Andrew Gelman revises Ron Unz

by on February 12, 2013 at 9:32 am in Data Source, Education | Permalink

I am afraid that any excerpt would be misleading, so I will simply tell you to read the whole thing.

whatsthat February 12, 2013 at 9:48 am

more like andrew gelman’s correspondent revises

mike February 12, 2013 at 10:04 am

It’s a funny aspect of modern politically correct journalism that almost every single person who commented on the article acted as if it was about anti-Asian discrimination (which Unz explicitly said he did not find) and nobody even mentioned this section (the whole second half of the article, really) even to denounce it.

Cliff February 12, 2013 at 11:12 am

Whaaa? “Unz explicitly said he did not find”??

JL February 12, 2013 at 10:29 am

Excellent, I love this kind of fact-checking. I voiced concerns about Unz’s Jewish numbers in the original thread, noting that his methods are very indirect and that he has a dubious track record with quantitative arguments (cf. reanalyses of his Hispanic crime and IQ articles). However, knowing Unz, I predict that he will refuse to admit that he has made any substantial mistakes.

Gelman’s correspondent doesn’t actually prove that Jews and non-Jewish whites are treated equally in admissions to elite schools. Perhaps they are but we really don’t know. It would be very interesting to see actual application and admissions data on this.

wiki February 12, 2013 at 10:44 am

I would really love to see the Ivies release data so that scholars could figure out what’s really going on. For me the big takeaway from Unz’s original essay was the collapse of Asian admissions to a narrow band in the Ivies starting around 93-94.

anon February 12, 2013 at 11:38 am

I would really love to see the Ivies release data so that scholars could figure out what’s really going on.

Will never happen. Among other things, it would show how they pander to celebrities, legacies, and politicians. See, e.g., “The Price of Admission: How America’s Ruling Class Buys Its Way Into Elite Colleges — And Who Gets Left Outside the Gates” by Daniel Golden.

If you want to see what a college that admits on merit looks like, see CalTech.

charlie February 12, 2013 at 12:00 pm

The issue is jewish outmarriage. I’d say at least half of the “jews” at my little ivy had a non-jewish father and would not be found on surname sorts. Some would identify as jews, some not, but no question the pernicious effects of outbreeding are having an effect.

Something in the same is true for asians. Japanese out marriage rates are incredibly high. For non-immigrant asian (ie born here) it is also just as high. CA is an outlier in having such a high native born asian population. In addition, you have an incentive NOT to identify your asian half as a result of said quotas.

It is all the narcissism of small differences. I don’t see much evidence of asians being harmed. The real scandal, as always, is there are less than 1000 African Americans college grads a year capable of being admitted to Ivy Leagues, and probably less than 10K capable of being admitted to “elite”. Is that much better than, say, 1890? The huge number of African immigrants is also changing those numbers.

Miley Cyrax February 12, 2013 at 2:50 pm

Yes, Asians aren’t being harmed. We’re actually helping them by requiring higher test acores and grades from them to be treated the same as other people.

Ray Lopez February 12, 2013 at 12:15 pm

I think the author of the link “Andrew” and the original poster make assumptions that are hard to verify, namely, that you can tell who a “Jew” is (which is disputed even amongst Jews, not to mention the state of Israel) from a surname. Hence the author Andrew states: “I’m a little miffed the list doesn’t include Rosenthal or Gelman, but hey, what can you do?” meant as sarcasm. BTW, there are Jewish origins even in Hitler’s family tree, and of course Hitler’s chauffeur and sometime admirer of Hitler’s niece and his propaganda chief were Jewish, as well as some other Nazis. Point being: it’s really hard to tell who is Jewish. By contrast, with Asians, it’s pretty easy don’t you think? ;-)

zbicyclist February 12, 2013 at 12:25 pm

Not by checking names. Consider my neighbors: Sue Lee is white. The Gyllenhaals are half Korean.

msgkings February 12, 2013 at 2:20 pm

You don’t mean the sibling actor Gyllenhaals do you? Their dad is Swedish their mom is American Jewish.

zbicyclist February 12, 2013 at 5:05 pm

These are cousins of the actor Gyllenhaals.

