Using a dataset from dairy farms in Germany that combines two types of welfare measures, namely welfare quality protocol (WQP) measures and production economic and herd-management data, this study aims to validate the use of production economic and herd-management data to proxy dairy cow welfare measures. The paper implements two multivariate estimation approaches of Seemingly Unrelated Estimation and Canonical Correlation Analysis. Data from on-farm animal welfare assessments based on WQP require time intensive collection and are typically unavailable for research based on large-scale panel datasets. On the other hand, survey data on production economic and herd management are available for such analysis, especially in European countries, but their informational value regarding animal welfare is debated. In this paper, we were able to establish relationships between the four WQP principles (feed, health, housing, behaviour) and variables from production economics and herd-management data. We find that concentrated feed, building costs, cell counts, milk fat content, calving intervals, and age at calving have strong links to the different principles of the WQP measures. In conclusion, our findings support the use of already existing and routinely collected production economic and herd-management data from dairy cows to enable an analysis of farm animal welfare on a larger scale using panel data. (c) 2022 The Authors. Published by Elsevier Ltd on behalf of Institution of Chemical Engineers. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).