Fixing the race, ethnicity, and national origin questions on the U.S. Census

question6

In his new book, “What is Your Race? The Census and Our Flawed Efforts to Classify Americans,” former Census Bureau director Ken Prewitt recommends taking the race question off the decennial census:

Screen Shot 2013-08-10 at 10.48.52 PM

He recommends gradual changes, integrating the race and national origin questions while improving both. In particular, he would replace the main “race” question by a “race or origin” question, with the instruction to “Mark one or more” of the following boxes: “White,” “Black, African Am., or Negro,” “Hispanic, Latino, or Spanish origin,” “American Indian or Alaska Native,” “Asian”, “Native Hawaiian or Other Pacific Islander,” and “Some other race or origin.” Then the next question is to write in “specific race, origin, or enrolled or principal tribe.”

Prewitt writes:

Screen Shot 2013-08-10 at 10.56.05 PM

His suggestion is to go with these questions in 2020 and 2030, then in 2040 “drop the race question and use only the national origin question.” He’s also relying on the American Community Survey to gather a lot of the demographic information that is useful for so many purposes.

108 thoughts on “Fixing the race, ethnicity, and national origin questions on the U.S. Census

  1. To me, no “fix” that fails to acknowledge the the 1.6 billion South Asians are different from the 1.6 billion East Asians is the least bit credible. Those are markedly different origins.

    (Note that “brown” can refer to Latin American or to South Asian, but not East Asian.)

    And if we are asking about origin, why don’t we ask about when that was? I would love the census to focus more on how long a person’s family has been in the US. (most recent? first? median? I don’t know. something.)

    ***********************************

    This “fix” does not really address problems with the construct. Are we talking about color? origin? ethnicity? language?

    • South Asians used to be classified with Caucasians/Whites (as Afghans, Persians, Arabs, and Turks still are) until the Reagan Administration in 1982 responded to requests from Indian immigrant businessmen to make them eligible for minority business development loans and for racial preferences on government contracts by merging them into the Oriental/Asian category.

      Old fashioned physical anthropology and new-fangled DNA studies both say that the old classification system is more scientific than the one introduced by Reagan. Also, it sure doesn’t seem to have helped Republicans win the Indian-American vote.

      • Which way do the Indian-American vote? Anecdotally, I feel the lot that are businessmen or executives drawing high salaries seem to vote more Republican. What do the stats show?

        • Rahul,

          Think of Indian-Americans as being like Jews. I mean, generally. That’s probably your best first guess in anything, until you find more information.

          In fact, Indian-Americans are generally quite well educated — due to our immigration policies in the 60’s and since. There is a big kernel of truth in the stereotypes and jokes about their being doctors and engineers. They have tended more towards the professions than finance or business. (In my view, that’s a product of the values of their families, and we selected for education on the way in.)

          So, it is not surprising that they tend to vote for the Democratic party, for that reason alone. The tend to be higher income and higher education. Gelman’s works tells us what that means. Additionally, many are often mistaken for being black or hispanic, based upon their skin color.

          The hundreds I know personally (my wife takes me to a LOT of weddings) are quite aware that they are easily identified as minority, and do not think that the GOP is interested in them. Of course, the ones I know do not constitute a random sample and might not be representative of the larger Indian-American population.

          From what I recall, and I don’t have any stats handy, Indian-Americans vote pretty heavily Democratic. Like in excess of 60%, and I think in excess of 75%.

      • Steve,

        Are you suggesting that the Reagan administration made this change simply to curry political/electoral favor with Indian-Americans?

        Are you suggesting that, for social purposes in our society, that Indian-Americans are not a minority group?

        • Yes, that’s the history of the Reagan Administration’s change. Indian immigrant businessmen wanted to qualify for minority business development privileges with SBA low interest loans and government contracting, so they wanted to be reclassified out of the Caucasian majority. The same year, the Reagan Administration classified Hassidic Jews as eligible for minority privileges.

          “Are you suggesting that, for social purposes in our society, that Indian-Americans are not a minority group?”

          They appear to be the highest income ancestral group that the Census Bureau measures. (There are other unmeasured groups that might be higher, such as Episcopalians or Reform Jews, but the Census Bureau never asks questions about religion.)

          If the government awards money and prizes to people who insist they belong to a minority, it’s hardly surprising that more people will insist they belong to a minority.

          Obviously, blacks are a minority, but there are huge numbers of people in America today, such as South Asians and Hispanics, who could go either way. In the past, their representatives tended to insist they were more or less white, but after the Nixon Administration codified affirmative action, their lobbyists insisted they were minorities.

          Hispanics, like Nixon’s best friend Bebe Rebozo, got a special deal: they could identify as white to mollify their Castillian racial pride, but still benefit from minority development privileges on the grounds that they were the one ethnicity that the government cared about. (Thus, everybody else in America gets lumped into the non-Hispanic category.)

        • Steve,

          I am going to try to keep this focused. That is, I am going to try to keep this on census questions, their history, use of the data they provide and other census question related things.

          If you think that I’ve dropped any of your points on this issue, please let me know. If you would like to continue discussion of other points, we can do that elsewhere. (You might let me know by twitter, in both cases, so we do not waste everyone else’s attention on such flags.)

          1) You seem to believe that the reason these Ethnicity question was added — and for changes to the Race question — has been the desires of minority elites to get access to set asides and preferences. (Obviously, you should correct me if I am wrong about that.) Do you think that this is the ONLY reason? The major reason, in addition to many smaller ones? One of many major reasons, in addition to many smaller ones? Or one of a constellation of reasons?

          2) You mentioned the Reagan administration’s desire to…I’m unclear why the administration went along with this. Can you explain how this fraction (i.e., “Indian immigrant businessmen”) of a fraction (South Asian businessmen in the US) of a small faction (South Asians in the US) that has voted to Democratic over the years was able to convince the Reagan Administration to make this change?

          3) My guess is that you disagree with minority set-asides and all forms of affirmative action. Is that correct? Is that objection the entire root of you problem with Race and Ethnicity questions on the census (apart from questions about their legality, I mean). in the absence of such programs, would you be ok with Race and Ethnicity questions? (Let’s assume we’re talking about less problematic questions.)

