Data quality is a critical factor in scientific information systems, especially taking into account the fact that the methods used to capture data are constantly being revised and improved which means that data collected over time may have variable quality. We present an approach that we implemented for giving feedback on data quality and report on a study of its use in the food sciences. We distinguish between two main types of data quality feedback, one concerning the validation of data at input time and the other with analysing the quality of data already stored in a database. We propose a general data quality framework and analysis toolkit that allows users to configure both the data quality metrics and how these metrics are visualised. We describe how the toolkit was integrated into a system for the management of food composition data before presenting the results of a questionnaire-based study used to evaluate both the data quality framework and how feedback on data quality is presented to the users.