Premium
Constructing an integrated model of the antecedents of adolescent smoking
Author(s) -
Wilkinson Dawn,
Abraham Charles
Publication year - 2004
Publication title -
british journal of health psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.05
H-Index - 88
eISSN - 2044-8287
pISSN - 1359-107X
DOI - 10.1348/1359107041557075
Subject(s) - psychology , extraversion and introversion , theory of planned behavior , psychosocial , personality , developmental psychology , big five personality traits , structural equation modeling , path analysis (statistics) , cognition , self esteem , clinical psychology , social psychology , control (management) , statistics , management , mathematics , neuroscience , psychiatry , economics
Objective: Reviews have called for integrative, theoretically informed models of the ‘hundreds of associations’ (Miller & Slap, 1989, p. 131) between psychosocial measures and adolescent smoking (e. g. Tyas & Pederson, 1998). Such a model was tested. Method: A prospective, classroom‐based survey measuring previously identified correlates of smoking allowed comparison of the strength of relationships between antecedents and smoking status six months later. The prospective sample included 225 13 to 14‐year‐olds. Measures of behaviour‐specific cognitions derived from the Theory of Planned Behaviour (TPB)—as well as personality, self‐esteem, parental support and parental control, sociodemographic factors, and descriptive norms—were included. Relationships between antecedents were explored using path analyses. Results: High initial rates of smoking were observed. Of the variance in smoking six months later, 56% was explained by seven direct predictors: intention, perceived ease of smoking, estimated number of friends smoking, percentage of older brothers smoking, self‐esteem, extraversion and car access. Discussion: Results emphasize the importance of behaviour‐specific cognitions specified by the TPB but suggest that other factors, including extraversion and self‐esteem, need to be included in models of the antecedents of smoking. The findings also imply that some antecedents, such as parental support, may indirectly influence adolescent smoking through their effect on other variables.