This Age, Period, and Cohort Effects Essay example is published for educational and informational purposes only. If you need a custom essay or research paper on this topic, please use our writing services. EssayEmpire.com offers reliable custom essay writing services that can help you to receive high grades and impress your professors with the quality of each essay or research paper you hand in.
Age, period, and cohort are variables often used in social research that are so closely interrelated that the effects of one cannot be studied without consideration of the effects of the others. Each variable is a perfect linear function of the other two, which means that when any two are statistically held constant, the third has no variance. It follows that the effects of all three cannot be simultaneously estimated with any conventional statistical analysis – a phenomenon known as the age-period-cohort conundrum.
The age-period-cohort conundrum is important because all three variables are important for the explanation of a wide range of social and psychological phenomena. Age, the amount of time passed since an entity came into existence (by birth, in the case of human individuals), almost always needs to be an independent or control variable when human individuals are the units of analysis. Almost as important are cohort, the time when an entity came into existence, and period, the time when measurement was taken. All three are closely and causally related to a wide range of influences on human characteristics and behaviors.
To illustrate the APC conundrum, consider a hypothetical case in which a cross-sectional study of adults shows that at one point in time there was a positive linear relationship of age to support for a certain social policy. From these data alone one cannot tell whether the relationship reflects age or cohort effects, or both, because age and cohort are perfectly correlated in cross-sectional data. Now, consider panel data showing that specific individuals on average became more supportive of the policy as they grew older. From these data alone, one cannot tell whether the change resulted from period or age related influences, or both, because in panel data age and period are perfectly correlated. These two sets of data together suggest positive age effects, but they do not prove such effects because there is a logically possible alternative explanation. They could have resulted from positive period effects offset at each age level by opposite-signed cohort effects. This explanation seems rather improbable, because according to theory and some empirical evidence, most period and cohort effects on attitudes result ultimately from the same influences and thus should usually be reinforcing rather than offsetting. However, ”usually” is not ”always,” and thus a confident conclusion about age effects is not warranted without consideration of other relevant information.
This hypothetical example illustrates the importance for attempts to disentangle age, period, and cohort effects (cohort analyses) of theory and what Converse (1976) has called ”side information” -information other than the APC data at hand. Good cohort analysis is not ”plug and play” but rather requires human judgment at each stage of the process.
- Converse, P. E. (1976) The Dynamics of Party Support: Cohort Analyzing Party Identification. Sage, Beverly Hills, CA.
- Glenn, N. D. (2005) Cohort Analysis, rev. edn. Sage, Thousand Oaks, CA.