Applying the Bradley–Terry–Luce method to P–E fit

Abstract

This study attempted to increase the size of the correlation between person–environment (P–E) fit and job satisfaction by rescaling the instrumentation of the Theory of Work Adjustment using the Bradley–Terry–Luce method and a probability-based fit index. This approach worked as well as, but failed to outperform, the currently used correlation-based fit index. However, a probability-based fit index offers the advantage of being intuitively easier to understand than a correlation-based index. The very high correlation between the correlation-based fit index and the probability-based fit index suggests that both assess the same construct. It is argued that, considering the restriction of range common to all assessments of job satisfaction, the correlation between fit index and job satisfaction may not represent an upper limit, but rather a lower bound of the relationship between P–E fit and job satisfaction.