ABSTRACT. Discrete‐choice theory models, and more specifically the logit model, have been applied widely recently in analyzing migration patterns in a zonal system. We argue that spatial effects and more specifically the relative location of zones are not taken into account in such analyses. We, therefore, introduce a measure of spatial structure and advocate its usage as a predictor of migration in such models. In an example of intrametropolitan migration in Toronto, we demonstrate that this variable is not only significant but also it improves the performance of all the other variables with the greatest impact on the distance between zones. In addition, inclusion of this variable improves the overall performance of the model in terms of residuals. Copyright © 1988, Wiley Blackwell. All rights reserved

[No Keywords available]

evaluation residential mobility Research Methodology Canada, Toronto logit model intrametropolitan migration Evaluation Methodology Americas spatial structure discrete choice models population methodology demography Research Design developing country Population Dynamics Developing Countries statistics Migration, Internal measurement Developed Countries spatial distribution North America geography Canada Western Hemisphere theoretical model Models, Theoretical Article rural-urban migration migration Geographic Factors developed country Demographic Factors research Estimation Technics Emigration and Immigration Evaluation Studies Distance Northern America Health Services Accessibility health care delivery