This type of reference is usually based on the Dale and Krueger studies or others that tried to replicate findings. The full abstract of the original 2002 study is below:
https://cdn.theatlantic.com/static/mt/assets/business/dalekrueger_More_Selective_College.pdf . Note that when the study controlled for average SAT score selectivity of colleges students applied to, not the one they attended, the increase in earnings for highly selective colleges became insignificant (after controls). Applying to selective colleges appeared to be more relevant for future earnings, rather than being accepted to one or attending one (in addition to other controls including individual student score, gender, race, .SES,… ). This might relate to the type of person who applies to highly selective colleges tending to be the type of person who also is more likely to target career fields associated with a higher salary.
“Estimates of the effect of college selectivity on earnings may be biased because elite colleges admit students, in part, based on characteristics that are related to future earnings. We matched students who applied to, and were accepted by, similar colleges to try to eliminate this bias. Using the College and Beyond data set and National Longitudinal Survey of the High School Class of 1972, we find that students who attended more selective colleges earned about the same as students of seemingly comparable ability who attended less selective schools. Children from low-income families, however, earned more if they attended selective colleges.”
The author notes that low income families were an exception and tended to have a small benefit from attending a selective college beyond what could be explained by the controls. He writes the following. I suspect much of this effect relates to some additional missing relevant controls that impact lower income groups more than others, such as net price to families after FA grants.
"For example, based on the self-revelation model in column 3 of Table VII, the gain from attending a college with a 200 point higher average SAT score for a family whose predicted log income is in the bottom decile is 8 percent, versus virtually nil for a family with mean income. "
Also note that this is an older study, and a lot has changed since the class of 1972. They repeated the study in a somewhat newer sample at https://www.nber.org/papers/w17159.pdf , which found similar results.
My more recent dataset example from the NYT tax sample shows relatively little difference in earnings between rich and poor at both elite and non-elite, well known colleges. Some examples of other colleges mentioned in this thread are below. There are significant differences in income at the different colleges, but those differences occur for both rich and poor students. For example, a poor kid at MIT averages 23 percentiles higher than a poor kid at Wayne state, while a rich kid at MIT averages 24 percentiles higher than a rich kid at Wayne state. The difference in earnings is similar for both rich and poor students.
I expect the bulk of this 23-24 percentile difference between MIT and Wayne State relates to individual student characteristics, rather than attending the school itself. One such characteristic is student quality, so I listed ACT score. Students with higher ACT scores tend to have higher average earnings, and MIT has more high ACT score students than Wayne State. Student quality can also be a barrier for certain higher salary fields, such as pursuing a MD. As highlighted in my earlier post, one of the most influential factors is college major and planned career field. MIT likely has a larger portion of students pursuing CS/engineering and other fields associated with a higher salary than Wayne state. I could list many other relevant factors.
Average Percentile Income at Age 34
MIT: Poor student: ~76th Rich student: ~82nd – ACT = 34/36
Ivies: Poor student: ~72nd, Rich student: ~78th – ACT = ~32/~35
U of Michigan: Poor student: 69th, Rich student: 76th – ACT = 30/34
U of Illinois: Poor student: 65th, Rich student: 75th – ACT = 26/32
UNC: CH: Poor student: 63rd, Rich student: 71st – ACT = 27/33
Wayne State: Poor student: 53rd, Rich student: 64th – ACT = 21/27