          4) Putting aside your objections in #3, for a moment. Work with me. Would you think that given these sorts of programs that using the census to count minorities’ and ethnicities’ distributions in the country is an inappropriate use of the census?

          5) Is your objection to the use of race and ethnicity questions on a census as a rule? Or is your objection because these are problematic categories? Or is it that the force behind (past and perhaps future) writing/including them is itself so problematic?

          I have a lot of question which, in another forum, I might ask about clarifying your objections to minority protections, set asides and “preferences,” but they are not appropriate here, as they do not link to the census question issue directly.

          (And more in response to your various comments shortly, If I find things on point to respond to.)

        • “1) You seem to believe that the reason these Ethnicity question was added — and for changes to the Race question — has been the desires of minority elites to get access to set asides and preferences. (Obviously, you should correct me if I am wrong about that.) Do you think that this is the ONLY reason?”

          There are a lot of reasons.

          A major reason that Nixon pushed for codifying racial and ethnic preferences was because “minority capitalism” (e.g., racial and ethnic preferences for minority businessmen) was a major plank of his 1968 campaign. Here’s a 1973 New York Times wire service article on how Nixon politicized his Minority Capitalism campaign:

          http://news.google.com/newspapers?nid=1964&dat=19731206&id=LXkyAAAAIBAJ&sjid=W7cFAAAAIBAJ&pg=1270,3421121

        • And here’s an informative book chapter on how Nixon’s “minority capitalism” campaign came to focus less upon blacks and more upon Hispanics:

          http://books.google.com/books?id=5VdIdVMJ_5AC&pg=PA144&lpg=PA144&dq=nixon+%22minority+capitalism%22&source=bl&ots=roDmXjb5ix&sig=_enXGssVtnmJ-lhEHop_wpyJyVY&hl=en&sa=X&ei=RGMNUsf3FsuJiwKNjoGYBQ&ved=0CD8Q6AEwAw#v=onepage&q=nixon%20%22minority%20capitalism%22&f=false

          The reparations theory to justify Nixon codifying preferences for blacks, who were overwhelmingly the descendants of American slaves, made a lot of sense. Reparations made less sense for Hispanics, who hadn’t been slaves, had suffered less severe discrimination, and some of whom, such as Nixon’s best friend Bebe Rebozo, were members of the American elite. But Nixon thought he had a better chance to converting Hispanics than blacks to voting Republican.

        • “2) You mentioned the Reagan administration’s desire to…I’m unclear why the administration went along with this. Can you explain how this fraction (i.e., “Indian immigrant businessmen”) of a fraction (South Asian businessmen in the US) of a small faction (South Asians in the US) that has voted to Democratic over the years was able to convince the Reagan Administration to make this change?”

          It seemed like a good idea at the time to the Reaganites.

          South Asian business owners seemed like “natural Republicans,” so why not buy their eternal support by doing them this obscure little favor? After all, Nixon’s “minority capitalism” had helped cement Cuban support of the GOP, so why not buy off South Asian leaders by this tiny technical change of reclassifying them from white to Oriental? Sure, it didn’t make much sense scientifically, but politics isn’t about science, it’s about doing your potential friends favors so they’ll do you favors.

          Sure, from the perspective of 2013 it seems like a big mistake for the GOP — putting the most naturally conservative set of South Asians on the racial preference gravy train just strengthened their determination to insist upon their minority status and thus support the Democrats as the affirmative action party.

          But the Reagan Administration’s error was in not being cynical enough. Back then, Reagan could count on Vietnamese and Taiwanese voters supporting him in California for anti-Communist reasons. It didn’t appear all that likely that the future main division in American politics was the core of the electorate versus all the fringe groups? Surely, an Obama-like coalition would naturally crack up as the disparate competing Democratic interest groups wound up at each other’s throats. Reagan enjoyed vast success employing divide and conquer strategies against the old FDR coalition. Thus, an Obama-style coalition would seem like a house of cards easily knocked over.

          What the Reagan Administration didn’t adequately foresee was something very few analysts today have fully come to grips with: that the centrifugal aspects of an Obama-style coalition of blacks, single white women, gays, immigrants, and so forth could be held together, at least for awhile, by stoking resentments that the various fringe groups of the electorate hold against the demonized core (the core being well epitomized by Mitt Romney: white rather than nonwhite, straight rather than gay, married rather than single, successful rather than unsuccessful, and so forth and so on). The 2012 Obama campaign was scientifically divisive in whipping up anger among non-core groups at straight white males.

          It greatly helps the Democrats that the federal government insists so much upon dividing up the population along race and ethnic lines and rewarding groups for being minorities. You get more of what you pay for.

        • Steve,

          It seems that you’ve got a real problem staying on topic. It is not that you have demonstrated any ability to out-argue anyone. Rather, it’s that you keep bringing in issues that others decline to address because they are off topic.

          Furthermore, you have phrased/presented the issue(s) you repeatedly bring in such inflammatory/insensitive/combative terms that it — intentionally or not — so lowers the discourse as to discourage intelligent debate.

          You just called blacks, hispanics, women, <30, jews, east asians, south asians, highly educated, lgbt, non-southern poor and urbans — correct me if I've left anyone out — "fringe groups." Twice.

          And you did so despite the fact that it did nothing to advance or clarify your position on the current topic (i.e., census questions).

          You have repeatedly confused analysis with fact and opinion with analysis. (Of course, there are places for all three in intelligent and informed debate, but only when each is acknowledged for what it is.)

          I have no idea how much of this is willful. I have no idea if your are trolling, looking for bait. I don't know if you actually think that you are the best debater on the block. Regardless which o these it might be, you need to consider how you take part in discussions on such educated and intelligent forums and Andrew Gelman's blog.

          I tried to steer the debate back to the topic. I presented non-judgemental questions to clarify positions and areas of disagreement. You have elected — again — to try to hijack this discussion to focus on your "conservative" objections to…well, you know what you are objecting to.

          At this point, I am electing to walk away. While I particularly enjoy having the last word, I am perfectly willing to cede that to others, or you, in this case.

