Construct validity of a work environment impact scale

This study examined the construct validity and the internal consistency of a newly developed assessment, the Work Environment Impact Scale (WEIS). After administration to 20 individuals with psychiatric disabilities, Rasch analysis was utilized to scrutinize the data. For this study, criteria for determining unexpected person/item responses were based on the following criteria: MNSQ > 1.3 and $\underline{t}>2.0$. Persons/items with MNSQ < 0.7 and $\underline{t}<-2.0$ were examined to enhance instrument precision but were not considered misfit. Results suggest that the WEIS is an appropriate and valid instrument to utilize with workers who have psychiatric disabilities. Overall, the items appeared to match the worker's need for performance, satisfaction, and well-being as the mean of persons measure is 0.30 ± 0.49 logits more than the mean of items measured. In addition the hierarchical order of items is consistent with literature identifying environmental press and affordance for workers with psychiatric disabilities. WEIS items constitute a uni-dimensional construct given that the summary statistics for both item and person had a MNSQ of 1.00 and 1.02 respectively and both $\underline{t}$ values were -0.2. However, three items exceeded the established criterion for being too informative given the MNSQ < 0.7 and a $\underline{t}<-2.0$., suggesting these needed to be revised to enhance the quality measurement of the instrument. One hundred percent of the workers fit the expected response pattern of the Rasch model suggesting that workers with greater satisfaction, performance and health had a higher degree of match with his/her occupational environment. In summary, anecdotal data suggested that the WEIS provided clinically relevant information useful for planning of work-related interventions or reasonable accommodations.