The police stress inventory: Does it measure stress?

Research on stress among law enforcement personnel has not kept pace with the interest in this subject. This study had two primary objectives: to extend the psychometric evaluation of a modified version of Lawrence's Police Stress Inventory (PSI) and to evaluate empirically its concurrent convergent and discriminant validity. Data were collected from 259 certified North Carolina law enforcement officers. Factor analysis and various psychometric analyses raised some concerns about the structural integrity of the PSI. Further analyses revealed that the PSI correlated more highly with a job satisfaction measure than with another measure of stress and, thus, failed to exhibit discriminant validity. These findings suggest that the PSI is not a valid measure of stress. It is argued that research findings about stress have serious implications for policy and practice and that considerable care must be taken in developing and selecting instruments that purport to measure stress among law enforcement officers.