We examine the measurement of country-level institutions in international accounting research. Despite a large body of literature linking national institutions to financial reporting outcomes, the construct validity and reliability of the institutional measures used in cross-country studies have rarely been assessed. We show that widely cited measures of investor protection, legal quality, and equity market development have been adopted without validity or reliability assessment and used inconsistently across published studies, with the same indicator labelled as a measure of different constructs in different papers. Using forty-eight candidate indicators from six international databases for seventy countries, we conduct an exploratory factor analysis followed by a confirmatory factor analysis and identify three latent constructs that pass conventional thresholds for convergent reliability and discriminant validity. The constructs are empirically distinct, indicating that measures routinely treated as interchangeable in published work measure different latent dimensions. We offer a validated three-factor measurement model that researchers can use to operationalise country-level institutions in future cross-country studies and discuss the implications for the interpretation of existing studies.
Alsabah et al. (Fri,) studied this question.