Restraint influence on state anxiety and personality assessment inventory responding
Standardized test administration is key to a test’s validity, but often thwarted in correctional environments when policy or staff require that examinees wear restraints. Restraints such as handcuffs may impact participants’ state anxiety, stress, frustration, and irritability, as well as the profile validity of their responses. Participants in this study were randomized to cuffed, partial-cuffed, or uncuffed conditions. They completed a state anxiety measure, Positive and Negative Affect Schedule (PANAS), and a multiscale (trait) inventory, the Personality Assessment Inventory (PAI). Due to sample size limitations, experimental conditions (cuffed and partial-cuffed) were combined into the restrained condition for analyses and compared to their uncuffed (unrestrained) counterparts. The author hypothesized that individuals who were restrained at any point in the study would report higher state anxiety (PANAS NA) and have elevated PAI scales related to anxiety (ANX), frustration (STR), and irritability (MAN-I) when compared to individuals who were unrestrained. In addition, it was hypothesized that those who were restrained at any point during the study would have elevated PAI validity scales (ICN, INF, and NIM) when compared to their unrestrained counterparts. Notably, the power of the study was limited by sample size; post-hoc power analyses rendered .06 to .26 power across comparisons, which indicated a low likelihood of identifying significant effects. Independent samples t-tests found no significant differences between the restrained and unrestrained conditions on the aforementioned comparisons, and small Cohen’s d values for four of the six comparisons. While the study’s findings do not strongly inform practice recommendations, they bring attention to the frequent issue of unstandardized test administration in forensic contexts.