        • “It seems that you’ve got a real problem staying on topic.”

          Dear ceolaf:

          You seem ill-tempered, poorly informed, and ill-equipped to think about non-cliched ideas. You constantly complain that I introduce broader perspectives that you’ve never heard of and don’t want to think about.

        • “My guess is that you disagree with minority set-asides and all forms of affirmative action. Is that correct?”

          I’ve several times published the suggestion that the best politically attainable compromise would be to preserve affirmative action for the descendants of American slaves and for registered members of American Indian tribes, as reparations for the two main historical victims of America, while eliminating it for everybody else.

          This would involve some refinements along the lines raised by Henry Louis Gates and Lani Guinier when they pointed out in 2004 that black students getting into Harvard were less and less descendants of American slaves and more and more individuals from foreign elites or with a white parent or white grandparents. In other words, Michelle Obama is rightfully entitled to reparation in admission at Harvard, but Barack Obama shouldn’t have benefitted from racial preferences.

          By eliminating all immigrant groups, this would have the advantage of stabilizing the racial ratio of preference beneficiaries to benefactors. When Nixon codified affirmative action around 1970, there were about 7 potential white benefactors for every 1 potential black beneficiary in the population: a ratio that whites could reasonably afford.

          That black to white ratio hasn’t changed much over the decades, but Nixon’s adding Hispanics and Asians, and the subsequent mass immigration has radically altered the racial/ethnic ratio. Today, the number of white babies who are expected to serve as affirmative action benefactors who miss out on opportunities in order to help nonwhite affirmative action beneficiaries is down to 1 to 1. The most reasonable way to preserve affirmative action for blacks and American Indians is to eliminate it for everybody else.

          After all, the fact that immigrants who have chosen America, warts and all, are given preferences over native-born Americans is morally bizarre.

  2. Why not just drop all these ancillary questions right away from the next census? What’s the downside?

    I feel we are suffering census bloat. Why not do the bare minimum that’s constitutionally required in the census and use other, cheaper ways to survey and gather any & all other socio-economic indicators that we feel are important to know about?

    • I agree with the spirit of this comment but I’m not sure the other ways are in fact cheaper. Once you are going to enormous expense to ask some questions about every single person in the U.S., the marginal cost to ask another question is pretty small. There may be legal or constitutional issues about asking questions that aren’t needed for the task, but pragmatically it makes a lot of sense. From that standpoint there’s no question you’d like to ask some additional questions, it just comes down to how many and which ones.

      • I disagree in this sense: The cost of the last census was $13 billion. The marginal cost to ask non-essential questions could well be “pretty small” as a fraction, yes, but that’s still a couple hundred million dollars.

        My point is that the alternative does not need to be a census-mimicking every-person survey. That’s wasteful. I suspect for a fraction of the money saved you could do a fairly sophisticated sample survey. Pragmatically, how accurate do we really need, say, the racial composition of the nation anyways?!

        • I think your math is off.

          “A couple hundred million dollars” is 1/65th of the cost of the census. There is no way that the marginal cost of a question is 1/65th the cost of the whole program. (Let me know if you need me to explain why I think that’s not possible.)

          So, perhaps you are suggesting that all of the non-essential questions in the whole census program have a total cost of “a couple hundred million dollars.” That would mean that you think that you can get all of that data and analysis — not just from the short form, but from the long form, and with that degree of precision for all the levels and localities — for that amount of money?

          My guess is that you don’t understand how much data the census provides, or how important the granular data really is.

        • “So, perhaps you are suggesting that all of the non-essential questions in the whole census program have a total cost of “a couple hundred million dollars.”

          Yes. At least. Only my guess, though.

          “That would mean that you think that you can get all of that data and analysis — not just from the short form, but from the long form, and with that degree of precision for all the levels and localities — for that amount of money?”

          No. Not saying that. Just saying the degree of precision / granularity especially for “all localities” that neccesitates a full person-to-person survey (of non-essentials) may not even be necessary. How many locales actually used that data to implement policy and did they need super duper accuracy for that? If I think a county has 25% Hispanics, when in fact it actually has 22%, does that change my county-policy by that much?

        • Pulling fictional examples out of my ass….

          1) Given a large enough county, a 3% different can make a huge difference. If you have hundreds of thousands of people in the county, that’s thousands of whatever group you are talking about.

          2) The census gives us data down to the level of census tracts. That’s units of 4,000 people. When it comes to placing services, that kind of information is REALLY important.

          You sample survey might give you good state-level info, and “good enough” county-level for many purposes. But you would be losing a ton of information that is actually used. Admittedly, the uses I know about are social services (both governmental and NGO), but I imagine that the world of commerce and business uses that data as well.

        • My guess is that you don’t understand how much data the census provides, or how important the granular data really is.

          Whether or not some researchers and other users find ancillary census data useful or not is irrelevant. Article I Section 2 of the U.S. Constitution authorizes Congress to conduct a census every 10 years to determine the number of people and thus the number of people in Congress. The Constitution does NOT give Congress the ability to ask whatever questions it wants to ask for any random purpose.

        • That is certainly an important point to consider. There is serious scholarship on this point, none of which I have addressed.

          Rather, I was addressing Rahul’s point that we could accomplish the same thing with cheaper sample surveys. Both of us were addressing a different point that the legal question.

        • Even if so, do we want to tack on to the census every socio-economic question since the marginal cost would be so much lower than doing the alternative complete survey of 300 million people?

          That’s my point.

  3. I’m pretty certain there is neither any legal or constitutional issue with asking these questions. Section 5 of the XIV and section 2 of the XV amendments give Congress the power to do a lot of things. So while there may be constitutional problems with asking people what they had for breakfast (etc.), it’s unlikely “race and color” questions fail the test.

  4. The race data has been on the census since practically the beginning, and is actually pretty useful to the people who, in practice, are the primary consumers of the main census data. (That would be genealogists.) The “hispanic separate from race” issue has been known to be a kludge for a while – the reason for not changing it has to do with not breaking the time series. Changing that question would mess up a lot of longitudinal analyses of census/ACS data.

