator, M, again controlling for the covariates X:

Mediator=ν+αCondition+γ2X+εm (2)

The covariates that we controlled for included participants’ age, gender, race and participants’ baseline smoking susceptibility. In cases in which the outcome is continuous (Baron and Kenny, 1986), the typical mediation analysis estimates λ as the direct effect of the independent variable, αβ as the effect mediated, and λ + αβ as the total effect. The percentage of mediated effect through M is computed as 100%[αβ/(λ + αβ)]. In cases in which the outcome is binary, as in this study, estimation of these parameters is achieved using the counterfactual framework of causal mediation effects proposed by Imai et al. (2010).

Statistical analyses were conducted using R software (version 3.3.0). In particular, we used the R package for Causal Mediation Analysis (Tingley et al. 2014) along with a non-parametric bootstrap procedure with 10,000 replications to determine statistical significance.

Results

As Table 1 shows, participants in the two study conditions did not differ in their demographic characteristics or baseline susceptibility to smoke. Bivariate comparisons revealed that two of the three hypothesized mediators were significantly impacted by experimental condition. Participants in the cashier condition reported higher perceived smoking norms than participants in the hidden condition (M’s = 22.96 and 15.04, respectively; p = 0.05) and also spent more time in front of the cashier (M’s = 71.72 seconds and 55.3 seconds, respectively; p= 0.01). Perceived accessibility of cigarettes did not differ between conditions (p = 0.64).

Controlling for each of the hypothesized mediators, the logistic regression of the impact of study condition on future smoking susceptibility revealed that being in the power wall condition led to an increase in participants’ future smoking susceptibility relative to the hidden condition (Table 2). Results of the effect decomposition analysis showed that the total power wall effect increased future smoking susceptibility by 13 to 14% (depending on the mediator model), and that a significant portion of that effect (2.1%, CI=[0.1% −5.4%]; or, 14% of the total effect) was mediated by the increase in participants’ perceived smoking norms. The power wall effect on future smoking susceptibility was not mediated by the amount of time spent in front of the cashier. Perceived cigarette accessibility was not influenced by exposure to the power wall and consequently it did not mediate any of the total effect of condition. Figure 1 presents estimates from separate models evaluating each potential mediating variable. Because participants’ smoking behavior at baseline is correlated with smoking susceptibility, it also could influence the impact of power wall exposure. As such, we conducted a sensitivity analysis controlling for baseline smoking status; results were unchanged. An additional sensitivity analysis that removed participants that were susceptible to smoking at baseline yielded similar results.

Table 2.

Results of Logistic Regression Outcome Models and Linear Mediation Models Testing Hypothesized Mediators of the Relationship between Exposure to the Tobacco Power Wall and Future Smoking Susceptibility

Predictor Outcome model (logistic) Mediator model (linear)
Estimate SE Wald χ2 p Estimate SE t p
Hypothesized mediator: Perceived norms
Hypothesized mediator: Norm 0.03 0.01 2.72 0.01
Cashier condition 1.45 0.69 2.12 0.03 7.04 3.44 2.05 0.04
Race: Other −1.84 1.08 −1.71 0.09 6.90 5.06 1.36 0.18
 African American −0.42 0.67 −0.62 0.54 12.03 3.95 3.05 0.00
Male 0.40 0.57 0.70 0.49 −0.82 3.49 −0.24 0.81
Age −0.13 0.17 −0.76 0.45 6.39 0.90 7.11 0.00
Baseline smoking susceptibility 5.20 1.06 4.93 0.00 3.50 5.00 0.70 0.49
Hypothesized mediator: Perceived accessibility
Hypothesized mediator: Access 0.47 0.21 2.21 0.03
Cashier condition 1.47 0.61 2.39 0.02 −0.14 0.21 −0.67 0.50
Race: Other −1.24 0.90 −1.37 0.17 0.07 0.30 0.25 0.80
 African American −0.27 0.62 −0.43 0.67 −0.06 0.24 −0.27 0.79
Male 0.11 0.54 0.21 0.84 0.39 0.21 1.84 0.07
Age −0.04 0.14 −0.28 0.78 0.25 0.05 4.67 0.00
Baseline smoking susceptibility 3.85 0.81 4.76 0.00 1.04 0.30 3.52 0.00
Hypothesized mediator: Time spent at the cashier
Hypothesized mediator: Time at cashier 0.01 0.01 0.81 0.42
Cashier condition 1.23 0.61 2.01 0.04 17.46 6.24 2.80 0.01
Race: Other −1.18 0.88 −1.35 0.18 25.24 9.10 2.77 0.01
 African American −0.33 0.62 −0.53 0.59 11.70 7.12 1.64 0.10
Male 0.23 0.53 0.43 0.67 −1.40 6.32 −0.22 0.82
Age 0.09 0.13 0.65 0.52 0.31 1.63 0.19 0.85
Baseline smoking susceptibility 4.14 0.80 5.15 0.00 10.27 8.99 1.14 0.25
Open in a new tab

