Migration researchers have demonstrated relatively limited engagement with experimental research. The strongest engagement has been with natural experiments, which observe actual migrant behaviour, typically analysing migration behaviour before and after particular events, but have limitations in terms of control over the ‘experiment’. In contrast, researchers have relatively neglected the contribution of true experiments involving the manipulation or simulation of changes in the environment or institutions that shape migration. The paper compares the relative strengths of two types of true experiments – field and laboratory – in terms of their validity and reliability, reviews their current application in migration studies, and sets out opportunities to develop future research around these methodologies. Copyright © 2015 John Wiley & Sons, Ltd.