Search

The assessment of noncognitive traits is challenging due to possible response biases, “subjectivity” and “faking.” Standardized third-party evaluations where an external evaluator rates an applicant on their strengths and weaknesses on various noncognitive traits are a promising alternative. However, accurate score-based inferences from third-party evaluations requires disentangling score variance due to raters versus applicants by utilizing a multilevel factor analysis (MFA). To date, MFA is highly underutilized in the measurement field. In this study, we apply the MFA to analyze third-party evaluations using data from the Personal Potential Index (PPI). The PPI is a third-party measure used to evaluate graduate school applicants noncognitive traits to help inform admissions decisions. We analyzed 12,693 ratings of 6,249 applicants divided into two randomly selected subgroups. We conducted multilevel exploratory factor analysis with one subgroup and tested the hypothesized structure with the othe...

Differential item functioning (DIF) analyses have been used as the primary method in large-scale assessments to examine fairness for subgroups. Currently, DIF analyses are conducted utilizing manifest methods using observed characteristics (gender and race/ethnicity) for grouping examinees. Homogeneity of item responses is assumed denoting that all examinees respond to test items using a similar approach. This assumption may not hold with all groups. In this study, we demonstrate the first application of the latent class (LC) approach to investigate DIF and its sources with heterogeneous (linguistic minority groups). We found at least three LCs within each linguistic group, suggesting the need to empirically evaluate this assumption in DIF analysis. We obtained larger proportions of DIF items with larger effect sizes when LCs within language groups versus the overall (majority/minority) language groups were examined. The illustrated approach could be used to improve the ways in which DIF analyses ...