display | more...

The cohort effect is the effect on a person's life based on the time period in which they lived, or grew up. For example, if you came of age in the 1940s, then you were less likely to have premarital sex than those who came of age in the 1990s, however you were more likely to have polio.

A cohort is an age group, like a generation. If you are 21 right now then you are of the 1980/84 cohort. The cohort effect is the effect on you because you are of the 1980/84 cohort. For example, you may have taken ketamine as a recreational drug, and you probably never had the measles.

In statistics, primarily medical statistics, the cohort effect is important when looking at data over a long period of time. If statisticians detect a certain variable, such as ecstasy use, rise substantially during a period of time, then they can assume those coming of age during that time period are under the cohort effect when it comes to that variable. Also, taking the cohort effect into account, a statistician would not consider those who grew up in the 1940s to be healthier because they didn't use ecstasy.  The difference is that the age group who came of age in the 1940s was under a different cohort effect than those in the 1990s.

Although statisticians define a cohort age group by their birth years, studies involving the cohort effect do not always focus on adolescents of the era. The cohort effect of a certain aspect can include the dying age group, the infant age group, or the elderly age group. For example, one study has shown that the 1920/24 cohort in the Netherlands had a considerably high mortality rate while they were in their 40s and 50s (during the 1960s).

Sometimes statisticians shrug off data, citing the cohort effect. For instance, we would not assume that high school students in the 1973 were more able to deal with internet addiction than students in the 2003 because the internet was not available to the 1955/59 cohort, unlike the 1985/89 cohort.

see: epidemiology



Log in or register to write something here or to contact authors.