نظریه خودتأییدی و تصاویر هشدار دهنده سیگار کشیدن

Abstract Introduction The present study examined self-affirmation theory, cigarette smoking, and health-related images depicting adverse effects of smoking. Previous research examining self-affirmation and negative health-related images has shown that individuals who engage in a self-affirmation activity are more receptive to messages when compared to those who do not affirm. We were interested in examining the extent to which self-affirmation would reduce defensive responding to negative health images related to cigarette smoking. Methods Participants included 203 daily smokers who were undergraduate students at a large southern university. Participants completed a battery of questionnaires and were then randomly assigned to one of four conditions (non-smoking image control, smoking image control, low affirmation, and high affirmation). Analyses evaluated the effectiveness of affirmation condition as it related to defensive responding. Results Results indicated that both affirmation conditions were effective in reducing defensive responding for those at greatest risk (heavier smokers) and those more resistant to health benefits associated with quitting. Conclusions Findings are discussed in terms of potential public health implications as well as the role defensive responding plays in the evaluation and processing of negative health messages.

3. Results 3.1. Evaluation of overall defensive responding to smoking images Preliminary analyses using a one-way between subjects ANOVA with post hoc comparisons utilizing the Tukey HSD test revealed no baseline differences with respect to age, number of cigarettes smoked, nicotine dependence, perceived health benefits, or quit intentions across the four groups. Descriptive statistics for the sample are presented in Table 1. Table 1. Means and standard deviations for the total sample and each group at baseline. Group Variable Total sample M (SD) High affirmation group M (SD) Low affirmation group M (SD) Smoking image control group M (SD) Non-smoking image control group M (SD) Age 23.07 (4.99) 22.63 (3.55) 22.44 (5.04) 23.07 (4.92) 23.17 (6.16) Number of cigarettes smoked per day 11.55 (9.25) 11.41 (10.27) 11.58 (9.72) 11.24 (8.46) 11.61 (8.65) Nicotine dependence 3.82 (1.79) 3.76 (1.91) 3.63 (1.82) 3.98 (1.62) 3.84 (1.72) Perceived health benefits 5.22 (1.74) 5.13 (1.66) 5.26 (1.91) 5.29 (1.74) 5.20 (1.67) Quit intentions 4.35 (1.60) 4.64 (1.50) 4.17 (1.55) 4.26 (1.74) N/A Note: A one-way between subjects ANOVA, using post hoc Tukey corrections, revealed no significant differences between the groups on any of the study variables. Table options A one-way between subjects ANOVA was conducted to examine the differences in overall threat when compared to ratings of the non-smoking related control images. Threat and unpleasantness were used to create a composite threat score. Results revealed that there was an overall significant effect of the ratings for the three conditions viewing the smoking images when compared to ratings of the control images, F(3, 199) = 59.34, p < 0.001. Post hoc comparisons using the Tukey HSD test indicated that mean scores for the high affirmation condition (M = 3.37, SD = 1.39), the low affirmation condition (M = 3.62, SD = 1.36), and the smoking control image condition (M = 3.81, SD = 1.59) were significantly different than that of the non-smoking related control image condition (M = 1.54, SD = 0.86). Note, the group evaluating control images were not included in the primary analyses, thus the primary analyses were based on a sample of 153. Overall, these results suggest that smokers in all conditions rated the health-related images as more threatening than the control images. 3.2. Affirmation group and threat A one-way ANOVA was used to test whether there were differences in the mean threat ratings of the smoking related images based on group. Overall, there were no significant differences across the three experimental groups that evaluated the smoking-related images, F(2, 152) = 1.13, p = .325. Thus, an overall effect for group was not supported. 3.3. Affirmation group, number of cigarettes smoked, threat (H1a) To test the hypothesis that the number of cigarettes smoked would moderate the association between group (high affirmation, low affirmation, and smoking image control) and threat ratings, a hierarchical multiple regression analysis was conducted and all variables were mean centered. Condition was operationalized by two dummy coded variables representing high affirmation and low affirmation versus the control condition. Thus, threat was examined as a function of condition, smoking, and two 2-way product terms representing high affirmation versus control × smoking and low affirmation versus control × smoking. The results revealed significant interactions between high affirmation versus control, t(147) = 2.98, p = .003, and between low affirmation versus control, t(147) = 2.56, p = .011. Tests of simple slopes were conducted using procedures described by Aiken and colleagues ( Aiken and West, 1991 and Cohen et al., 2003). High and low smoking values were specified as one standard deviation above the mean and one standard deviation below the mean, respectively. Fig. 1 presents graphs of these interactions as well as standard errors. Ratings for overall threat among lighter smokers did not differ from the control (M = 2.73) in neither high (M = 3.08) nor low (M = 3.22) affirmation groups, t(147) = − .94, p = .34; and, t(147) = − 1.29, p = .20, respectively. In contrast, significant differences were found among heavier smokers in threat ratings between the high affirmation group (M = 3.66) and the control group (M = 4.92), t(147) = 3.32, p < .001, and between the low affirmation group (M = 3.99) and the control group, t(147) = 2.43, p = .02. These results suggest that being affirmed decreased threatening ratings of the smoking images for those who smoked more. Moreover, heavier smokers in the control group reported the highest (greatest resistance) image ratings. Number of cigarettes smoked as a moderator of threat. Note: Error bars reflect ... Fig. 1. Number of cigarettes smoked as a moderator of threat. Note: Error bars reflect standard