101. COHORT ANALYSTS' FUTILE QUEST: STATISTICAL ATTEMPTS TO SEPARATE AGE, PERIOD AND COHORT EFFECTS.
- Author
-
Glenn, Norval D.
- Subjects
AGING ,COHORT analysis ,DEMOGRAPHY ,AGE groups ,LONGITUDINAL method ,SOCIAL sciences - Abstract
The article presents statistical attempts to separate age, period and cohort effects. The great bulk of the literature on cohort analysis is devoted to methods or attempts to separate statistically age, period and cohort effects. The effects are named for the kinds of influences which produce them; age effects are produced by influences associated with the aging process, period effects by influences associated with each period of time and cohort effects by influences associated with membership in each birth cohort. Regardless of how cohort data are examined, two kinds of effects are confounded with one another. Age and cohort effects are confounded in cross-sectional data by age, age and period effects in intra-cohort trend and, period and cohort effects in trend data for each age level. Space limitations preclude the author's dealing with all of the literature reporting statistical attempts or methods to unconfound these effects. The author's caution that their method has limitations, but the cautions are vaguely worded and the most important of them is buried near the middle of the article.
- Published
- 1976
- Full Text
- View/download PDF