Cohort Analysis

Cohort analysis, a general strategy for examining data rather than a statistical technique, has become increasingly popular in the social sciences in the past few decades, especially during the last quarter of the 20th century. Useful in assessing the consequences of aging (of humans or other entities) and in understanding social and cultural change, cohort analysis has also attracted attention because it presents an unusually intriguing methodological challenge.

In its broadest sense, cohort analysis is any quantitative research that uses a measure of the concept of cohort and relates that measure to one or more additional variables. A cohort, in turn, consists of those individuals (human or otherwise) who experienced a particular event during a specified time. The kind of cohort most often studied is the human birth cohort—those persons born during a given year, decade, or other period of time. If the term cohort is used without a modifier, it is usually understood that the referent is a human birth cohort. Otherwise, the event commonly experienced by the individuals is used an as adjective to identify the kind of cohort, as in retirement cohort. The events that define cohorts may range from marriage to joining an organization, from entering a graduate program to becoming a parent for the first time. The individuals, if not humans, may be marriages, organizations, textbooks, movies, or other entities that came into being during a particular time.

The usual and more restricted meaning of cohort analysis is that it is any attempt to estimate age, period, and cohort effects on a dependent variable, such as some kind of behavior or attitudes, and such attempts require data from at least two cohorts for at least two points in time. An age effect results from growing older, a cohort effect from cohort membership, and a period effect from influences that vary through time. For example, an enduring preference for the kind of music that was popular when one was a teenager or young adult is a cohort effect, whereas changes in attitudes in response to an economic downturn are period effects. It is obviously important to estimate age effects in studies of human aging, and it is also important to estimate each kind of effect to understand social and cultural change.