Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




S. Martin Taylor


The Community Intervention Trial for Smoking Cessation (COMMIT) was the largest community-based effort to date aimed specifically at the reduction of smoking. COMMIT involved 11 matched pairs of intervention and comparison communities from geographically diverse regions of North America. One community from each of the matched pairs was randomly allocated to receive a standardized intervention protocol implemented between 1989 and 1993. The main trial goal was to achieve higher quit rates amongst heavy smokers in the intervention communities compared to the comparison communities.

Intermediate trial goals were to increase the priority of smoking as a public health problem and increase social norms that support nonsmoking in the intervention communities. These goals were informed by contemporary health promotion theory which suggests that longstanding changes in behaviour are best achieved by altering social environments to support healthy choices. This thesis presents a conceptualization of the links between attributes of individuals, communities and attitudes toward smoking which provides the basis for empirical investigations of: COMMIT's success in promoting attitude change; the nature of the relationship between attitude and behaviour change for smoking; and the role of community context in constructing attitudes.

Five research objectives are addressed using several analytic and descriptive techniques: 1) to develop indices to measure smoking attitudes and attitude change; 2) to document changes in smoking attitudes over the course of the COMMIT trial; 3) to determine covariates of individual attitude change; 4) to examine the nature of the relationship between attitudes toward smoking and smoking behaviour; and 5) to describe the role of community context in shaping attitudes and to begin to account for the geographic variability in attitudes in COMMIT communities.

Attitude change was assessed primarily from prevalence surveys administered in 1989 (n=9,875) and 1993 (n=14,117). Data from cohorts of smokers and nonsmokers (n=5,450) were also used to examine covariates of attitude change and for the analysis of the attitude change - behaviour change relationship. Community profile data and legislative tracking were employed in the descriptive analyses of the relationship between community context and smoking attitudes.

Valid and reliable measures representing the primary constructs of belief in the priority of smoking as a public health problem and norms and values concerning smoking were developed. Attitude change, assessed by a cross-sectional approach, revealed that there were substantial changes in favour of stronger antismoking attitudes in both the intervention and comparison communities. Intervention effects were limited to changes in heavy smokers' beliefs about the seriousness of smoking as a public health problem. The strongest covariate of attitude change was shown to be quitting smoking during the trial for both measures, while changes in beliefs about smoking as a public health problem were demonstrated to be greater for less educated groups in the intervention communities. The attitude change - behaviour change analyses revealed support for traditional theoretical positions in the debate. Conclusions reached, however, suggest that the type of behaviour change being measured (i.e., quitting or starting smoking) and the measure of attitude affect the results of attitude-behaviour analyses. Lastly, 'place' was shown have an important independent effect on attitudes toward smoking. Through case studies of several COMMIT communities, place attributes identified to have an important impact on individual smoking attitudes were economic reliance on the tobacco industry, socio-economic and cultural composition and legislative/political climates.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Geography Commons