Assessing the validity and reliability of measurements when evaluating public policy

Abstract

AbstractA substantial aspect of scientific research involves linking concepts to observations using measurements. This exercise has raised questions among researchers of whether or not measurements “truly” and “reliably” capture ideas and observations. We address this question by setting out a methodological standard on how to assess the validity and reliability of measurements. We do this by examining measurements that evaluate public policy, arguing that this topic is gaining increasing attention from political science researchers and policymakers. The analysis concerns measurements of the level of transparency and accountability of lobbying laws, central to recent regulatory policy research. We conduct convergent validation, content validation and reproducibility tests on four indices applied to 13 regulations found worldwide. By doing so, the article provides scholars with an evaluation of measurements of lobbying laws’ robustness, while offering methodological and theoretical lessons of value to larger regulatory and public policy scholarship.