We Need to Compare, But How? Measurement Equivalence in Comparative Public Administration
Public Administration Review, 75(1): 36-48, 2015
43 Pages Posted: 15 Oct 2014 Last revised: 15 Jul 2015
Date Written: 2015
There is an increasing interest to study public administrations, public managers or citizens interactions with, and views towards government from a comparative perspective in order to put theories to test using cross-national surveys. However, this will only succeed if we adequately deal with the diverse ways respondents in different countries, and regions perceive, and respond to survey measures. In this article, we provide an examination of the concept of cross-national measurement equivalence in public management, and how to proceed in establishing equivalence. We examine two different methodologies that test, and correct for measurement non-equivalence, namely 1) multiple group confirmatory factor analysis, and 2) item response theory. These techniques are then used to test, and subsequently establish the cross-national measurement equivalence of two popular measurement constructs, citizen satisfaction with public services, and trust in public institutions. Results show how appropriately dealing with non-equivalence accounts for different forms of biases, which would otherwise stay undetected. In doing so, this article contributes to the methodological advancement in studying public administration beyond domestic borders.
Keywords: Citizen satisfaction, institutional trust, comparative public administration, item response theory, measurement equivalence, multiple group confirmatory factor analysis
JEL Classification: D12, D18, D4, D63, D8
Suggested Citation: Suggested Citation