Measuring the Effects of Correctional Officer Stress on the Well-Being of the Officer and the Prison Workplace and Developing a Practical Index of Officer Stress for Use by Correctional Agencies

The prison work environment creates higher levels of stress among correctional officers than is found among employees in other occupations, and elevated levels of stress have significant debilitating effects on the well-being of the officer. Less is known about the effects of officer stress on the well-being of the prison workplace, but ample evidence from other occupations suggests that elevated levels of stress will have significant debilitating effects on officer job performance. With this knowledge, correctional agencies' EAP/ HR are trying to find a practical method to identify those officers in need of therapeutic intervention. This project will (1) advance current knowledge about the effects of correctional officer stress on the prison workplace and, (2) create and validate a practical 'stress index' developed on the basis of administrative data routinely collected and maintained by the correctional agency.
The first phase uses multiple regression modeling to test a path model of the independent effects of stress on the officer's attitudinal, emotional, and behavioral well-being and how the independent and combined effects of these individual-level outcomes affect officer task performance, organizational
citizenship behavior, and counterproductive workplace behavior. In both Arizona and Texas, a random sample of 250 correctional officers (N=500) will be interviewed to collect a self-reported measure of stress, individual-level stress effects, and organizational-level stress effects. These
self-reported data will be augmented with information on each surveyed officer available in the correctional agency's administrative files.
Since frequent surveys of all officers to identify those individuals with high stress levels are not practical, agencies need a measure of stress based on administrative data already being collected. Based on first-phase findings, structural equation modeling will be used to develop a 'stress index' based on the differential weights of salient administrative data. This index then will be validated by comparing the findings of interviews conducted with a purposive sample of officers identified as having a high stress level (N=150) and a random sample (N=150), for a total of 300 interviews in each state. The self-reported measures of stress and stress outcomes among the correctional officers are the 'gold standard' against which to compare the index scores, and the predictive validity of the stress index will be established if the stress level and individual stress effects are significantly greater within the purposive sample than within the random sample. ca/ncf