Abstract
Mixed methods research - the combination of qualitative and quantitative data within the same research design to strengthen causal inference - is gaining prominence in the social sciences, but its benefits are contested. Social scientists and philosophers have sought to cash out the epistemic rationale of mixed-methods research but none of the available accounts adequately captures the epistemic gains of mixing methods within a single research design. We argue that what matters is variety of evidence, not of data or methods, and that there are distinct epistemic principles grounding the added value of variety of evidence for causal inference.
