<p>Again, the national averages are quite irrelevant to this discussion. We’re talking about comparisons among schools that by their very nature are atypical. The schools at the top of the survey have far more than the national averages of engineering graduates. It doesn’t matter what percentage of those engineering grads go on to graduate programs; if engineering grads still make up a disproportionately large (relative to the national average) share of their graduates who never go beyond a Bachelor’s, that’s going to skew the reported lifetime earnings and ROI for that school higher. In short, it is still the case that all the data are telling us, at best (assuming away problems with the underlying Payscale data) ,is that if you’re only going to get a Bachelor’s degree, you’ll make more money in engineering than in other fields. Which we already knew. </p>
<p>Just as an example of how unrepresentative some of these schools are, here’s what Haverford says about its biology majors:</p>
<p>
</p>
<ul>
<li>This is out of a graduating class of about 300, so roughly 8-15% of the class are bio majors. Let’s assume an average of 10%, the figure reported on their latest CDS. So that’s 30 bio majors/year, of whom 9 to 12 attend medical school. If Haverford is true to national averages, an equal number of medical school admits would come from other fields, so that’s 18 to 24 medical school admits per year, or 6 to 8% of the graduating class. That is clearly much higher than the national average.</li>
</ul>
<p>I’m quite certain that the national average of biology majors who eventually attend medical school is well below 30-40%. I’m less certain what percentage go on to other graduate studies, but I’m quite confident that Haverford’s combined percentage of biology majors who end up in medical school and as matriculants in other graduate programs–ranging from 50% to 70%, according to these figures–is well above the national average. Consequently, in taking into account the value of a Haverford biology degree, it makes absolutely no sense to focus just on the minority of biology graduates from Haverford who never go beyond a Bachelor’s degree. Yet that’s exactly what this survey does. </p>
<p>It’s also interesting to note that Haverford actually does quite well in the ROI survey–#19 among private colleges and universities, despite reporting zero engineering majors. But just as with CMC, that may be more an artifact of limited/unrepresentative/unverified/self-selective-and-self-reported data than any reflection of reality.</p>
<p>Addendum: This could easily be a case where a small group of bored undergraduates or deviant alumni sitting around on a Saturday night with nothing better to do decided to “juice” the Payscale data by filling in some fictional career paths of fictional alumni. In a low-N environment it would be extremely easy to do that on Payscale, where everything is self-reported and unverified. I’m not accusing anyone at CMC in particular, just suggesting reasons why the Payscale “data” should not be taken seriously.</p>
<p>But really: how does a school go from “not enough data” in year 1, to a $764K ROI in year 2, to very nearly double that ($1.341 million) in year 3? It’s just not credible. It’s either gamed, or it’s hugely influenced by one or two unrepresentative high-earning reports (e.g., at Michael Jordan levels) trickling in. Either way, it’s sufficient grounds to toss the reported data as simply not credible.</p>
90% is much higher than I expected. I appreciate the reference. I agree that the advanced degree percentage is quite high for some colleges. It seems some selective LACs are especially high.</p>
<p>So using the above estimates gives the following totals for social sciences:
MA - 50k (NCES)
MBA - 29k (estimate based on 16% of test takers * number of MBA degrees reported by NCES)
JD - 25k (estimate based on 58% of test takers * number of JDs)
PhD - 10k (NCES)
MD - 2k (MCAT Matriculants)
Total = 116k </p>
<p>Engineering:
MS - 65k (NCES)
MBA - 32k (estimate based on 18% of test takers * number of MBA degrees reported by NCES)
PhD - 10k (NCES)
JD - 2k (estimate based on 4% of test takers * number of JDs)
MD - 2k? (MCAT groups engineering into "other)
Total = 111k</p>
<p>There are more than twice as many undergrad social science majors as engineering majors earning bachelor’s degrees, yet the totals for advanced degrees above show roughly the same number of engineering majors as social science majors, suggesting that a much larger percentage of engineering majors are pursuing advanced degrees than social science majors. I realize I am counting students who pursue multiple advanced degrees multiple times, ignoring students who do an MA/MS/PhD in a completely different field from their undergrad, and not considering that different rates of MBA/JD applicants matriculate in different undergrad fields. It’s quite unlikely that these differences would be enough to change the conclusion that a large number of engineering majors are pursuing advanced degrees, nor that a larger portion of engineering majors are pursuing advanced degrees than social science majors.</p>
I don’t dispute this. In fact I made a similar comment about the difference in Payscale ROI primarily relating to different colleges having different portions of majors in fields with higher undergrad salaries (most notably tech) in my first post of the thread.</p>
<p>The comment that triggered the percent engineering majors advanced degrees discussion implied that few engineering majors pursue advanced degrees compared to other fields, something I disagree with.</p>
<p>And do you also agree that any assessment of the value of a Swarthmore Bachelor’s degree that is based exclusively on the lifetime earnings of the 10% of Swarthmore graduates who do not go on to earn advanced degrees is pretty much worthless, a bad joke? As in, not remotely representative of how most Swarthmore graduates view, or actually use, that degree, which for them is primarily a ticket to graduate school admission?</p>
<p>The analogous methodology would be to assess the value of HS diplomas by looking only at the lifetime earnings of those HS graduates who never pursued higher education beyond a HS diploma. The 90% of graduates of an elite prep school who went on to college, and the 50% of graduates of that school who went on to top-25 ranked colleges, might well object that judging the value of their prep school diplomas by looking only at the career earnings of their slacker classmates, the 10% who never went to college, is absurd because it relies on unrepresentative data. I think they’d have a point.</p>
<p>The Payscale ROI numbers are not meaningful for a large number of reasons, which have been discussed in this thread. In my opinion the most serious problem is estimating ROI by assuming that salary differences are entirely due to the college attended with no influence from major selection, location, or the individual student.</p>
<p>I don’t have as big a problem with not considering students who pursue advanced degrees. Obviously the ROI and average salary is not accurate for that group, but grouping undergrad degrees together is necessary to make an apples-to-apples comparison. If they wanted to include advanced degrees, then they’d need to compare the same advanced degree group at different schools. For example, they might compare the average salary of BA → JDs at multiple colleges, looking at both the BA and JD school. For similar reasons, it’s necessary to group similar majors together, as has been previously discussed. I expect they don’t group by majors or advanced degrees because they don’t have enough data – either it’s not reported or they don’t have a large enough sample size. As has been discussed, the small sample size is particularly problematic for smaller colleges.</p>
<p>However, only a few majors are compared, and some are aggregations of different majors with different employment characteristics (e.g. “engineering”, “business”, “English/humanities”). They also only list what they consider the top 20 in each major or aggregation. I.e. the information provided only barely scratches the surface of what a useful amount of information of this type would be.</p>
<p>Another thing that happens at some of the top colleges is that students are recruited by businesses (including consulting firms) for jobs that have no connection to the student’s major. I guess this does reflect positively on the ROI of a degree from one of those schools, but I think it is just another example of the complexity of the situation.</p>