A preliminary development and validation of a measure of safety performance
Abstract
Safety researchers have devoted extensive attention to safety performance behaviors. However, current safety performance models have yet to differentiate between safety citizenship behaviors directed towards the organization and those directed towards individuals. This might be a potential oversight, considering that citizenship behaviors targeted at different beneficiaries might be associated with different antecedents. As such, the purpose of the present study was to develop and validate a new safety performance scale. Items from existing measures formed the item pool and those tapping into the proposed dimensions were selected. Next, items were pilot tested using an online panel of 333 employees from various safety-related industries. A 4-factor structure emerged after exploratory factor analysis and the scale was further refined using reliability analysis and item response theory analysis. Finally, confirmatory factor analysis was conducted to replicate the factor structure using data from 137 employees. Theoretically related variables were correlated with the safety performance dimensions to establish the nomological network. Results supported the 4-factor structure of the new safety performance scale and construct validation hypotheses were largely supported. Implications, study limitations, and directions for future research are discussed.
Degree
M.S.
Advisors
Sliter, Purdue University.
Subject Area
Occupational health|Occupational psychology|Organizational behavior
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.