<p>@poetgrl:
I see class gaps widening until a disruptive event happens in the 2020’s that will lead to class gaps shrinking for the rest of our lives. Hopefully it’s not too violent.</p>
<p>Higher education will be much disrupted over the next 20 years, regardless.</p>
<p>Wow Purple Titan. You do have all the answers. Do you have rigorous numbers to back those predictions?</p>
<p>
No worries. Our militarized police will keep us all in order.</p>
<p>“The thing is, we actually know how well students of various schools do in getting in to elite professional schools or getting PhDs or winning student awards or in becoming great in their field (“American leaders”).”</p>
<p>But that’s not what I and the article are talking about. There’s no debate about the fact that the Ivies concentrate future leaders. The question is whether that advantage is equalized once the Princeton grad and the Oklahoma State grad get into Harvard Law. Is the OSU superstar at a long-term disadvantage at that point? Harvard Law probably knows. My further guess is that if Harvard Law saw much of a disadvantage, they’d be admitting a lot fewer Tier IV grads. They admit a lot of them, just not many from any one school.</p>
<p>If you look just at salaries, I wonder how much of the Tier IV disadvantage is explained by regionalism. Some of my lower-paid classmates were Tier IV grads, and they are lower paid because they went home to Alabama or Montana after HLS. They are highly successful and have their dream jobs in the locations they want, but their salaries would not compare well with NY/DC salaries, even for people doing government and not-for-profit work in NY & DC.</p>
<p>
</p>
<p>Neither would their COL </p>
<p>@Hanna: True. These salary figures don’t take in to account regional differences either. Looking at the guys only (because there may be cultural preferences leading certain groups of women to choose to be underemployed more or less), the difference between TierI and TierIV bachelors getting TierI MBAs or MDs certainly can be swamped by regional preferences/COL differences.</p>
<p>They’re bigger for JDs and PhDs, but still, $180K in Huntsville goes much further than $255K in NYC.</p>
<p>Interestingly, of men who get public school (TierIII) MBAs, those with TierIV bachelors actually outperform those with TierI bachelors.</p>
<p>@uskoolfish:</p>
<p>Read Peter Turchin.</p>
<p>
</p>
<p>The author tracked the “most competitive” and “highly competitive” schools in the Carnegie class(es) covered by each of the first 3 tiers. Apparently, only one school in Tier III (public “Research I” schools) happened to be “most” competitive when she conducted this study. More schools in Tier 1 and 2 happened to be “most” competitive. Still, 7 of 29 in Tier 1 are only “highly” competitive.</p>
<p>I suspect that if we increase the average selectivity gap between Tier 1 and Tier 3 (by excluding all the “highly” competitive schools from Tier 1 and the one “most” competitive school from Tier 3), we also will increase the income gap (even for those students who also attend T1 grad schools). By how much? I don’t know. Maybe not by much at all. </p>
<p>@tk21769 :</p>
<p>You suspect, but you actually have no idea. And that’s because the author dumps a load of schools below “highly competitive” in to TierIII but only a handful of those schools in to TierI.</p>
<p>The “most competitive” and “highly competitive” schools could actually have <em>no</em> difference and the averages could still come out the way they do.</p>
<p>OK, let me illustrate by example:</p>
<p>Suppose that you have 3 types of students divided in to 2 sections of a class.</p>
<p>Type A and Type B both get 100% on their test. Type C get 80%.</p>
<p>First section has 18 Type A students, 14 Type B students, and 8 Type C students.</p>
<p>Second section has 1 Type A student, 10 Type B students, and 48 Type C students.</p>
<p>The first section would have an average test score of 96%.
