ABSTRACT
This article analyzes if, and to what extent, the public service motivation (PSM) construct has an added value to explain work motivation in the public sector. In order to address the specificity of PSM when studying work motivation, the theoretical model underlying this empirical study compares PSM with two other explanatory factors: material incentives, such as performance-related pay, and team relations and support, such as recognition by superiors. This theoretical model is then tested with data collected in a national survey of 3,754 civil servants at the Swiss municipal level. Results of a structural equations model clearly show the relevance of PSM. They also provide evidence for the importance of socio-relational motivating factors, whereas material incentives play an anecdotal role.
ACKNOWLEDGEMENTS
This article is part of a broader research project (n° 100012–116083) financed by the Swiss National Science Foundation. The authors would like to thank the reviewers and the editor of IPMJ for their astute comments and suggestions, which contributed to the improvement of the present article. A previous version was presented to the 14th International Research Society on Public Management Conference in Bern 2010. The authors are grateful for the discussion they had with the participants.
Notes
*Classified in years.
Note. In grey, correlations among items measuring the same construct. Mean correlations = 0.1227. Mean shared variance = 0.0369.
**Sig. 0.01 (bilateral); *Sig. 0.05 (bilateral).
*Based on summative indexes.
a Set to (1) for identification purpose.
***Statistically significant at p < .001.
***Statistically significant at p < .001.
***Statistically significant at p < .001.
In this study, the measured dimensions of PSM were: “attraction to policymaking,” “commitment to the public interest,” “self-sacrifice,” and “bureaucratic governance.” The latter dimension was developed by Vandenabeele (Citation2008a) to suit the European context. It reflects traditional principles of European public services such as equality, equity, or continuity.
3 for Skewness and 10 for Kurtosis.
Delery and Doty (1996) on strategic human resource management, the work of Jurkievicz and Massey (Citation1997) on motivational factors. We were also inspired by the items designed by Spector (Citation1985) for job satisfaction subdimensions and by the battery developed by Bourcier and Palobart (Citation1997) on work recognition.
The reflective-formative issue concerns the specification of a model of measurement (for an extensive discussion, see Jarvis, MacKenzie, and Podsakoff Citation2003). It refers to critical issues faced when constructing scales: direction of causality between items and scales, correlations between the items, possibility to delete items of the same scale, and location of the measurement error term. Of course, when turning to a second order construct, as PSM is, the issues are the same and refer both to the relations between items and subscales (or dimensions), and to the relations between the dimensions and the construct. For PSM, the pending question refers to whether or not PSM is a second-order formative construct. The underlying issues are whether or not the four subdimensions should be part of the PSM construct, and how those subdimensions are related.
Results of the rotated (varimax) factor analysis show that at the exception of the “self-sacrifice” and the “commitment to the public interest” dimensions of PSM that load on the same factor (an overlap that is well known in PSM research; Wright Citation2008), all the items are related to their respective constructs.
Fit coefficients for the model without PSM: χ2 = 191.750, df = 24, p-value = .000, TLI = .954, CFI = .976, NFI = .972, p-value of close fit = .975, RMSEA = 0.043.