In data linkage studies, selection bias can occur if the records of certain subgroups of individuals have different linkage rates to other groups. These differences in the accuracy of linkage (‘linkage errors’) may impact the assessment of study outcomes if people from certain groups are more or less likely to have the risk factors or outcomes of interest. This chapter explores some of the common issues affecting quality in the data linkage process. The types of linkage errors discussed in the chapter are: missed matches from missing linkage variables, missed matches from inconsistent case ascertainment, and false matches. The chapter presents a narrative review of the literature describing how the limitations of data linkage can impact the quality of research findings. It provides some recommendations for practice to improve the linkage process and the interpretation of results based on linked data.