cohort analysis

cohort, cohort analysis

The term cohort originally referred to a Roman military unit, but it is now used to identify any group of people with a time-specific common experience, such as graduating from school in the same year, or cohorts defined by time of marriage or widowhood. Cohort analysis refers to any study in which there are measures of some characteristics of one or more cohorts at regular intervals after the defining event.

Cohort analysis as a method of research was developed by demographers and applied primarily to the study of fertility. The most common type of cohort analysis uses age-groups (birth cohorts), for example five- or ten-year age-bands, to study mortality rates. The individuals within the bands move through the ageing process together as a cohort, and often become identified as a distinct group, such as the ‘Baby-Boomers’ or ‘Ageing-Hippies’. This approach is especially common in secondary analysis , as age is a commonly recorded item of information in population census data-sets, and in data from registers and administrative records. Cohort analysis can also be applied to repeated cross-sectional survey data-sets, where samples are large enough to distinguish a number of cohorts defined by (for example) age, or year of first childbirth. Proxy cohort data or analysis is achieved by tracking the characteristics of ten-year age-groups through successive decennial population censuses or equivalent large data-sets.

The critical problem of cohort analysis is to differentiate age, cohort, and period effects. Age effects are associated with growing older; cohort effects are common to people born at the same time; period effects are due to the shared experience of particular historical events-for example the Second World War. Unfortunately there is no simple way of disentangling these. This is important substantively: because the effects are confounded, different interpretations of change are usually possible (for example, observed cultural, political, or social changes may be due to cohort effects or ageing), and conclusions have to be correspondingly tentative. When informed by sound theory, however, cohort analysis is a powerful analytic technique (see, ‘Career Opportunities in the Federal Republic of Germany: A Dynamic Approach to Study Life Course, Cohort and Period Effects’, European Sociological Review, 1986).