    • Back in 1956, the Census proposed asking about religion. Jewish groups made a stink about being counted, and that was the end of that proposal.

      Interestingly, that has had major real-world impacts. Because the government does not count by religion, you can only sue for disparate treatment by religion. You can’t sue for disparate impact by religion the way discrimination lawsuits are constantly filed over disparate impact (e.g., today’s NYC stop and frisk ruling) and by ethnicity (if you are Hispanic). Why not? Because the government doesn’t count how many members there are of each religion, so it can’t apply the EEOC’s Four-Fifths rule to determining whether an institution has some explaining to do about their hiring or firing patterns.

      The Ethnicity question was added in the 1970s to get around the problem that most influential Hispanics (e.g., today, Marco Rubio, Bob Menendez, etc.) don’t want to identify as non-white, but they still want to be given racial preferences — hence, they now get ethnic preferences!

      • This is untrue.

        There are plenty of discrimination lawsuits based on disparate impact theories for religion. For example, Sikhs suing over head/headgear rules for certain public employees (e.g., bus drivers). That is, disparate impact suits even against public employers.

        There is no hard and fast rule that the only valid source of information about the demographics of the population is the US census. Plaintiffs can cite any source of information they want, and it is us to the defendant to try to refute that information, and up to the jury to weight that out. Even when census information is introduced, the defendant can still refute it and the jury still has to weigh the arguments.

        So, what could Sailer possibly be mean?

        Well, there is an adverse impact theory (not the disparate impact theory) that is relevant to the EEOC’s four-fifths rule (but not the complement to the disparate treatment rule). This is not about definitive proof, but rather one particular kind of evidence. However, Sailer has misrepresented it.

        * It is about the selection procedure, which is dependent upon the applicant pool. By Sailer’s explanation, municipalities would be sued every day for having fewer than 80% as many women as men. Gyms could get in trouble for having fewer than 80% as many female trainers as male trainers. Daycare centers could get in trouble for having fewer than 80% as many men as women. Etc. etc. etc..

        * This four firths/80% standard has been fading in importance. It was invented in the 1970’s, and through the last 3 years it has become less and less credible in the courts.

        To bring this back to the real topic:

        The question is where the information/data comes from. Without information about the make up of the relevant population, some theories simply cannot be used in lawsuits. And the US Census can provide some of that information. But it is not the only source, and quite often it is not a useful source at all. It simply does not collection information needed for most discrimination lawsuits because the relevant population is some applicant pool — or perhaps larger pool of qualified potential applicants — rather than the general population.

        ****************

        As for the slam on influential Hispanics, I am not going to let that go.

        Marco Rubio and Bob Menendez had nothing to do to do with adding the Hispanic category to the census, and Sailer knows that. This is a gratuitous and incredibly illogical/ungrounded/irrational (some would use other words) on these two men. I don’t have to have an ounce of esteem for either one of them to know that Sailer’s theory for why they are to blame for this census is just entirely idiotic.

        If Sailer wants to talk about uses of census data that’s fine — though he should try to get his facts straight. If he wants to talk about about the origins of particular census questions, that’s fine — though he should get his facts straight.

        But if he wants to psychologize political leaders by ethnicity, he should probably do that elsewhere. It certainly is not appropriate for this comment thread, given the topics at hand. I don’t think it is appropriate for any comments on this blog, though that certainly is not my call.

        I would also suggest that Sailer spend more time reading and thinking about the nature of ethnic preference. When American born and raised Hispanics are booed and told to “go home” when they speak at public events, it becomes clear that the overt “racism” is alive and well, strongly suggesting that covert “racism” is prevalent as well. Any number of disciplines have found evidence of racial/ethnic preferences for whites, from our standards of beauty on up. If someone wants to question the appropriateness of legal protections for particular groups that attempt to address perceived or claimed preferences *against* them, it should be done so on those terms. It’s fine to question their appropriateness as a response. One can even intelligently question the existence of particular claimed preferences or disadvantages that are use to justify the legal protections. But Sailer’s formulations are not intellectually defensible.

        *********************

        One last comment on Mr. Sailer.

        I was wondering who this man is, so I went to google to look him up. I typed in “Steven Sailer.” Google’s first suggested autocompletion was “Steve Sailer racist,” then “Steve Sailer vdare” and “Steve Sailer twitter.” As I this is my first encounter with anyone names “Steve Sailer,” I cannot claim to know whether any of that is apt. I do not even know if that Steve Sailer is this Steve Sailer.

        • “But if he wants to psychologize political leaders by ethnicity, he should probably do that elsewhere.”

          This blog is run by a political scientist and politicians analyzize themselves by ethnicity all the time. See for example “Dreams of My Father” by Barack Obama. I guess the rule is this is ok if it’s flattering, but a warcrime otherwise.

          I appreciate the correctives, but can’t stand these attempts to shut people up. Sailer’s a pretty normal guy. He doesn’t threaten people because of their ethnicity, or treat them poorely in his personal life, or hire/fire people based on their ethnicity or anything. To treat regular people who aren’t hurting anyone as if they are the second coming of Hitler, to be hounded out of the public square, is “just entirely idiotic”. Let the man speak and then correct him.

        • Note that I said that this thread was not the right place for it, and that it is not my call as to whether it is a appropriate for comments on the this blog generally.

          Do you disagree with either of those statements?

        • Was only on this thread talking about politician — whether they “analyzize themselves by ethnicity” or not — before Sailer’s comment.

          I allowed for the possibility of it being appropriate on other threads (in response to other blog posts). My point was on this particular post, the conversation that was going on before he injected the names of particular politicians into the discussion. (politicians, as I have said, who had no personal connection to the point he was making about the 1970’s.)

        • Dear Ceolaf: Obviously, I didn’t argue that Rubio, who was a small child during the Nixon administration was one of the white Hispanics who got the feds to declare Hispanic not a race but an ethnicity. I cited him as an example of a current beneficiary. Back during Nixon’s term, beneficiaries included the President’s best friend Bebe Rebozo, a white Cuban.

          The point of course is that Hispanic preferencse tends to do a lot more good for white Cubans than for mestizo Mexicos or black Dominicans, even though the affluent Republican white Cubans need preferences the least.

