UBC Theses and Dissertations
Assessing discrimination in a police recruit assessment center Tinsley, Paul N.
The overall concern of this study is that of substantive equality, as defined by Canadian law, in the employment context, and the specific goal of this study is to provide a model to assess (and prevent) unlawful systemic discrimination in an assessment center. Because discrimination is essentially the same, wherever it occurs, the model proposed in this study is also useful for assessing discrimination in employment selection generally. In the employment context, evidence of systemic discrimination is often limited to selection patterns, and so this study argues that statistical analyses can be particularly useful. Since the Supreme Court adopted the effects theory, where intent is immaterial and the focus is on results, such analyses are likely to become an appealing alternative to traditional arguments of exclusion and disproportion. The analytic model proposed here suggests two general phases to a legal analysis of discrimination. First, there is the preliminary phase, which consists of three interrelated steps: identifying the applicable selection procedure, identifying the relevant legal issue, and identifying the appropriate groups for comparison. Second, there is the assessment phase, which consists of two sequential steps: comparing the groups of interest on the dimension of interest to determine if differences exist, and analyzing observed differences to determine if they are legally or practically significant. It is in this phase that statistical analyses can be especially helpful in an assessment of systemic discrimination. To test its utility, the proposed model was applied to the Justice Institute of British Columbia Police Academy assessment center (where entry level police applicants are screened) to determine whether the assessment center discriminated on the basis of sex. Of particular interest to the Police Academy is that the results indicated no sex discrimination, but notably the results also indicated that the proposed model provides a practicable and relatively uncomplicated way to assess discrimination. Moreover, consistent with the goal of prevention, this study demonstrates how a reliability assessment can provide important information about the potential for discrimination in employee selection, thereby providing employers with the means to be more proactive than otherwise possible.
Item Citations and Data