A Systematic Review of Experimental Research on Public Service Motivation

Abstract

AbstractExperimental research has been increasingly recognized as the gold standard for causal inference. However, despite the call for the adoption of experimental approaches to the study of public service motivation (PSM) – one of the most researched topics in the field of public administration – we find that the number of such studies is small in this first-ever review of systematic assessment of experimental research on PSM. Reviewing the universe of such experimental studies − 41 in total – by their topic and type, first, we find that they are concentrated on the analysis of the efforts of PSM and in survey experiments by type. Second, while many studies confirm that PSM positively influences various outcomes, they also show that other intrinsic and extrinsic motivations often outperformed PSM. Third, while acknowledging that methodologically, the randomization of treatment has been well adopted in recent years, we highlight some key recommendations for further improvements regarding the type of experiment, sample characteristics, measurement, and treatment sequences based on our review.Keywords: Experimentsfield experimentpublic service motivation (PSM)social desirability biassurveysystematic review Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 For the parsimony and replicability of our case selection, we followed the outlined process while limiting subjective researcher decisions as much as possible. That said, we acknowledge that we may have missed some closely related studies on PSM that did not directly use the term or did not involve experimental manipulation and random assignment of PSM or other key variables of interest that might causally affect PSM.2 The coding process included all three coauthors to thoroughly review the coding results. At the initial stage, all three coauthors jointly searched, screened, and reviewed the relevant articles. After obtaining 41 articles, two of the coauthors regularly held meetings – both face-to-face and via online – to discuss the coding results and decide which categories to code. In this process, the two coauthors independently read the articles and cross-checked the results. After the cross-validation by the two coauthors, the third coauthor conducted the final review.3 We categorized each study based on the information it provided as much as possible. For the most part, this was relatively easy since most studies explicitly described the type of experiment they conducted. One study (Sun et al., Citation2019) did not explicitly identify the type of experiment employed, and we used our discretion to categorize the study as a lab experiment.4 One article was categorized as both a lab and survey experiment. See Tepe et al. (Citation2022).5 Ranking fifth highest with 20 out of 400 studies (5% of the total), according to Ritz et al. (Citation2016).6 Table A1 in the Appendix provides more detailed analyses of each study.7 Detailed information available in the Table A2 in the Appendix8 Detailed information available in the Table A2 in the Appendix9 We did not find this to be an issue among the four studies that examined the determinants of PSM. Detailed information available in the Table A2 in the AppendixAdditional informationFundingThe research was funded by the Office of Development Research & International Cooperation at the KDI School of Public Policy and Management.Notes on contributorsKee Hoon ChungKee Hoon Chung is an Assistant Professor at Department of Public Administration, University of Ulsan.Inbok RheeInbok Rhee is an Associate Professor at KDI School of Public Policy & Management.Cheol LiuCheol Liu is a Professor at KDI School of Public Policy & Management.