We have chosen this topic because dealing with data quality in the health sciences is characterized by an impressing paradox. On the one hand, achieving a high data quality is regarded as being essential for valid scientific inference. On the other hand, inconsistent attention has been paid to the definition, assessment, and reporting of data quality. Corresponding standards are mostly lacking, despite numerous theoretical and applied works of relevance. To make matters worse, neither the evaluation, nor the resulting data quality of data bodies are commonly transparent.

This workshop therefore consists of two parts:

First, a theoretical part on definitions of data quality and data quality frameworks. Second, an applied part on tools, statistics, and application examples to assess data quality and general data properties.

Presentations cover the handling of data quality in research data collections as well as in electronic health records and other sources that have not been primarily collected for research purposes. We cover assessments in ongoing data collections as part of a data monitoring as well as assessments between the end of data collections and the start of substantive statistical analyses. The latter is commonly referred to as initial data analyses (IDA), which specifically assesses the suitability of some data body for a defined research purpose.

Beyond invited theoretical and practical lectures we welcome contributions of participants to the above mentioned topics.