It has been suggested that parish-based reconstitution studies may underestimate the true age at marriage because they do not normally include data for emigrants who may be expected to behave differently from individuals who remain in their natal parishes. This study uses data from C18-19th parish registers of north-west Germany to estimate the difference in age at marriage between leavers and stayers. The difference is not significant for males; although that for females is significant, it is small and the consequence of failing to include migrants is likely to be negligible for most studies. However, it is shown that there is also an independent effect on age at marriage that is due to the woman's natal social (economic) status; historical demographic studies that ignore this dimension may risk confounding two different effects.