        • Steve,

          No, that’s untrue.

          “The Ethnicity question was added in the 1970s to get around the problem that most influential Hispanics (e.g., today, Marco Rubio, Bob Menendez, etc.) don’t want to identify as non-white, but they still want to be given racial preferences — hence, they now get ethnic preferences!”

          You cited them as “influential Hispanics,” which I take to mean “HIspanic leaders.” You linked them to such people in the 1970’s, even though they were not in such positions in the 1970’s and each may disagree with such people in a variety of ways.

          In fact, you cited two sitting US senators as example of problem from the 1970’s. You did not simply cite them as beneficiaries, but as contemporary exemplars of this old problem. And, as written, the problem was that that a group of “influential” people wants a pair of possibly contradictory things.

          In fact, I have seen no evidence that this change in the census questions was a creation of Hispanic US senators, despite your implication that they are to blame. And even if the vast numbers of Hispanic US Senators at the time were to blame, it would STILL be inappropriate to name drop Rubio and Menendez, as they had nothing to do with it.

          Further, whomever you really meant by “influential Hispanics” in the 1970’s, they were actually NOT US Senators. Whatever their means of influence, it was not as members of the world’s most deliberative body.

          Rubio and Menendez are not the most famous Hispanics in the US today. Perhaps Jorge Ramos. Perhaps one of the Lopez entertainers. It’s actually really hard to see how either Rubio or Menendez are influential at all.

        • ceolaf: I wouldn’t have predicted that I would find myself compelled to defend one of Steve Sailer’s statements. Huh.

          Here’s what he wrote: “The Ethnicity question was added in the 1970s to get around the problem that most influential Hispanics (e.g., today, Marco Rubio, Bob Menendez, etc.) don’t want to identify as non-white, but they still want to be given racial preferences — hence, they now get ethnic preferences!”

          You read this as promoting a theory that Marco Rubio and Bob Menendez are somehow to blame for the presence of the Hispanic ethnic category in the census. To me that seems to be a misreading of what Sailer wrote. Marco Rubio and Bob Menendez are cited only as current examples of influential Hispanics who would like to have special preference even while identifying as white; they are not cited as having anything to do with the census in the 1970s. Let me rearrange the words a bit to show you how I interpreted it:

          Most influential Hispanics (for example, today, Marco Rubio, Bob Menendez [and others in the past, presumably]) have a problem: they don’t want to identify as non-white, but they still want to be given racial preferences [or at least, special preferences of some kind]. In the 1970s, the ethnicity question was added to get around this problem — thereafter, influential Hispanics got [to identify as white and still receive] ethnic preferences.

          I express no opinion about the factual accuracy or moral status of Sailer’s statement, although I will say that the passive voice of “the ethnicity question was added” is lazy writing.

        • ceolaf says:

          “Rubio and Menendez are not the most famous Hispanics in the US today. Perhaps Jorge Ramos.”

          Univision anchorman Jorge Ramos is a classic example of how Latin American culture prefers blue-eyed whites. Ramos looks like Anderson Cooper crossed with C-3PO:

          http://www.immigrationbn.com/Portals/52046/images/E–Documents%20and%20Settings-mpollak-My%20Documents-My%20Pictures-My%20Pictures-Website-jorge_ramos_01-resized-600.jpg

          It’s absurd for Ramos’ children to benefit from ethnic preferences, but that’s the way it is.

        • Rahul: Well, admittedly I’m probably used to people a little rougher around the edges than most academics, but here some examples to illustrate where I draw the line:

          Normal: regular middle class Americans, warts and all. People who live and let live even though they have strong opinions about things. People who try to affect change through regular politics, exposition and debate.

          Abnormal: violent extremists, head chopping Muslim fundamentalists, cross burning KKK members, murderers, people who use physical intimidation to achieve political ends, gang members, those who participate in mob violence and so on.

          As it happens I’ve seen real extremists up close and personal. They aren’t even remotely like Sailer. He’s very clearly in the Normal category.

          So, where do you draw the line?

        • I draw the line a lot earlier than you I guess. :) I include all the people under the Abnormal class you succinctly summarized and then some more. Sailer included.

          To me, people like Sailer are more insidious than a railing fundamentalist because they can be persuasive (and clever) and cause so much more damage. The evil is less patently obvious.

        • I don’t think whether or not he is “normal” is particularly dispositive.

          More generally, there have historically been all kinds of widely held ideas and beliefs — eminently “normal” — that were quite problematic, nonetheless.

        • Rahul,

          “The evil is less patently obvious”

          Seriously? Sailer would be considered a “liberal” in about 90% of the countries on this planet. Even his views on immigration are more “liberal” than 90% of the people on this planet. And I don’t just mean famously xenophobic places like Japan either. Check out Mexico’s laws against immigration from Guatemala for example. By any reasonable standard he’s clearly a part of the western classical liberal tradition in the same sense that both Democrats and Republicans are, as opposed to totalitarian traditions like communism, fascism, absolute monarchies, or religious fundamentalism. He merely disagrees with you. That’s it. Last time I checked disagreeing with “Rahul” doesn’t make you evil. It just makes you not-Rahul.

          Nick: Unfortunately Frequentists don’t seem to differ from Bayesians in any measure other than they just can’t imagine probabilities which aren’t frequencies. It would be so much easier if they were just dumb or dressed funny or something.

          Ceolaf: being “normal” is pretty dispositive. There’s all kinds of problems, weirdness, and just plain “full of crapness” in everyone. There is however a pretty clear difference between normal law abiding people, who don’t treat their fellow citizens poorly, engaging in civil debate and people like Osama bin Laden/Stalin/Mao/Hitler. To take someone in the former category and act like they’re in the later is just pure posturing and is obviously done in order to shut down debate. I’m glad he said what he really thinks and that you responded. Nothing bad came from it. Nothing whatsoever.

        • Incidentally, here’s an anecdote about how people in other countries really view immigration when it involves people coming into their country.

