Smoking cross-culturally: Risk perceptions among young adults in Denmark and the United States

Abstract

Research examining smokers’ understanding of their smoking risk reveals that smokers acknowledge some risk but often deny or minimize personal risk. We examined risk perceptions of lung cancer among smokers and non-smokers in a smoking-lenient (Denmark) and a smoking-prohibitive (the United States) culture. Participants were 275 Danish students attending trade schools (mean age 22.6 years) and 297 US students attending community colleges in Florida (mean age 23.6 years). Results revealed cross-cultural differences suggesting that Danish smokers showed greater risk minimization than US smokers. In addition, in both countries the risk of a typical smoker was rated as lower by smokers than non-smokers, and smokers rated their personal risk as lower than they rated the risk of the typical smoker. Cross-cultural differences in moralization of smoking might be one explanation for these findings.

Keywords:

• risk perception
• smoking
• cross-cultura

Acknowledgements

We would like to thank James Shepperd and Jostein Rise for comments on an earlier draft of this manuscript.

Notes

Notes

1. There are no theoretical reasons for expecting gender differences in smoking risk perceptions; in fact the literature on risk perception rarely shows gender differences. However, because there were relatively more men in the Danish sample (70%) than in the US sample (46%), we examined the possibility that gender interacted with country of origin for each of the three primary dependent variables: personal risk, non-smoker risk, and smoker risk. A 2 (gender) ×2 (country) ANOVA was conducted for each of the three dependent variables. Results revealed no main effects of gender or country × gender interactions (ps > 0.16). Therefore, the analyses did not further examine gender.

2. Statistical adjustments are sometimes recommended when a large number of outcome variables are tested. Such adjustments were not appropriate here for two reasons. First, we tested a small number of a priori hypotheses–hypotheses which were based on a large literature of similar research. Second, researchers argue that Bonferroni adjustments should not be used because they increase the likelihood of type 2 error, they test the null hypothesis which is usually irrelevant to the researcher, and it is illogical to make statistical adjustments based on the number of tests performed (Perneger, 1998). Rothman (1990) and Feise (2002) also argue that it is the best policy to not use Bonferroni adjustments when making multiple comparisons.