errors. † indicates differences between the high affirmation and control group, ** indicates differences between the low affirmation and control group. Figure options 3.4. Affirmation group, perceived health benefits, threat (H2a) Parallel analyses were conducted to evaluate whether affirmation effects varied as a function of perceived health benefits. Threat was examined as a function of condition, perceived health benefits, and two 2-way product terms representing high affirmation versus control × perceived health benefits and low affirmation versus control × perceived health benefits. The results revealed that the interactions between high affirmation versus control, t(147) = − 2.10, p = .04, and low affirmation versus control, t(147) = − 2.20, p = .03, were significant. Fig. 2 presents graphs for this interaction as well as standard errors. Significant differences in threat ratings were found among those with lower perceived health benefits between the high affirmation (M = 3.73) and control groups (M = 4.82), t(147) = 2.65, p = .0037, as well as the low affirmation (M = 4.05) and control groups, t(147) = 2.13, p = .03. In contrast, for those who believed there were more health benefits associated with quitting smoking, there was no difference in threat ratings between the high affirmation group (M = 2.9) and the control group (M = 2.86), t(147) = − .10, p = .91. There was also no difference among those higher in perceived benefits in the ratings between the low affirmation group (M = 3.1) and the control group, t(147) = − .95, p = .34. Taken together, these results suggest that being affirmed lowered threat ratings of the smoking images for those who believed that there were fewer health benefits associated with smoking cessation. Further, consistent with the findings above, participants low in perceived health benefits in the control group reported the highest (most threatening) overall image ratings. Perceived health benefits as a moderator of threat. Note: Error bars reflect ... Fig. 2. Perceived health benefits as a moderator of threat. Note: Error bars reflect standard errors. † indicates differences between the high affirmation and control group, ** indicates differences between the low affirmation and control group. Figure options 3.5. Affirmation group and quit intentions A one-way ANOVA was used to test the next set of hypotheses identifying whether there were differences in the mean ratings of quit intentions based on group. Overall, there were no significant differences across the three experimental groups, F(2, 152) = 1.27, p = .28. Thus, an overall effect for group was not supported. 3.6. Affirmation group, number of cigarettes smoked, quit intentions (H1b) Next, we examined whether affirmation effects on quit intentions varied as a function of smoking levels. We expected affirmation effects to be more evident among heavier smokers. Quit intention was thus examined as a function of condition, smoking, and two 2-way product terms representing high affirmation vs. control × smoking and low affirmation vs. control × smoking. The results indicated that the interactions between high affirmation vs. control, t(147) = 2.70, p = .007, and low affirmation vs. control, t(147) = 2.13, p = .035, were significant. Fig. 3 presents graphs for these interactions as well as standard errors. No differences emerged in quit intentions for lighter smokers between the high affirmation (M = 4.96) control groups (M = 5.37) nor between the low affirmation (M = 4.65) and control groups (t(147) = − .97, p = .33; and, t(147) = − 1.68, p = .09, respectively). In contrast, there was a significant difference in quit intentions among heavier smokers between the high affirmation group (M = 4.33) and the control group (M = 3.1), t(147) = 2.90, p = .004. There was no significant difference among heavier smokers in reported quit intentions between the low affirmation group (M = 3.7) and the control group, t(147) = 1.41, p = .16. The results suggest that being highly affirmed increased intentions to quit for heavier smokers. Number of cigarettes smoked as a moderator of quit intentions. Note: Error bars ... Fig. 3. Number of cigarettes smoked as a moderator of quit intentions. Note: Error bars reflect standard errors. † indicates differences between the high affirmation and control group. Figure options 3.7. Affirmation group, perceived health benefits, quit intentions (H2b) Finally, analyses were conducted to evaluate whether affirmation might vary as a function of perceived health benefits. Quit intention was examined as a function of condition, perceived health benefits, and two 2-way product terms representing high affirmation vs. control × perceived health benefits and low affirmation vs. control × perceived health benefits. The results revealed that the interactions were significant between both high affirmation vs. control t(147) = − 2.10, p = .04, and low affirmation vs. control t(147) = − 2.42, p = .02. Fig. 4 presents graphs for these interactions as well as standard errors. For participants with lower perceived health benefits, there was a significant difference in self-reported quit intentions between the high affirmation group (M = 4.24) and the control group (M = 3.12), t(147) = 2.72, p < .01. There was no difference among participants with lower perceived health benefits in quit intentions between the low affirmation group (M = 3.74) and the control group, t(147) = 1.54, p = .12. Moreover, for participants who believed there were greater health benefits associated with quitting smoking, no differences were found in quit intentions between the high affirmation (M = 5.18) and control groups (M = 5.31), nor between the low affirmation (M = 4.59) and control groups (t(147) = − .32, p = .75; and, t(147) = − 1.86, p = .07, respectively). Taken together, these results suggest that being highly affirmed increased quit intentions for those who believed that there were fewer health benefits associated with smoking cessation. Perceived health benefits as a moderator of quit intentions. Note: Error bars ... Fig. 4. Perceived health benefits as a moderator of quit intentions. Note: Error bars reflect standard errors. † indicates differences between the high affirmation and control group.