A systematic review of experimental studies in public management journals

Introduction Before the 1990s, there was very little public management research that employed experimental methods, despite some evidence of early experimentation in the classical roots of public management as discussed in Chapter 3. Margetts (2011), in her review of public management studies using experimental methods, found only a scattering of experiments since 1960 in leading journals at that time (including Journal of Public Administration Research and Theory, Public Administration Review, Public Administration, and Public Management Review). In an early statement of the case for experiments, Bozeman argued that perhaps the lack of an experimental tradition was ‘because public policy management researchers are more interested in the problem than in problem-solving techniques’ (1992, pp. 440–1). But the situation has begun to change in more recent years, as this chapter documents, with a marked increase in experimental studies in the top public management journals. In this chapter, we take a systematic look at these trends and the characteristics of experimental studies published in top public management journals, focusing on the period 1990–2015. How We Did Our Systematic Review The studies in our systematic review come from the top 20 journals in Google Scholar's category of public policy and administration and are listed in Table 2.1. Using Google Scholar's category provides a wider range of journals than previous reviews (Anderson and Edwards 2014; Bouwman and Grimmelikhuijsen 2016; Margetts 2011) and avoids relying on personal judgment, although a few of the journals in the Google Scholar category are more public policy journals than public management journals. The Google Scholar ranking we use is based on the h5-index, which is the h-index for the most recent five-year period, where h is the highest number such that h articles have at least h citations. The h-index is designed to provide a score that avoids over-weighting a journal with just a few highly cited articles, or one with a great many articles with just a few citations (because the journal simply publishes more articles than other journals). In other words, journals do better in the h5-index ranking when they publish more highly cited articles.