Design and validation of a screening method to identify accident - proneness in industry

Investigations show that the scattering of accidents in individuals is not the same, and three quarters of all accident occur to one quarter of accident prone individuals. The aim of the study was to design and validate a questionnaire for screening out the accident prone individuals.
This study in 5 stages of item generation, content validity assessment, reliability, construct validity, and criterion validity was performed among 230 workers who had and who had not experienced accident in 19 industries of Isfahan, Iran. For data analysis SPSS and Amos 22 were used.
195 items were identified during item generation stage. In content validity evaluation stage 13 items were removed, 15 items were modified and 79 items were merged and convered to 14 items. Internal consistency (Cronbach's alpha) of the remained items was 0.86. The results of confirmed validation showed that comparative fit index was acceptable. (CFI=0.99, GFI=0.96, RMSEA=0.066, RMR= 0.022). T test showed that the mean scores of the workers who had accident were significantly higher than the workers who had not experienced accident (p The result of confirmed validation showed that there is a good agreement between the proposed model and the field data. Reliability and construct validity showed that the current questionnaire is suitable for screening out the accident prone individuals.