Text above is from @nxthompson, writing at the Washington Monthly before he jumped to the New Yorker and then took the helm at Wired magazine:https://washingtonmonthly.com/2000/09/01/playing-with-numbers/ …
Show this thread
Always remember: When Harvard, Yale & Princeton didn't top early US News #BestColleges rankings, the statistician was replaced & algorithm adjusted.pic.twitter.com/rUKWlUvdEj
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This year US News says #1 Princeton #2 Harvard #3 Yale (also #3: MIT, Chicago, Columbia)pic.twitter.com/pTxVvlPSsg
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Can I ask, where is this from?
New conversation -
-
-
1/ “According to sources close to US N&WR, a bitter internal struggle broke out when it became clear that
@Caltech was going to come out on top in 1999 after the rankings had been changed (based on NORC stats advice)” original algorithm from ‘88 was also p-hacked to push Ivies up -
Are college rankings gamed? Of course! “The introduction of U.S. News’ category of “percentage of alumni who give” also significantly affected fundraising...At one West Coast college, I was told, alumni who have not given money in five years have been reclassified as dead”
- 1 more reply
New conversation -
-
-
I hate this BS. So many outstanding universities, in some cases whom excel over these schools, are diminished in significance for nothing other than branding obsessions. It's also counter to the purpose and exercise of higher education.
-
You've actually alluded to the concerns in my remarks—such a dilemma is in part, the product of a problematic ranking system that elevates a small handful of colleges to serving as *the* passport to desirable opportunities.
- End of conversation
New conversation -
-
-
The "Weapons of Math Destruction" book by
@mathbabedotorg has a whole chapter on this. The US News college ranking fiasco has contributed significantly to the inflation in college costs as schools chase the ranking.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The baseball metaphor actually gets at a really fundamental question in these statistics. A metric that tells you nothing new is useless. A metric that disagrees with established wisdom on everything is probably wrong. The sweet spot is something like 80% known, 20% novel.
- 1 more reply
New conversation -
-
-
This is one of the many reasons why pedigree is not the right metric to hire people.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The baseball analogy is perfect. That's the plot of Moneyball, right there. I'd like to know which colleges really ARE that .220-hitting utility MVP! c
@AthleticsThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Ideally I'd like to see two measures: "value add" and "reputation". The first would measure the extent to which graduates of an institution outperform what their grades, test scores, parents' income and parents' education would otherwise predict. Perf. metric TBD. 1/2
-
The second would measure the signal strength of a given institution's degree. Regardless of what I actually *learn* at a school, I might want to know how much cachet a degree from a given institution carries. 2/2
-
Yes, had something like this in mind. To truly do apples-to-apples, though, you'd need to do some sort of cost-of-living adjustment based on where graduates settle, and take into account the mix of degrees awarded by each school (e.g. MIT is going to be very STEM heavy).
-
Without digging too deep into the paper, looks like degree-mix may have been accounted for.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.