Measurement invariance in careers research: Using IRT to study gender differences in medical students' specialization decisions

The current study demonstrates the use of item response theory (IRT) to conduct measurement invariance analyses in careers research. A self-report survey was used to assess the importance 1,363 fourth-year medical students placed on opportunities to provide comprehensive patient care when choosing a career specialty. IRT analyses supported measurement invariance across gender. Additional analyses indicated that compared with men, women placed significantly greater importance on opportunities to provide comprehensive patient care. This in turn predicted career choice, with women being more likely than men to pursue primary care specialties. This study extends the careers literature both methodologically and substantively. Methodologically, this study exemplifies how and why to use IRT to assess measurement invariance prior to comparing groups on career-related attitudes. Substantively, this study is the first to demonstrate that the importance placed on comprehensive patient care mediates the effect of gender on intentions to pursue primary care specialties.