          When I was in Anbar Province Iraq, one of the problems we had to deal with was a group of several thousand “Palestinians” who tried to leave Iraq when the sectarian violence flared up. They got to the a border crossing with Syria and Syria refused to admit them. So they just stayed there in the no-man’s land between Syria and Iraq. Since this was out in the deep desert, US Marines had to give them food and water or they would have all died within a week. Some of them were stuck there for years.

          So who were these “Palestinians”? They were indistinguishable from any other Arabs in Iraq and had been living in Iraq for so long at least one new generation had been born and raised there. They originally came from Jordan I believe.

          The whole time they were Iraq (decades), no Iraqi government would give them citizenship and Syria wouldn’t even let them in the coutry. We tried everything we could to find a country that would take them, but none of the other countries include Arabs would take them in.

          That’s a good illustration how the rest of the world really views immigration. You can be far along the scale of “immigration restrictionist” in the US and still be more liberal most the rest of the planet.

        • @Entsophy:

          He merely disagrees with you. That’s it. Last time I checked disagreeing with “Rahul” doesn’t make you evil.

          Well, if only it was only me. Do you really think I’m expressing a novel outlook on Mr. Sailer?

          PS. I disagree with a lot of people (e.g. you) without thinking of them as evil. :)

        • Yes, well my mother, first wife, and current wife are all foreign born naturalized US citizens. The later from Afghanistan of all places. So I guess I have the good fortune of only disagreeing with you on things people aren’t hyper sensitive too. If that wasn’t the case, everyone would demonize me as an American Pol Pot ready at any moment reignite a new round of the Killing Fields, and the fact that I’d be the exact same freedom loving, rights respecting, classical liberal wouldn’t count in my defense one bit.

          Most of the planet holds one view or another far more extreme than Sailer. Some of them no doubt are evil. But most are just regular ol’ people.

        • “Most of the planet holds one view or another far more extreme than Sailer.”

          I feel there’s a crucial difference between some illiterate peasant’s instinctive xenophobia and the sophisticated, malicious version Sailer spouts.

        • Rahul is concerned about by evilness:

          “To me, people like Sailer are more insidious than a railing fundamentalist because they can be persuasive (and clever) and cause so much more damage. The evil is less patently obvious.”

          Of may be I’m clever and persuasive because I’m right.

          After all, I have thought longer and harder about these questions than almost anybody else. And I’m pretty good at seen through piffle and logically reducing questions to their essences.

        • Rahul,

          Immigration is a policy decision. It has about the same affect on people and about the same moral character as any one of a hundred other policy decisions: abortion, tariffs, use of immanent domain, blah blah blah and so on.

          Why does disagreeing with Rahul on this policy decision make you vile beyond belief, but disagreements on the others are fine?

        • @Entsophy:

          “Why does disagreeing with Rahul on this policy decision make you vile beyond belief, but disagreements on the others are fine?”

          By your “vile beyond belief” you mean my characterization of Sailer as evil, yes? Just making sure.

          If so, disagreement on immigration policy is hardly the reason. Actually, I’m curious what you think is my opinion on immigration policy. Humor me please.

        • On who’s right and who’s wrong here, I have my guesses but I will keep them to myself.

          But posting as “Anonymous” and asking who is Steve Sailer, with all the innuendo, is poor practice. Anonymous may have very good reasons for keeping anonymous, and that’s fine by me — and, immensely more to the point, consistent with the practice of this blog. But staying anonymous and writing that last paragraph is not on, I suggest, so I trust Anonymous will retract it as no more than a cheap shot. Who Steve Sailer is is not the point; it’s whether he’s right or wrong.

        • I have already owned the comment.

          I did not mean to post it anonymously, and do not know how that happened, and there is — to my knowledge — no way to correct it.

          However, I see nothing to retract. It is an honest retelling of what happened when I tried to up Mr. Sailer at that point in the writing of that post. I make quite clear that I lack the knowledge to judge the aptness of those searches, or even their relevance to this Steve Sailer.

          Had I made the story up…well, I wouldn’t do that. But if I had, that would be worthy of a retraction. Had I been saving it in my back pocket for a while and just pulled it out right here, that would probably be worthy of a retraction. If I used those facts as the basis for my criticism of Sailer or his comments, that would be worthy of retraction. But none of those are the case.

          I address his comments on their (lack of) merits first. And then I separated a piece of information that I encountered (i.e., how google suggests autocompleting searches that start with his name).

          That kind of information fits into “consider the source.” But I did not dismiss his comments based on ad hominem attacks, rumors or any discussion of his person.

          Yes, by their inclusion I am suggesting that they are relevant to the discussion at hand. I stand by that. Clearly, many other people have searched for “Steve Sailer Racist,” though I would not have done that myself. Because I do not know that man, and have not paid attention to his (other) writing, I cannot even begin to judge whether or not he is racist. However, if this comment is typical of his writing, I can see why some people might ask the question. But I do not know whether this comment is typical of his writing.

          Are you, Mr. Cox, suggesting that that last paragraph to which you object is untrue, dishonest or in any way misrepresentative of anything? Are you suggesting that it is irrelevant to the rest of my comment? Are you suggesting that we should not consider who offers ideas/comments?

          My own unique pseudonym does not generate any suggested autocompletions. My legal name’s first autocompletion is for another spelling of the name, followed by “click2profit”, “sanaria” and “toronto” (those last three with the correct spelling) — none a reference to me. Gelman’s are “blog” “columbia” and “stan”.

          Actually, I wrote Gelman to ask if I’d gone too far, as it is his blog and I trust him to make the judgment. I thought the comment acceptable, but I wanted to double check.

          I certainly am open to revisiting the question. Can you explain why the comment is out of bounds or worthy of retraction. I truly would like to hear more.

        • You can snipe at, or more precisely about, Steve Sailer because he’s visible. He can’t reply in kind. I don’t think he should and I don’t think you should, but the playing field was not, and is not, level if you exploit his visibility and hide who you are.

          “Steve Sailer racist” is not what you said but why did you cite it? I call that innuendo. Your argument is strong enough without it. I am glad you agree that ad hominem attacks are out of bounds.

          I object only to your last paragraph. I am not expert on any of these matters, but I tend to agree with your main stance here.

