Survey Research in Public Administration: Assessing Mainstream Journals with a Total Survey Error Framework

Abstract

Survey research is a common tool for assessing public opinions, perceptions, attitudes, and behaviors for analyses in many social science disciplines. Yet there is little knowledge regarding how specifi c elements of survey research methodology are applied in practice in public administration. Th is article examines fi ve mainstream public administration journals over an eight-year period regarding current methodological practice, organized around the total survey error framework. Th e fi ndings show that survey research in the fi eld of public administration features mainly small-scale studies, heavy reliance on a single data collection mode, questionable sample selection procedures, and suspect sample frame quality. Survey data largely are analyzed without careful consideration of assumptions or potential sources of error. An informed evaluation of the quality of survey data is made more diffi cult by the fact that many journal articles do not detail data collection procedures. Th is study concludes with suggestions for improving the quality and reporting of survey research in the fi eld. S ince McCurdy and Cleary (1984) fi rst discussed quality issues in doctoral dissertations in public administration, a number of other scholars have explored the quality of research methodologies used in the fi eld (e.g.,

Development of Public Administration and its Research in Slovenia through the Lenses of Content Analysis of the International Public Administration Review

“…On the other hand, it is worth mentioning that these results are completely in line with the results of other CAs in the field. Namely, methodological weaknesses in PA research were identified in CAs conducted by Perry & Kraemer (1986), Lee et al (2009), andWalker et al (2014).…”

“…In the field of PA, CA was first conducted among doctoral dissertations in the field (Lee et al, 2009), but later often extended to research papers. However, previous CAs in this field were mainly focused on US-based PA research, while there has never been a CA conducted in the field of Slovenian PA research.…”

“…The methodological part of our code-book is mainly derived from Lee et al (2009). The classification used for the analysis of topical orientation is our own; even though we are aware that comparison of the results is of great importance, we believe that the popularity of the topics is not the same in the US and Central/Eastern Europe (due to different historical/political/systemic background elaborated in previous chapters).…”

“…The origin of CA in the PA research lies in the analysis of quality of doctoral dissertations in the field (Lee, Benoit-Bryan, & Johnson, 2009). Later, CA was extended to research papers many times by many researchers.…”

What makes public-private partnerships work? Survey research into the outcomes and the quality of cooperation in PPPs

“…So, the survey represents approximately the whole population of officially known Dutch PPP projects up to 2014. By including almost the entire population in our study, we avoid many of the issues with regard to representation as described in the total survey error framework (see, for example, Groves andLyberg 2010 or Lee, Benoit-Bryan, andJohnson 2012). Coverage or sampling errors, which arise in the process of selecting a sample from a target population, are therefore most likely not present in our study.…”