Structural Validity, Reliability, Measurement Invariance, and Construct Validity of the Personal Growth Initiative Scale-II: A Systematic Review Using COSMIN Methodology

Personal Growth Initiative (PGI), a cognitive and behavioural skill set for intentional growth, is often measured using the multidimensional Personal Growth Initiative Scale-II (PGIS-II). This systematic review examined its structural validity, reliability, measurement invariance, and construct validity. A literature search across multiple databases identified 15 studies evaluating the psychometric properties of the PGIS-II. Study quality was assessed with the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) Risk of Bias checklist (version 3.0). The specified psychometric properties were evaluated based on COSMIN criteria for good measurement properties, and a modified Grading, Recommendations, Assessment, Development, and Evaluation (GRADE) approach assessed the certainty of evidence. Findings showed structural validity and internal consistency variability, with some modifications to the first-order four-factor model lacking theoretical justification. A second-order four-factor structure and a bifactor model demonstrated moderate to high-quality evidence. Test–retest reliability was confirmed only for the Chinese (moderate quality) and Turkish (low quality) versions. Measurement invariance was supported for the English and Portuguese first-order four-factor models, as well as the English bifactor model. Construct validity was examined for seven versions, yielding mixed results. While some versions may be suitable for specific contexts, inconsistencies in psychometric evidence emphasise the need to further clarify the instrument's structure and psychometric performance in diverse populations through high-quality studies. Nevertheless, based on our findings, the bifactor model appears to be the most suitable structure for the PGIS-II, as it captures both the general PGI construct and its subdimensions, at least for now.