METHODS: This study applied the 28-item IMPROVE-FL questionnaire in a national survey of 29 medical schools with quota sampling to gather evidence of construct validity (via confirmatory factor analysis and hypothesis testing) and assess its internal consistency (via construct reliability analysis). Subsequently, a smaller study was conducted in two medical schools, one public and one private, to establish evidence of test-retest reliability.
RESULTS: These analyses reported that the finalised 23-item IMPROVE FL had excellent model fit and evidence of construct validity (β = .64-.87; x2/df = 3.86; RMSEA = .05; SRMR = .03; CFI = .95; TLI = .94; GFI = .93; AGFI = .91), confirmed the expected correlation with fixed mindset (r = -.116, p r = .327, p