Public service motivation measurement - Testing an abridged version of Perry's proposed scale

This article proposes and tests a three-dimension, shortened instrument for public service motivation based on Perry's (1996) exploratory 24-item scale. A confirmatory factor analysis is used with both a diagonally weighted least squares and a modified version of maximum likelihood estimation that are more appropriate for ordinal indicators. Data from the National Administrative Studies Project, consisting of mail surveys of managers engaged in information management activities working in state-level primary health and human service agencies, is used (n = 274; response rate = 53%). Results indicate good support for the shortened scale compared to Perry's original work. Findings corroborate Perry's theorized dimensions and items in the exploratory instrument. An exploratory analysis evaluating modifications to the tested model suggests possible improvements in two dimensions but does not undermine the overall, supportive results and suggests the importance of continued item and scale development.