The spatial dimension in the definition of internal migration usually refers to the distance someone has to move to be regarded as a migrant. Lack of precise data on migration distances, however, has obliged migration researchers to use aggregate distance measures whose accuracy is largely unknown, raising potentially serious validity concerns for research. The aim of this paper is to examine the accuracy of standard aggregate measures of migration distance and to seek practical means for improving their validity. Employing uniquely detailed data where individual migration distances for an entire country's population have been measured with considerable accuracy, the paper compares variants of aggregate distance measures with the actual distance travelled by individual migrants. For the first time, the results shed empirical light on some of the weaknesses of aggregate migration distance measures and, more importantly, also point to their usefulness. The findings show that there is a significant potential to improve the accuracy of migration distance measures; practical suggestions for overcoming the difficulties of using aggregate distance measures are provided. Copyright © 2015 John Wiley & Sons, Ltd.