VALIDITY AND RELIABILITY OF BIOPSYCHOSOCIAL-RELATED MEASUREMENT SCALES AMONG LOW-INCOME MALAYSIAN SMOKER

Abstract

Introduction: Tobacco is the only legal product that kills a large proportion of its consumers when used as intended by its manufacturer. The effect of nicotine as a driving substance on smoking has been established for decades. Still, very little is known on how the biopsychosocial determinants relationship affects levels of nicotine addiction in smokers, especially in the urban low-income population. The study aimed to validate measurement scales related to biopsychosocial factors that will be used in the future study to evaluate biopsychosocial components that influence nicotine addiction among urban poor smokers. Methods: Exploratory factor analysis (EFA) using the principal component analysis with varimax rotation and Kaiser normalization was used to assess the factor structure. Then, the confirmatory factor analysis (CFA) was conducted to assess the unidimensionality, validity, and reliability of the latent construct. Results: EFA showed extraction of factors according to their original scales with all factor loading and communality's values were above 0.5. During CFA, factor loading less than 0.6 was deleted. Convergent validity verified by computing the Average Variance Extracted (AVE) for every construct range between 0.528 – 0.801. The Fitness Indexes achieved the required level (RMSEA=0.05, CFI=0.937, Chisq/df=1.7). Meanwhile, the Discriminant Validity Index range between 0.75-0.89, which is higher than the correlation coefficient value. Internal consistency assessed from Composite Reliability range between 0.714-0.965. Conclusion: The measurement scales are valid and reliable to assess the intended constructs among low-income male smokers in the urban area.