Social influences on smoking cessation: a comparison of the effect of six social influence variables Bas van den Putte, Ph.D. * , Marco C. Yzer, Ph.D., Suzanne Brunsting, M.A. Amsterdam School of Communication Research ASCoR, Department of Communication, University of Amsterdam, Kloveniersburgwal 48, 1012 CX, Amsterdam, The Netherlands Available online 9 December 2004 Abstract Background. This study brings together the fragmentary knowledge on social influence factors related to smoking cessation. An inventory of social influence variables shows six social influence factors. With the exception of descriptive norm, most of these are rarely studied in the context of smoking cessation. Methods . Regression analysis on the data of 2895 smokers was used to estimate the relative importance of social influence variables as predictors of intention. The moderating effect of quit history was studied by adding interaction terms to the regression analysis. Results . The regression analysis shows that subjective norm and injunctive norm, that is, the social norms on what ought to be done, are more important than descriptive norms, that is, the perceived smoking and smoking cessation behavior of others. This holds especially for smokers whose past cessation attempts quickly failed. Most smokers think that it is acceptable to smoke in most social situations, but simultaneously think that other people approve it if they quit smoking. Conclusions . Results suggest that health campaigns should incite social interaction to increase smokers’ awareness of social norms on the proper behavior. Also, studies into smoking cessation should take account of the various social influence factors outlined in this study. D 2004 Elsevier Inc. All rights reserved. Keywords: Smoking cessation; Health behavior; Social influences; Interpersonal relations Introduction In the past, relatively little research attention has been given to social influences on smoking cessation. This is remarkable because it has extensively been shown that the social environment is an important determinant of the onset of smoking [1,2]. Also, reviews of smoking prevention programs showed that interventions that con- centrated on social norms or social resistance skills were relatively more successful than other smoking prevention programs [3,4]. Recently, the social norms approach that aims to correct misperceptions of the social norm, by stressing that most people do not smoke, has gained some popularity [5,6]. An explanation for the scarcity of research into social influence factors on smoking cessation might be that many researchers in this field used social–psychological models that do not explicitly take social influence variables into account, for example, the trans-theoretical model [7], health belief model [8], or protection motivation theory [9]. An exception is the theory of planned behavior [10], which incorporates the subjective norm, but applications of this theory to smoking cessation are scarce. Studies that examined smoking cessation outside the context of these behavioral models used a wide variety of social influence variables [11–13]. Except for the proportion of smokers in the social environment, the effect of most social influence variables on smoking cessation was rarely empirically tested. In addition to differences in the conceptualization of social influence, previous research used diverse study samples such as patients of a medical center [14], visitors to a health promotion clinic [10], low SES African American 0091-7435/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2004.09.040 * Corresponding author. Fax: +31 20 5253681. E-mail address: s.j.h.m.vandenputte@uva.nl (B. van den Putte). Preventive Medicine 41 (2005) 186 – 193 www.elsevier.com/locate/ypmed