How can data from different sources be combined to improve the reliability of the dataset to produce robust results in animal welfare impact assessments?

There is a general consensus that improved animal welfare (AW) benefits both farmed animals and humans. However, the types of animal welfare measures that can be implemented at different stages of the value chain vary considerably. In order to reliably assess their socio-economic and environmental impacts, an appropriate dataset is essential. This study aims to report on the process of data collection for the assessment of animal welfare strategies in the broiler and pig production chains. Therefore, this study details the identification and validation of indicators, the development and validation of questionnaires, and the data collection process on different stages of the production chain. The study contributes to the debate on how to combine data from different sources to obtain a reliable dataset. The data collection strategy is illustrated by an intervention study on how weight sensors in pig fattening can alert possible diseases and avoid additional costs.