Monitoring the metropolitanization process

Abstract
Alternative approaches have led to different interpretations of the metropolitanization process in the United States. We identify and illustrate several methods and procedures for monitoring metropolitan-nonmetropolitan population change using the 1950–1980 U.S. decennial censuses. Two basic approaches are compared: constant area approaches and component methods. In addition, we assess the effects of changing metropolitan definitions on metropolitan-nonmetropolitan growth. The results clearly reveal that the underlying mechanics of metropolitanization not only are complex but have changed substantially during the 1950–1980 period. We conclude with observations regarding the use of these procedures in future research.