So here we have it. A nice Swedish name, but along the maternal line either Jewish or Korean, and all fine Americans.

Dan February 12, 2013 at 5:04 pm

The methodology actually seems much more defensible than you describe. IIRC Unz looked at the proportion of Jews with the listed names out of all Jews, according to the latest census, and extrapolated based on that.

Anon II February 12, 2013 at 6:49 pm

Actually, Unz only used that method to confirm the validity of his National Merit Scholar semi-finalists overall percentage of Jews that he had primarily determined by his much more subjective direct inspection method. It is the correspondent to Gelman who consistently used, instead, the non-subjective Weyl list of Jewish names to determine the % Jews among undergraduates at Harvard, NMS, etc. so that the numerators and denominator would all be obtained by the exact same method, regardless of how accurate this method actually is. That is the likely reason the correspondent’s NMS numbers for individual states don’t agree with Unz’s.

crimson February 12, 2013 at 7:53 pm

Unz stated that “my estimate of Jewish NMS semifinalists based on direct inspection was 5.95%, while the estimates based on Weyl Analysis were 5.92% and 6.03%, depending upon the particular subset of distinctive surnames selected,” so while Weyl Analysis results for individual states may not agree with Unz’s direct inspection method, it is still a mystery how Unz obtained a lower Jewish % for so many states compared to the results from Weyl Analysis if overall they differed by so little.

Ron Unz February 13, 2013 at 5:05 am


Meritocracy: The Yale Debate and Surname Analysis, February 1, 2013

Meritocracy: Response to Prof. Gelman on Jewish Elite Overrepresentation, February 13, 2013

JL February 13, 2013 at 7:04 am

“Others have trusted them, too” is a very weak defense of the dubious Hillel data.

lemmy caution February 13, 2013 at 6:36 pm

The trouble is, there is no standard system that Hillel: The Foundation for Jewish Campus Life uses to count Jews. Some data comes from students who self-identify, other comes from admissions offices’ extrapolation or “guesstimate” of those who identify plus those who don’t. Other campuses rely on historical data, and, most importantly, there isn’t a uniform definition of “who is a Jew.” A students’ definition of Jewishness might be different at Yeshiva U. and U.C. Santa Cruz.

Matt February 13, 2013 at 4:21 pm

To a layperson it seems that it should be quite trivial for Gelman to resolve Unz’s main counterarguments, the incomplete application of Weyl analysis to NMS states as listed on Gelman’s blog and whether Weyl analysis is applicable to a sample of the size of “the names of Harvard students from the classes 2009-2012″ (and preferably whether such a list is actually available to cross validate).

Hopefully he shall do so rapidly.

nb February 13, 2013 at 7:53 pm

Hi Matt,

I am Prof. Gelman’s correspondent who did the Weyl analysis for NMS data for several states. I wrote multiple replies on Prof. Gelman’s blog addressing this:

I performed Weyl Analysis on the names of Harvard students from the classes of 2009-2012 in order to estimate the % of Jewish students at Harvard College in Fall 2008, at which point Harvard College’s enrollment was ~6,600. Unz performed Weyl Analysis on the names of NMS semifinalists in 25 states with a total of 12,000 NMS semifinalists, so I’ll leave it up to the reader to decide on the applicability of Weyl analysis to the Harvard alumni names. The directory of Harvard alumni names is not publicly available, though I welcome any neutral party with access to to check my numbers. Weyl analysis yielded the result that 10-11% of Harvard undergrads were Jewish in 2008 vs Hillel’s estimates of 25+% that Unz used to compare to the % of Jewish NMS semifinalists derived via Weyl analysis/direct inspection. (Unz stated that Weyl analysis gave results within 0.1 percentage point of his direct inspection method.)

Comments on this entry are closed.

Previous post:

Next post: