How do we know Smith’s current USNews ranking doesn’t already reflect the Time’s source material? You can’t get any lower than last.
Both Time and USNWR use the same source material for earnings – CollegeScorecard (although different specific metrics within the CollegeScorecard earnings information).
USNWR’s current ranking includes 5% on earnings more than than HS grad, and no other direct earnings metics. The new ranking presumably includes 5% on earnings more than HS grad + 5% on earnings by major. The new rankings are expected to have an additional 5% weighting in earnings categories that is not present in current ranking.
So, in effect, USNews is measuring the same subject (median earnings) in twice as many ways as it did last year.
The Time Magazine pecking order goes something like this:
- Davidson - #37
- Middlebury - #89
- Hamilton - #92
- Barnard - #94
- Wesleyan - #111
- Grinnell - #215
- Vassar - #266
- Smith - #436
Just for fun, I predict that Williams holds onto #1 while there will be a new #2 in the LAC rankings with Amherst dropping a couple spots. Any other guesses or insider info out there?
Bucknell and / or Richmond ascend.
But who knows. It sounds like this metric is a tiny percentage of the overall grade.
That would be amusing.
Based on the underlying statistics, I expect Williams to place no lower than in a tie for first and Amherst to place no lower than in a tie for second.
In contrast, per major earnings figures can differ dramatically, such as by 37%, and therefore may serve as a greater distinguishing factor. This could be the case even at a modest weighting percentage.
Scrolling through the data available at College Scorecard for two very different but similarly ranked universities (UCLA and Rice), its hard to see how US News is going to make this work fairly.
UCLA being much larger has information available for a wide range of majors, many of which place surprisingly close to the overall national median earnings mark. There are some standouts in CS etc where earnings are very near the top, but many are more modestly placed. Rice on the other hand has information for a very limited number of majors, all of which are very high versus national medians, but are missing info for a majority of majors.
Not a statistician, but I dont know if you can use “statistical transformations” to make up for missing information without making some possibly unfair assumptions?
Ehhh. If they can reward a few, penalize a few - the logic is irrelevant.
Generating headlines - and PR from some and bluster from others - is what’s relevant to them. And thus staying on the ratings forefront, even if their ranking is - not necessarily a true reflection of the college landscape.
I agree it is hard not to be pretty cynical about this. No one is ever going to “peer review” their work, and I find it entirely plausible the marketing team will provide “feedback” to the statistical team until they are happy with what they will be selling this upcoming year.
In fact, this isn’t a prediction per se, but when the US News jumped on the upward mobility bandwagon a few years ago–what they are referencing when they say “produce more year-over-year movement in the rankings than either of the previous two editions,” as it was the edition before that where they made that change–on the one hand it generated a lot of news, as a variety of publics sort of gained relative to privates which realistically were in fact less committed to upward mobility. On the other, though, there was a decent amount of pushback from people who saw the change as US News as diluting its rankings as an indicator of superior institutional qualities, including academic excellence.
So I would not be surprised if this was then a swing of the pendulum back in favor of some privates with high academic reputations. Indeed, I would not be completely surprised if it swung past where it had been a few years ago.
But, who knows? All I am confident saying is that in the end, they can really make it do whatever they want it to do. And indeed, that they have already done that, and are just teasing their market with this vague preview.
Curious how they are going to manage not only missing information for some majors, but also how to compare schools with different bundles of majors. For example, many colleges and universities don’t have a film school. Some that do have produced amazing results. How does data from those schools compare to schools that don’t have that major at all? Is it only comparing it against other schools that have that same course of study? I assume that’s the only way you can deal with that discrepancy in comparability. IDK.
Yes.
If that’s the case (and that’s a bit “IF”), it will be interesting to see what happens to Swarthmore which is distinguishable from NESCAC largely due to its engineering program. Does that just get filtered out?
Then it would seem to be an advantage for the schools that specialize in courses of study that lead to great remuneration. You’re rewarded for not producing future leaders in the arts, for example, who often don’t make a lot of money in the first years of their careers in that you don’t have to worry that your artists make less than others because you don’t produce any. You just need make sure that you’re good at what you do, even if it’s narrow and limited.
Sounds like it, at least insofar as (most) other LAC comparators.
Swarthmore will be compared to schools like Bucknell, Lafayette and Union, should earnings information be available for engineering majors at all of these colleges. Swarthmore’s engineering program will be neutral (excepting a complex “statistical transformation”) when considering against those NESCACs without engineering programs.
Let the games begin!
This represents what U.S. News intends to control against. Schools prominent for, say, engineering and business will not be inherently advantaged; schools prominent for, say, film will not be inherently disadvantaged.
So in theory, if College A ranked top 10% in Film major earnings, and College B ranked top 10% in Accounting major earnings, both of those observations could have the same impact even though College A doesn’t have Accounting and College B doesn’t have Film, and Film and Accounting had different distributions of earnings such that top 10% looked different in each major respectively.
Yes, it took a while for some of us to sort that out. But I think I’m getting the hang of this now.