Conducting large-N Surveys on Policy Work in Bureaucracies: Some Methodological Challenges and Implications from the Czech Republic
Abstrakt
The article deals with methodological challenges in conducting large-N surveys on policy work in central public administration. Based upon current work on the topic as well as the author’s experience with a large-N survey in the Czech Republic’s ministries, these challenges include: 1) definition of theoretical sample, 2) obtaining sample frame and support from the ministry, and 3) methods of sampling and data gathering.
It is shown that traditional conceptions of “policy analyst“ are not adequate for research design because a) this term doesn’t include everyone who might be doing policy work, and b) because in some countries the title simply does not exist. It is argued that: 1) there should be closer connection between theory of policy work and empirical surveys, especially in terms of clearer definition of theoretical population; 2) it would be useful to expand the operationalization of policy work in public administration to include all civil servants except those who work exclusively in the internal operations of the ministry; 3) researchers should be more transparent in describing their methodology (how the sample frame was constructed, how the units from the sample frame were chosen etc.); 4) researchers should better reflect potential weakness of research design (especially sampling bias); 5) more effort to estimate data quality should be made; 6) surveys should be accompanied by qualitative in-depth studies.
The appendix of the article provides detailed information about the methodology of the Czech Republic survey.