        • 1) Clarification question: Is our difference in visibility because he is higher profile than I am, or is it because I use a pseudonym and he does not?

          In fact, I have been using this pseudonym consistently online for over 20 years. One (but only one) reason I use it that its consistent use best gives people the ability to link the various things I have said in the various places I have said them. Thus, they are best able to to judge me based upon what I have said if I continue to use it.

          Yeah, it’s a self-perpetuating kind of thing. How do I ever get off this crazy thing?

          In fact, I do not think that it is that hard to figure out my real name. But if it would make a difference, Alexander Mishra Hoffman.

          But I don’t think that makes a difference.

          2) I reported that it was out there. I cited it because google seems to indicate that there are a lot of people who think that he — or someone else by that name, as I acknowledged — is racist. As I said, I am not in a position to judge what is in his heart or his mind. But I think we can consider this wider view, though we should not be ruled by it.

        • “Who Steve Sailer is is not the point; it’s whether he’s right or wrong.”

          Why not? Knowing who Sailer is and access to his large body of prior work can be a good opportunity to calibrate one’s priors about how seriously to take anything he writes.

          Nothing wrong in being a good Bayesian about authors you are reading.

        • Threading is getting intricate here, but here is a reply to @ceolaf.

          Thanks very much for considering my comments carefully and temperately.

          1) My objection was, as explicitly stated, to your last paragraph when posting as “Anonymous”, which gives complete asymmetry of identifiability. You’ve changed the question to what difference it makes if you post as “ceolaf”. That’s a more complex question, but, simply, I wouldn’t have made the same comment if you had done that. “Anonymous” I now understand to be an accident.

          2) Your latest comment appears to me much milder and more reasonable than what I took you to be implying originally.

        • NIck,

          1) Even if I had done it as Anonymous, why would that matter? If the facts I reported were true — and they were — then why would it matter who I was? Sure, if would make it harder for someone to engage in an ad hominem attack against me, but obviously that’s not your point. Honestly, I do not see what about that particular paragraph makes it out of bounds for an anonymous post.

          2) OK, here are the two paragraphs that you read differently.

          “I was wondering who this man is, so I went to google to look him up. I typed in “Steven Sailer.” Google’s first suggested autocompletion was ‘Steve Sailer racist,’ then ‘Steve Sailer vdare’ and ‘Steve Sailer twitter.’ As I this is my first encounter with anyone names ‘Steve Sailer,’ I cannot claim to know whether any of that is apt. I do not even know if that Steve Sailer is this Steve Sailer.”

          “I reported that it was out there. I cited it because google seems to indicate that there are a lot of people who think that he — or someone else by that name, as I acknowledged — is racist. As I said, I am not in a position to judge what is in his heart or his mind. But I think we can consider this wider view, though we should not be ruled by it.”

          Neither paragraph endorses a judgment of Sailer. And both have disclaimers. The only difference I see that makes the latter softer is the explicit statement “we should not be ruled by it.” On the other hand, it is also more explicit about accusations of Sailer being a racist, and explicitly says that can consider that. (And obviously, it makes reference to the source of this information, rather than actually citing what I found.)

          I am curious why you read the latter as notably milder and more reasonable? Is it the difference between “I cannot claim to know whether any of that is apt” and “I am not in a position to judge what is his heart or his mind”?

          I truly am curious. To me — the author — they read very similarly in tone and and level of accusation.

        • “For example, Sikhs suing over head/headgear rules for certain public employees (e.g., bus drivers). That is, disparate impact suits even against public employers.”

          That’s not a statistical disparate impact lawsuit, that’s a disparate treatment lawsuit that the helmut rule would affect all pious Sikhs. There’s not need for counting exceptions. It’s like 1972 Plyer case argued over whether Amish could take their kids out of school after 8th grade.

        • Steve,

          Being married to a long time employment attorney (management-side) and educated in constitutional law at Harvard University (among other places), I tend to use legal terms the way lawyers and our legal system does.

          Your understanding of “disparate treatment” “disparate impact” and “adverse impact” seems quite confused. Certainly, you are confused about which term means what. But I think your confusion may be larger than that.

          So, though you are right that that example was not a *statistical* disparate impact lawsuit. But that does not make it a disparate *treatment* lawsuit. I checked with real experts before posting that, and they could not tell what you were talking about.

          In part, this might be due to your new (at 5:08am, today) introduction of the term “statistical disparate impact,” but I don’t think so. I think your confusion runs deeper than that.

          In fact, using statistical evidence in a disparate impact lawsuit does not require census data, as I said above. The overall population is usually not the comparison group — certainly not in employment lawsuit, like firefighter exams. Rather, as I said above, it is the relevant population — such as the applicants or test takers. Further, this 4/5’s rule (which I’ve usually seen as an “80%”, but we can use the resonances you seem to prefer) has been fading out for 30 years. (Years, that’s 30, not 3. That was a typo in the original post. One of many. Sorry about that everyone.) I believe it was established by the courts, not the EEOC, and the courts have been paying less and less attention to this threshold for quite some time.

          Perhaps you could offer a specific example of a lawsuit that cannot be filed because of the limits of the census questions. That is, perhaps if you could cite a particular successful lawsuit that fits your meaning of “disparate impact” that a religious minority group (e.g., Baptists) could not file.

        • “Perhaps you could offer a specific example of a lawsuit that cannot be filed because of the limits of the census questions. That is, perhaps if you could cite a particular successful lawsuit that fits your meaning of “disparate impact” that a religious minority group (e.g., Baptists) could not file.”

          Sure. Consider the numerous firefighter hiring/promoting discrimination cases, such as Vulcan Society against the Fire Department of New York. Vulcan Society is a pure statistical disparate impact case over whether the hiring exam unfairly hurts a race (black) and an ethnicity (Hispanic). There have also been numerous lawsuits over strength requirements for firefighter hiring because they have disparate impact on women.

          On the other hand, in many Northeastern fire departments, Catholics are far more represented than Protestants (as a glance at the names of the 20 white or Hispanic plaintiffs in the famous Ricci reverse discrimination suit in New Haven would suggest). Yet, in small towns in the Northeast, blue collar white Protestants are often active in volunteer fire departments. But they seldom get the good paying jobs in paid fire departments.

