Why construction workers involve in unsafe behavior: Part B: Scale validity and reliability

Abstract

Background and aim: Construction industry continues to rank among the most dangerous industries in the worldwide. The aim of this study was the development of a new scale for measuring the factors influencing on unsafe behavior in the construction industry from the perspective of experienced frontline workers, supervisors and managers. Methods: In the current qualitative and quantitative study, scale items were extracted from a qualitative study. Validity assessment was conducted through the content and construct validity. Correlation Coefficient (CC) and Cronbach’s alpha testes were used to internal consistency and stability, respectively. As well as, specificity and sensitivity assessment used to determining the cut of point. Results: In the accordance of results from the qualitative, twelve components with factor value over 0.42 were extracted. These factors were explained 62 percent of total variance of unsafe behavior in the construction industry. Cronbach’s alpha and CC were 0.948 and 0.901, respectively. Coordinates of specificity and sensitivity points on ROC curve indicated 120.5 cut off point for new scale. Conclusion: For measurement of the factor influencing on unsafe behavior in the construction industry, we must use a valid and reliable instrument. The current scale is valid and reliable as well as specific and sensitive for this reason.

Keywords

• unsafe behavior
• qualitative study
• factor analysis
• reliability
• new instrument
• construction workers.