Organizational Design and Public Management: Extending the Experimental Methods Agenda

Public management research draws on the full methodological repertoire of the socialsciences in the search for valid and reliable evidence. In the pursuit of ‘realistic evidence’there has been a recent turn towards the use of experimental methods (Margetts 2011).This project will build upon this shift to examine the validity of social scientificknowledge by replicating and extending experimental studies on key questions oforganizational design.The growth in public management studies employing experimental methods has notbeen matched by replication studies: to date none have been published. While challengesexist in publishing replications, there are growing examples in the social sciences (e.g.Nosek and Lakens 2014). Replications are not identical copies, rather they seek to testthe theories employed in prior studies. In this project replications will be ‘empiricalgeneralizations’ and ‘generalization and extension’ (Tsang and Kwan 1999):- Empirical generalizations draw on the same designs but uses different populationsand settings. This leads to the proposition that the different context for thereplications—Hong Kong as again Europe/USA—will result in findings in the samedirection but with weaker results. It is theorized that this will arise from the morecollective culture and Confucianism in Hong Kong.- Generalization and extension will see two extensions undertaken. The first addressesconcerns in the discipline about practical relevance of findings on studies using students,thus experiments will be extended to include managers. Second, the impact of incentiveson intrinsic and extrinsic motivation will be examined.The majority of experimental studies in public management have examinedorganizational design. We take this cue to inform the selection of the five replicationsthat will be conducted in this study and examine performance management (Weibel etal. 2010), work motivation (Brewer and Brewer 2011), decision-making (Nutt 2005), andrules and red tape (Kaufman and Feeney 2014; Scott and Pandey 2000). The study,therefore, seeks to contribute towards better organizational designs that produceimproved outcomes for public agencies by better understanding the generalizabilitypublic management knowledge.The project will be hosted in the Laboratory for Public Management and Policy, whichprovides the resources and facilities to undertake survey, vignettes, simulations andcomputer-based lab experiments. The study team, in keeping with best practice onreplication and extensions, draws authors as Co-Is from the original studies givingaccess to research materials, commentary on research design and interpretation ofresults.