Quantification of qualitative data using ordered probit models with an application to a business survey in the German service sector

This paper aims at providing business survey analysts with simple econometric tools to quantify qualitative survey data. We extend the traditional and commonly applied method proposed by Carlson and Parkin (1975) to capture observable survey respondent heterogeneity. We also discuss specification tests. The empirical analysis is based on business survey data taken from the ZEW's 'Service Sector Business Survey', a quarterly business survey in the German business-related service sector carried out since 1994.