The change in the intelligence of the average college graduate over the last fifty years approaches the IQ gap separating whites and blacks.
This is an inevitable consequence of increasing the share of the population that attends college. In the sixties, 10% of American adults had college degrees. Since then that figure has more than tripled, to 33% today.
To say we're well into the territory of diminishing returns is to understate the problem--we're past the point of negative returns. Most Americans in college today are not benefiting from being there. They're foregoing work to accrue debt for degrees that, if they increase earning power at all, do so only marginally and they're picking up an unhelpful sense of entitlement in the process.
GSS variables used: COHORT(1940-1949)(1950-1959)(1960-1969)(1970-1979)(1980-1989)(1990-1999), EDUC(16-20), WORDSUM, BORN(1)
* Values for each decade come from those born two decades prior, so the time of actual graduation is approximate. For example, the result for the 1960s comes from the wordsum scores of those born in the 1940s; the result for the 1970s from those born in the 1950s, and so on. The approach isn't perfect--some people graduate later in life and a few while still in their teens--but it is an improvement on previous approaches.
**Update** Restricting the age of those evaluated only very marginally lowers the mean wordsum for the earlier cohorts (less than half of 1 IQ point on average).
Also, to reiterate, this measures respondents by total number of years spent in school. There are some--more now than in the past, presumably--who spend eight years in college without ever actually getting a degree.