PSYCHOMETRIC PROPERTIES OF THE SMOKING CONSEQUENCES QUESTIONNAIRE ACROSS SEX AND DURING SMOKING CESSATION TREATMENT
MetadataShow full item record
Cigarette smoking is the leading cause of preventable death and disability in the United States. There has been a growing interest to understand cognitive processes that may influence smoking. One promising cognitive process linked to smoking behavior is smoking outcome expectancies. Although there is no standard instrument for assessing smoking outcome expectancies, perhaps the most commonly used instrument is the Smoking Consequences Questionnaire (SCQ). Despite the wide use of this measure, little work has been devoted to examining its psychometric properties and no work has examined its construct stability across sex or during smoking cessation treatment. Therefore, the current study aimed to replicate and extent psychometric testing of the SCQ by examining the assumed multi-factor structure, reliability, convergent/discriminant validity, and predictive validity, as well as assessing measurement invariance across sex at baseline, quit week, 1-month follow-up, and 3-month follow-up and longitudinal measurement invariance among baseline, quit week, 1-month follow-up, and 3-month follow-up with a sample of treatment seeking daily cigarette smokers. Results indicated that the originally proposed four-factor SCQ does not adequately explain covariance among items, particularly across the Negative Consequences and Positive Reinforcement/Sensory Satisfaction subscales. Using item reduction techniques, adequate model fit was obtained for a refined, four-factor solution. Construct validity of the refined measure was demonstrated, however limited predictive validity was observed. Finally, although measurement invariance was not appropriate to examine across the full measure, the original SCQ Negative Reinforcement/Negative Affect Reduction and Appetite-Weight Control subscales demonstrated measurement invariance across sex at two time points, as well as longitudinally across two time points. Overall, the present findings provide initial evidence for an item reduced model structure of the SCQ and suggests divergent model structures across sex.