Discussion

This study provides evidence that the tobacco power wall has its effects on adolescent smoking risk by increasing the perceived prevalence of smoking. That is, the tobacco power wall causes adolescents to believe that cigarette smoking is more normative, thereby increasing the chance that they will smoke in the future. Previous research has demonstrated that perceived smoking norms are related to smoking uptake among adolescents (Primack et al., 2007; Sussman et al., 1988; Olds et al., 2005) and suggested that awareness of tobacco marketing increases smoking risk by influencing perceived smoking norms (Brown & Moodie, 2009). Our study adds to this literature by demonstrating that exposure to the tobacco power wall influences smoking susceptibility through its influence on perceived smoking norms.

Time spent in front of the cashier was influenced by the presence of the tobacco power wall but it did not mediate the power wall effect on future smoking susceptibility. This finding is surprising given that studies have shown that increased exposure to cigarette advertising is associated with increased levels of adolescent smoking (Wellman et al., 2006) and that theory and research support the idea that mere exposure to a product can influence consumer behavior (Baker, 1999; Crano & Prislin, 2006; Zajonc, 1968). The average difference in time spent in front of cashier between conditions was only about 17 seconds, which might not have been large enough to induce changes in smoking susceptibility. Alternatively, adolescents in our sample had presumably experienced hundreds of exposures prior to enrolling in the study; it is possible that mere exposure effects are more likely to occur earlier in an individual’s life when the exposure is more novel.

The final mediator tested, perceived accessibility of cigarettes, was not associated with exposure to the power wall in our study. This finding conflicts with previous experimental research that has shown that adolescents exposed to POS cigarette advertising perceived cigarettes to be more easily accessible (e.g. Henriksen et al., 2002; Wakefield et al., 2006). This could be because in both conditions, the cash register had a sign stating that tobacco is not sold to shoppers under the age of 18 and that the cashier would card everyone who attempted to buy tobacco products. Nonetheless, independent of experimental condition, perceived accessibility was associated with increased susceptibility to smoke in the future, consistent with what has been found in other studies (Jackson et. al. 1997).

Our study has several limitations that should be considered when interpreting its results. First, the sample was recruited using media advertising, and boys were underrepresented in our sample compared with the youth population of the United States. This feature of our study may constrain the generalizability of our results. Second, although our dependent measure, susceptibility to future cigarette smoking, has been shown to be a potent predictor of smoking in adolescents in several studies (Choi et al., 2001), we did not measure actual smoking behavior in this experiment. Third, we were unable to control for the economic status of participants. Fourth, our use of single-item measures to quantify perceived smoking norms and accessibility to cigarettes may have affected our findings. Although we selected these measures to minimize response burden, single-item measures may be vulnerable to random measurement error and bias if their interpretation is unclear to participants. Finally, the environment of the RSL, though closely modeled after a real convenience store, is still an artificial one, and thus this laboratory-based experimental investigation would not necessarily provide a good environment to model the entire process of how POS influence adolescent smoking.

In summary, the current study provides evidence that the impact of tobacco power walls on adolescent smoking is mediated through perceived smoking norms. This finding implies that one way to reduce the potency of tobacco power walls could be through counter- marketing messages or media literacy interventions that provide corrective normative information (see Pinkleton, Austin, Cohen, Miller & Fitzgerald, 2007). Future research could evaluate whether such a counter-marketing approach is effective at diminishing the impact of the tobacco power wall on adolescents. Future research could also investigate whether the associations found in this study hold for the marketing of other tobacco products (e.g., electronic cigarettes, snus).

Table 3.

