Nicholas Kerswell, Grace Shuttleworth, Georgia Dellosa, Benjamin Bonenti, Kris Crane, Xan Dangaard, Timothy Piatkowski
The theorized five-factor structure of the S-UPPS-P was broadly supported. Model fit indices demonstrated acceptable comparative fit (CFI = 0.938, TLI = 0.927, SRMR = 0.084), though RMSEA was elevated (0.103). Negative urgency, positive urgency, and lack of premeditation demonstrated strong and consistent factor loadings and internal reliability (α = 0.76-0.80). By contrast, lack of perseverance and sensation seeking demonstrated weaker loadings and lower internal consistency (α = 0.65 and 0.63, respectively). Latent factor correlations were theoretically coherent and consistent with multidimensional models of impulsivity.
INTRODUCTION: Despite its growing use in Australian AOD clinical settings, limited local psychometric evidence is available for the Short UPPS-P in treatment-seeking populations. This study aimed to evaluate the factor structure and reliability of the Short UPPS-P to determine its clinical utility as a multidimensional measure of impulsivity in an Australian AOD treatment setting.
METHODS: De-identified baseline outcome measure data were obtained from clients accessing a large Australian non-government AOD treatment provider (N = 18,786). Confirmatory factor analysis (CFA) was conducted in R using weighted least squares mean and variance-adjusted estimation (WLSMV), treating S-UPPS-P items as ordered categorical indicators. Internal consistency and latent factor correlations were also examined.
RESULTS: The theorized five-factor structure of the S-UPPS-P was broadly supported. Model fit indices demonstrated acceptable comparative fit (CFI = 0.938, TLI = 0.927, SRMR = 0.084), though RMSEA was elevated (0.103). Negative urgency, positive urgency, and lack of premeditation demonstrated strong and consistent factor loadings and internal reliability (α = 0.76-0.80). By contrast, lack of perseverance and sensation seeking demonstrated weaker loadings and lower internal consistency (α = 0.65 and 0.63, respectively). Latent factor correlations were theoretically coherent and consistent with multidimensional models of impulsivity.
DISCUSSION AND CONCLUSIONS: Findings provide support for the broad applicability of the five-factor S-UPPS-P structure within an Australian AOD treatment context when analyzed using appropriate ordinal psychometric methods. However, weaker psychometric performance within the lack of perseverance and sensation seeking domains suggests some items may function less optimally in this clinical population. These findings support continued refinement and contextual adaptation of impulsivity measures for Australian AOD settings.