Tobacco Induced Diseases (Oct 2018)

The factor structure and factorial invariance of short form of smoking temptation for TTM framework in Iranian smoker population: Golestan Province

  • Abdurrahman Charkazi,
  • Bagher Pahlavanzadeh,
  • Rahman B. Ozouni- Davaji,
  • Gholamreza Sharifirad,
  • Alireza Abadi

DOI
https://doi.org/10.18332/tid/94530
Journal volume & issue
Vol. 16, no. 3

Abstract

Read online

Purpose TTM pattern has widely been used in smoking cessation interventions. This study was aimed to investigate the compatibility of a hierarchical factor structure of short form questionnaire developed by Velicer et al regarding situational temptation construct among smokers in Golestan province. Methods A cross-sectional study was conducted on 387 smokers that were included using inconvenient sampling method. At first, internal validity and reliability of the short-form questionnaire were examined. At second, invariance of hierarchical structure was tested in sub-groups of ethnicity, residence, education, stage of quit, starting age, and income. Measurement invariance (including five invariances) and structural invariance (including three invariances) were explored in each sub-group. Results Chi-square test for confirmatory factor analysis of hierarchical factor structure was statistically significant for all samples [χ^2 (24)=66.3,p<0.001]; while, other indicators reveled the good compatibility of factor structure (CFI=0.944, TLI=0.915, RMSEA=0.067 ((0.048, 0.087)) and SRMR =0.042). Values of the first order of factor loading were between 0.5 and 0.84. There was strong correlation between first order factors and second order factor such that it was 0.9 for positive and habitual social situations factors and 0.82 for negative effects factor. Cronbach's alpha coefficient for the whole model was also 0.803. Conclusions Given the findings, hierarchical factor structure of short-form inventory about temptation to cigarette smoking was confirmed by three factors that were in line with Velicer et al model. The fitting indices of the model presented that construct validity was observed in all samples and sub-groups, and can be likely used in smoking cessation programs. Funding Golestan University of Medical Sciences, Gorgan, Iran

Keywords