Causal Effect Decomposition of the Relationship between Exposure to the Power Wall and Future Smoking Susceptibility

Effect type Effect Decomposition
Effect Low CI Up CI p
Hypothesized mediator: Perceived norms
Total effect 14.3% 3.4% 25.2% 0.01
Average causal mediated effect 2.1% 0.1% 5.4% 0.04
Average Causal Direct Effect 12.1% 1.0% 22.9% 0.03
Hypothesized mediator: Perceived accessibility
Total Effect 13.2% 2.0% 24.8% 0.02
Average Causal Mediated Effect −0.7% −3.2% 1.3% 0.53
Average Causal Direct Effect 13.8% 3.0% 25.1% 0.02
Hypothesized mediator: Time spent at the cashier
Total effect 12.8% 1.5% 24.1% 0.03
Average causal mediated Effect 0.9% −1.5% 3.8% 0.43
Average causal direct effect 11.9% 0.4% 23.5% 0.04
Open in a new tab

Acknowledgments

Funding Source: The work presented in this paper was supported by the National Cancer Institute and FDA Center for Tobacco Products (CTP) (R01CA175209). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration.

Footnotes

Financial Disclosure: None of the authors have any financial relationships relevant to this article to disclose.

Conflict of Interest: None of the authors have conflicts of interest to disclose.

References

  1. Baker W. When can affective conditioning and mere exposure directly influence brand choice? Journal of Advertising. 1999;28:31–46. [Google Scholar]
  2. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology. 1986;51:1173–1182. doi: 10.1037//0022-3514.51.6.1173. [DOI] [PubMed] [Google Scholar]
  3. Brown A, Moodie C. The influence of tobacco marketing on adolescent smoking intentions via normative beliefs. Health Education Research. 2009;24(4):721–733. doi: 10.1093/her/cyp007. [DOI] [PubMed] [Google Scholar]
  4. Center for Disease Contorl, CDC. Youth Risk Behavior Surveillance—United States, 2015. MMWR. 2016;65(6) doi: 10.15585/mmwr.ss6506a1. [DOI] [PubMed] [Google Scholar]
  5. Cohen J, Planinac LC, Griffin K, Robinson DJ, O’Conner SC, Lavack A, Thompson FE, Di Nardo J. Tobacco promotions at point-of-sale: The last hurrah. Canadian Journal of Public Health. 2008;99:166–171. doi: 10.1007/BF03405466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Choi WS, Gilpin EA, Farkas AJ, Pierce JP. Determining the probability of future smoking among adolescents. Addiction. 2001;96:313–323. doi: 10.1046/j.1360-0443.2001.96231315.x. [DOI] [PubMed] [Google Scholar]
  7. Coughlin P, Janecek F., Jr . A review of R.J. Reynolds’ internal documents produced in Mangini Vs. R.J. Reynolds tobacco company, civil number 939359 – The case that rid California and the American landscape of “Joe Camel”. San Francisco: Milberg, Weiss, Bershad, Hynes & Lerach LLP; 1998. [Google Scholar]
  8. Crano W, Prislin R. Attitudes and persuasion. Annual Review of Psychology. 2006;57:345–374. doi: 10.1146/annurev.psych.57.102904.190034. [DOI] [PubMed] [Google Scholar]
  9. Dewhirst T. POP goes the power wall? Taking aim at tobacco promotional strategies utilised at retail. Tobacco Control. 2004;13:209–210. doi: 10.1136/tc.2004.009043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Federal Trade Commission. Cigarette report for 2012. Washington, DC: Federal Trade Commission; 2015. [Google Scholar]
  11. Feighery EC, Ribisl KM, Clark PI, Haladjian HH. How tobacco companies ensure prime placement of their advertising and products in stores: interviews with retailers about tobacco company incentive programmes. Tobacco Control. 2003;12:184–188. doi: 10.1136/tc.12.2.184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Henriksen L, Flora JA, Feighery E, Fortmann SP. Effects on youth of exposure to retail tobacco advertising. Journal of Applied Social Psychology. 2002;32:1771–1789. [Google Scholar]
  13. Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychological Methods. 2010;15(4):309–334. doi: 10.1037/a0020761. [DOI] [PubMed] [Google Scholar]
  14. Jackson C, Henriksen L, Dickonson D, Levine DW. The early use of alcohol and tobacco: its relation to children’s competence and parent’s behavior. American Journal of Public Health. 1997;87:359–64. doi: 10.2105/ajph.87.3.359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lipperman-Kreda S, Grube JW, Friend KB, Mair C. Tobacco outlet density, retailer cigarette sales without ID checks and enforcement of underage tobacco laws: Associations with youths’ cigarette smoking and beliefs. Addiction. 2016;11:525–532. doi: 10.1111/add.13179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. McNeill A, Lewis S, Quinn C, Mulcahy M, Clancy L, Hastings G, et al. Evaluation of the removal of point of sale tobacco promotional displays in Ireland. Tobacco Control. 2011;20:137–143. doi: 10.1136/tc.2010.038141. [DOI] [PubMed] [Google Scholar]
  17. National Association of Convenience Stores (US) What is a convenience store? 2015 Retrieved from http://www.nacsonline.com/Research/Pages/What-is-a-Convenience-Store.aspx.
  18. Olds RS, Thombs DL, Tomasek JR. Relations between normative beliefs and initiation intentions toward cigarette, alcohol and marijuana. Journal of Adolescent Health. 2005;37(1):75. doi: 10.1016/j.jadohealth.2004.09.020. [DOI] [PubMed] [Google Scholar]
  19. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Merritt RK. Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychology. 1996;15:355–361. doi: 10.1037//0278-6133.15.5.355. [DOI] [PubMed] [Google Scholar]
  20. Pinkleton BE, Austin EW, Cohen M, Miller A, Fitzgerald E. A statewide evaluation of the effectiveness of media literacy training to prevent tobacco use among adolescents. Health Communication. 2007;21:23–34. doi: 10.1080/10410230701283306. [DOI] [PubMed] [Google Scholar]
  21. Pollay RW. More than meets the eye: on the importance of retail cigarette merchandising. Tobacco Control. 2007;16:270–274. doi: 10.1136/tc.2006.018978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Primack BA, Switzer GE, Dalton MA. Improving measurement of normative beliefs involving smoking among adolescents. Archives of Pediatrics and Adolescent Medicine. 2007;161(5):434–439. doi: 10.1001/archpedi.161.5.434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Robertson L, McGee R, Marsh L, Hoek L. A systematic review on the impact of point-of-sale tobacco promotion on smoking. Nicotine & Tobacco Research. 2015;17:2–17. doi: 10.1093/ntr/ntu168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Sargent JD. Smoking in movies: impact on adolescent smoking. Adolescent Medicine Clinics. 2005;16(2):345–370. doi: 10.1016/j.admecli.2005.02.003. [DOI] [PubMed] [Google Scholar]
  25. Shadel WG, Martino SC, Setodji CM, Scharf DM, Kusuke D, Sicker A, Gong M. Hiding the tobacco power wall reduces cigarette smoking risk in adolescents: Using an experimental convenience store to assess tobacco regulatory options at retail point-of-sale. Tobacco Control. 2016 doi: 10.1136/tobaccocontrol-2015-052529. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Substance Abuse & Mental Health Services Administration, SAMSHA. 2003 National Survey on Drug Use and Health. U.S. Department of Health & Human Services; 2004. [Google Scholar]
  27. Sussman S, Dent CW, Mestel-Rauch J, Johnson CA. Adolescent nonsmokers, triers, and regular smokers’ estimates of cigarette smoking prevalence: when do overestimations occur and by whom? Journal of Applied Social Psychology. 1988;18:537–551. [Google Scholar]
  28. Tingley D, Yamamoto T, Hirose K, Keele L, Imai K. Mediation: R package for causal mediation analysis. Journal of Statistical Software. 2014;59(5) doi: 10.18637/jss.v059.i05. [DOI] [Google Scholar]
  29. Unger JB, Rohrbach LA, Cruz TB, Baezconde-Garbanati L, Howard KA, Palmer PH. Ethnic variation in peer influences on adolescent smoking. Nicotine and Tobacco Research. 2001;3:167–176. doi: 10.1080/14622200110043086. [DOI] [PubMed] [Google Scholar]
  30. Unger JB, Shakib S, Cruz TB, Hoffman BR, Pitney BH, Rohrbach LA. Smoking behavior among urban and rural Native American adolescents in California. American Journal of Preventive Medicine. 2003;25(3):251–254. doi: 10.1016/s0749-3797(03)00193-4. [DOI] [PubMed] [Google Scholar]
  31. Wakefield M, Germain D, Durkin S, Henriksen L. An experimental study of effects on schoolchildren of exposure to point-of-sale cigarette advertising and pack displays. Health Education Research. 2006;21(3):338–347. doi: 10.1093/her/cyl005. [DOI] [PubMed] [Google Scholar]
  32. Wellman RJ, Sugarman DB, DiFranza JR, Winickoff JP. The extent to which tobacco marketing and tobacco use in films contribute to children’s use of tobacco. Archives of Pediatric & Adolescent Medicine. 2006;160:1285–1296. doi: 10.1001/archpedi.160.12.1285. [DOI] [PubMed] [Google Scholar]
  33. Zajonc Robert B. Attitudinal effects of mere exposure. Journal of Personality and Social Psychology. 1968;9(2 Part 2):1–27. [Google Scholar]

RESOURCES