The second section would have an average test score of 83.7%.</p>
<p>Yet Type B students got the exact same test score as Type A students. </p>
<p>You understand the statistics behind my reasoning now?</p>
<p>BTW, if you look at page 7, you’ll notice that the Barron’s tiers are pure obfuscation. She’s tiering solely using Carnegie and by private/public . . . . which Carnegie doesn’t do (there are 7 public Carnegie I LACs which she excludes from TierII and relegates to TierIV). The whole way she tiers reeks of an agenda.</p>
<p>The comments by Chris Avery of the Kennedy School in the NYT article are interesting and may go a long way to explain any disparity Hersch’s finds. (Emphasis in the below quotes is mine.)</p>
<p>
</p>
<p>
</p>
<p>In other words, much of the disparity may be in connections and in the ability to navigate the upper echelons of business, law, medicine or academia. A brilliant but blue collar kid who attends Directional State U. Is going to be at a disadvantage, but then again so will one at Harvard, the difference being that Harvard will have not only a lower percentage of blue collar kids, but a much, much high percentage of high flyers- kids with the social capital and family connections to find opportunities that may not be available to their equally capable and accomplished peers.</p>
<p>@Sue22:</p>
<p>Yep. Harvard would help with development of an understanding of the acceptable social mores while directional U would not. That’s consistent with Dale&Krueger as well.</p>
<p>I still haven’t seen evidence to dissuade me of the notion that going to an elite private (vs. elsewhere) for undergrad would help most those who come from disadvantaged/URM/immigrant backgrounds that are far removed from present-day American upper-middle-class norms as well as brilliant slackers who get in while it really doesn’t matter (outside of a few career paths) where a driven upper-middle-class kid who’s at least motivated & confident enough to apply to elite privates chooses to go to school.</p>
<p>
</p>
<p>Krueger and Dale (2011):
“For black and Hispanic students and for students who come from less-educated families (in terms of their parents’ education), the estimates of the return to college selectivity remain large, even in models that adjust for unobserved student characteristics.” (<a href=“http://www.nber.org/papers/w17159”>http://www.nber.org/papers/w17159</a>)</p>
<p>
</p>
<p>She doesn’t “dump” schools below “highly competitive” into either tier. She deliberately constructs 4 groups such that “the share of schools rated by Barron’s as most or highly competitive is significantly different between groups” (p.6). Isn’t that the whole point? T1 > T2 > T3 > T4, for the most part, for both selectivity and earnings. So I would expect that, if we re-define these groups to increase the selectivity difference, we also would increase the earnings difference (even after controlling for ability in the way she does). However, the trend may be non-linear (maybe starting off flat with a sharp take-off at some point pretty far down the selectivity scale, which would not be inconsistent w/Krueger & Dale). Indeed, as PT suggests above, there may be no difference between “highly competitive” and “most competitive”. </p>
<p>
</p>
<p>However, the construction of the tiers starts with the assumption that 40 private RUs > 159 private LACs > public RUs > whatever she did not think of to put in the other tiers. So Syracuse (tier 1) > Amherst (tier 2) > Virginia (tier 3) > Dartmouth (tier 4) in her tiering. Granted, that tiering may reflect popular belief among many in this forum, but even those who think private is always better than public will likely find some issues with her tiering.</p>
<p>If her point was to make claims about selectivity, then she should have arranged the tiers by selectivity, not by public/private or RU/LAC. And then she would have isolated the schools treatment effect from its selection effect by comparing cohorts of students with similar credentials who chose more or less selective schools.</p>
<p>Take a statistically significant number of students from the same socioeconomic backgrounds with substantially similar grades and test scores who were admitted to the same colleges as each other to study the same majors. Sort by which college they actually choose to attend, correcting for the effect their net cost has on their college selection. In other words, control as much as possible to isolate the effects of their choice of college. Then follow them over their entire lives and see if there is any measurable impact of choice of college on their careers, family, health, income, etc.</p>
<p>That is a study I would be very interested in seeing.</p>
<p>The study FCCDAD describes is similar to the Krueger and Dale studies, but with additional constraints. Hersch challenges K&D because they did not study students of similar ability whose choices varied enough in selectivity. I think it would be very hard to find enough students who were (a) admitted to top schools, and (b) chose to attend much less selective schools, while satisfying all other constraints AND tracking them over their entire lives. </p>
<br>
<br>
<p>That is one reason why I am fascinated by the UT vs. Texas A&M comparison.
Not having attended either - I still find it interesting to look at the numbers</p>
<p>UT and A&M have similar cost (A&M slightly cheaper)
They are almost exactly the same size, and in the same state (less than 2 hours apart)
UT has somewhat higher test scores (25/31 vs 23/29)
They have similar percentage engineering students (14% A&M vs. 12%)
UT has much more selective students by class rank
Graduation rates are very similar.
A&M students have higher long term salaries. Undergraduate outcomes seem as good generally.</p>
<p>That makes no sense unless Texas A&M is doing some things very well</p>
<p>I don’t have time to read through the replies, so I apologize if this is redundant, but this is the line in the article that seems most important to me. Speaking of undergraduates at Tier 1 schools:</p>
<p>“Their children are instructed from a young age not only in academics but also in the social nuances of making contacts and building relationships.”</p>
<p>I think this is a huge reason why the undergraduate degree appears more important than the graduate degree in this study. If undergraduates were closely connected to a professional mentor, perhaps this trend could be reversed.</p>
<p>@2018RiceParent :</p>
<p>Where are you getting your data from? When I look at Payscale’s 10-19 years out salaries for alums of both schools, I see an average of $93.4K for UT-Austin and an average of $90.9K for A&M:</p>
<p><a href=“University of Texas (UT) - Austin Salary | PayScale”>University of Texas (UT) - Austin Salary | PayScale;
<a href=“Texas A&M University - Main Campus Salary | PayScale”>Texas A&M University - Main Campus Salary | PayScale;
<p>Both UT and A&M are in the same Barron’s selectivity category (category 3).
FWIW, they are in the same Carnegie class (Research I). Both are public, too.</p>