Cohort Analysis of Survey Data: A Procedure for Studying Long-term Opinion Change

Two or more surveys taken a number of years apart may show a trend in opinion or at least a major change. However, such a change may be due to a shift in opinion among the members of a population, a change in the composition of the population itself, or a combination of these two types of change. An attempt to disentangle the two types of change is presented in the following article. The author is Assistant Professor in the Department of Sociology at Columbia University. This study is related not only to the development of panel procedures at Columbia but also to the author's studies of the engineering, legal, and other professions.