As a test of the two major theoretical models, two groups of recalled police officers who were laid off approximately 1 and 2 years, respectively, were administered the short-form Rokeach Dogmatism Scale. Employment status during layoff was considered essential in comparing authoritarian levels between the two groups. It was hypothesized that if authoritarian levels of both groups, regardless of layoff employment status, did not differ significantly, then the predispositional model would be supported. If the groups' levels of authoritarianism did vary significantly, then the socialization model would have merit. Although race and age are associated with authoritarianism levels, the results support the predispositional model. Implications from the findings are discussed. (Publisher abstract)