Evaluation of the construct and internal validity of the Professional Identity Questionnaire: A Rasch analysis

It has been suggested that occupational therapy practitioners with a poor sense of professional identity experience more difficulties in the workplace and may be less likely to implement evidence-based, occupation-focused interventions. Poor sense of professional identity has also been linked to increased risk of burnout. Although professional identity is considered important, few measures are available and none have been rigorously evaluated. The project was established to evaluate the construct and internal validity of one such measure, the Professional Identity Questionnaire (PIQ).

The PIQ includes 10 questions, each rated on a 5-point Likert scale. Rasch analysis was used to evaluate several aspects of construct and internal validity of the PIQ. These included rating scale validity, unidimensionality, statistical validity, differential item functioning, overall construct validity and person response validity. Correlations were also calculated to explore the relationship between the Professional Identity Questionnaire and a commonly used measure of burnout, the Oldenburg Burnout Inventory (OLBI).

A total of 118 occupational therapists working in mental health completed the PIQ and OLBI. Initial analyses suggested that the rating scale validity and targeting of item difficulties to person abilities were not optimal. Revision of the rating scale structure improved the overall rating scale validity and item targeting. Analyses using the revised scale structure showed generally good alignment with quality control criteria established for this study. Fit statistics for one item were less than optimal and some differential item functioning was present. Correlations between PIQ and OLBI total scores were -0.53 for the original PIQ rating scale structure and -0.46 for the revised rating scale structure.

Results from this study provide preliminary evidence for the construct and internal validity of the PIQ, especially when using the revised scale structure. However, further research is required before the PIQ can be recommended for widespread use.