Predictors of Physician Assistant Faculty Intent to Leave Academia: A Rasch Regression Analysis.

Abstract

The purpose of this research was to investigate whether a previously developed and validated measure of intention to stay in academia for physician assistant (PA) faculty could also be used to inform PA faculty intent to leave academia and return to full-time clinical practice. A secondary analysis of the national survey data of PA faculty (n = 427) was conducted using a Rasch regression model. The 19 items from the previously validated Supportive Environment Scale were examined as potential predictors of the dichotomous outcome of intent to leave academia. In this analysis, PA faculty who were somewhat or very likely to leave academia were compared with those who were not at all likely to leave academia. Five items from the Supportive Environment Scale emerged as a linear meaningful measure that predicted intent to leave. These predictors included the following: recognition by administration, support for scholarly work, a fair promotion process, a sense of institutional community, and support of the PA program by administration. Rasch prediction maps were produced that allowed for powerful visualization of prediction and identification of areas for intervention. The intent of PA faculty to leave academia and return to clinical practice can be meaningfully predicted by 5 items. Importantly, these items are in the purview of institutional administrators and can be used to target PA faculty retention efforts.