POSC323 Improving a Rating Scale: Applying Rasch Analysis to Student Pharmacists′ Perceptions of Their Professional Identity

Professional identity(PI) shapes how individuals perceive, evaluate, and assert their professional responsibilities. However, there is limited evidence for the psychometric properties of the instruments used to evaluate PI among student pharmacists. This study aimed to evaluate the psychometric properties of an instrument to assess student pharmacists’ perception of their professional identity. Data from a previously conducted cross-sectional survey were used to assess validity and reliability of student pharmacists’ perceived PI. Students indicated their agreement on a 6-point scale(strongly-disagree, disagree, somewhat-disagree, somewhat-agree, agree, strongly-agree) to identifying as scientists, medicine-advisor, clinical-practitioner, social-carer, medicine-maker, medicine-supplier, leader, businessperson, unremarkable-character, or entrepreneur. Validity and reliability of the instrument were assessed using the student and item separation index(SI) and reliability coefficients(RC). Linear regression assessed the association of students’ characteristics with perceived PI. Analyses were conducted in Winsteps and SAS. The analysis included 118 students. The primary analysis suggested inadequate scale functionality; thus, item responses(somewhat-disagree and disagree; somewhat-agree and agree) were merged. With modified item anchors, the data fit the model with infit-and outfit-mean-squared-values. After removing the most misfitting item, the item-person map showed appropriate distribution of item difficulty, identifying medicine-adviser and entrepreneur as the easiest and hardest items to agree with, respectively. Students’ perceptions mismatched with item hierarchy on the measurement continuum and exhibited some ceiling effect, thus additional items reflecting students’ higher perceptions of PIs are needed. While students’ RC was 0.98 and SI 1.73, item RC was 0.94. Students who reported receiving formal PI training were 0.30 (p=0.0001) times more likely to have high perception of PI than those who did not. The instrument exhibited fair validity and reliability for assessing the perception of PI among student pharmacists. Additional items to fill identified item gaps are needed to improve the PI instrument, with further assessment of this revised instrument.