          The statistical arguments that white Protestants suffer from discrimination in firefighter jobs would be very similar to the arguments successfully advanced that blacks, Hispanics, and women suffer from discrimination as show by disparate impact. But, I can’t recall any lawsuits on this ground. Maybe they would work, but there seems to be very little interest in the whole topic. I’ve been interested in firefighter discrimination lawsuits for at least 20 years or so, and I don’t recall ever hearing white Catholic v. white Protestant disparity ever come up. I’m sure it has, somewhere, but, in general, religion is vastly less used as a category in firefighter suits than race, ethnicity, or gender.

        • Steve,

          I’ve already addressed this. I’m sorry that you missed it.

          Such suits are based upon the applicant pool, rather than the general surrounding populations. The analysis is done on the applicant pool (or qualified potential applicants).

          The lack of census data on religion in the general population is not at all a deterrent to filing such suits as you describe, because such data wouldn’t really be useful in such suits. More importantly, there is nothing in law (case or statute) to prevent such suits, given appropriate data.

          There are all kinds of possible explanations for the disparity you cite, but that is not for this comment thread. Rather, the point is that the lack of such law suits simply cannot be traced to the nature of our census questions.

          So, I ask again, can you please name a specific example of a lawsuit that cannot be filed because of the limits of the census questions?

          (btw: I’ve always thought that a really smart applicant would get his/her demographically similar friends to apply for such jobs and/or take such exams. Thus, s/he could alter the applicant pool. If a minority make concentrated efforts like this, it would require a major change in how we consider certain kinds of evidence of bias. But I’ve never seen credible stories of this kind of effort. The point is, it’s about the demographics of the application pool, not the general population, and the census cannot give us data about a particular applicant pool.)

        • “Such suits are based upon the applicant pool, rather than the general surrounding populations.”

          No, in firefighter hiring cases, much of the complaint is that not enough minorities join the applicant pool. Statistics on the general population in the region are frequently cited to show that legally protected groups are in some insidious way being discouraged from applying.

        • The Sikh lawsuits against motorcycle helmet laws (because they interfere with their turbans) are similar to the Supreme Court’s famous 1972 Plyler v. Doe case in which the Justices exempted Amish parents from having to send their children to school beyond eighth grade. The statistical question of whether the laws impact Sikhs or Amish aren’t of major concern. It’s simply assumed by all parties that the laws contradict the dictates of the religions in question and then questions like religious liberty vs. general welfare are debated.

          In contrast, the 2009 Vulcan Society firefighter hiring case in NYC was 100% statistical in terms of the evidence that the judge accepted as decisive in taking control of FDNY testing away from the city.

          Similarly, the evidence used in week’s stop and frisk decision against the NYPD is almost (although not quite) 100% statistical.

  5. The Republicans need to either abolish or expand the ethnicity category if they ever get their hands on the Executive Branch again. Why should the only Ethnic categories be Hispanic or Non-Hispanic? Either get rid of ethnicity altogether or tell everybody to pick whatever ethnicity they want, and then have the government enforce disparate impact anti-discrimination rules for each and every ethnicity. Slovaks make up less than four-fifths of their expected rate in your workforce? The EEOC wants to know why!

      • I know that, but few racial/ethnic preference benefits are tied to the Ancestry question, which isn’t normally part of the Census. What the Census mostly cares about is sex, race, and ethnicity (i.e., Hispanic/Latino or nonHispanic). I’ve never heard of a disparate impact lawsuit based on the Ancestry question, although maybe there are a few.

    • It is not true that the question about hispanics is the only ethnic question on the census. Black, for example, refer to people with several different degrees of African ancestrality but rarely 100 % African and often with quite a lot of european ancestry, such as Obama who is 50 % white. The blacks are perceived as race – which is another story – but that is also true for the hispanics – who in practice have a racial status, despite their genetic diversity. And, finally, the same is for the whites: for example, for a long time, Irish were not considered white, and so were the germans, italians, welsh, etc. Thus white often refer to ethnic group as well as oppose to race. In really today we have much better ways – and also increasingly cheaper – to access someone racial ancestrality, assuming that matter for something, such as health purposes. The ethnic questions are the ones that would matter more because they refer to the community someones lives are embedded, their lives styles, etc. But even that is often distorted for example, with patently white people marking hispanics in top jobs such as universities to be counted as diversity. Sadly, these people were never among the unprivileged so that the whole thing is just fake.

      • An irate Anonymous says: “Marco Rubio and Bob Menendez had nothing to do to do with adding the Hispanic category to the census, and Sailer knows that.”

        Of course I know that: that’s why I used the word “today” in “most influential Hispanics (e.g., today, Marco Rubio, Bob Menendez, etc.)” to give some current examples of “white Hispanics” (a lot whiter than George Zimmerman, by the way). Back during the Nixon Administration when Hispanic ethnicity was made official, the most influential Hispanic was probably Bebe Rebozo, President Nixon’s best friend.

        • This is a good example of the dumbing down of intellectual discourse in America: ceolaf is having trouble keeping up in the argument, so he feels completely entitled to play the Disgust Card.

        • There’s probably a good reason for disgust having evolved as part of the range of human emotions.

          Sometimes, disgust may be justified.

        • Steve,

          Do you really think that I have failed to respond to your arguments on the merits?

          (Obviously, I’ve not gotten to this afternoon’s batch, yet.)

          1) Do you really think that you made any responses to my arguments that I failed to pick up before correcting this mistake on your part?

          2) Are you unaware that it was you who raised the issue of tone, not me? Is that why you mistook my correction of your mistake with “playing the Disgust Card”?

          (btw: the “Disgust Card” is not a particularly well known or widely used term. Google only has 680 hits for it, and most of them seem to be entirely different than this usage. Did you just independently coin term here, or is it something you’ve seen before?)

          So, If you don’t like discussion of tone, don’t go there, yourself.

          But most certainly, don’t falsely paint others has having trouble keeping up. You’ve dropped most of my arguments — to use a debate term — and may well drop many more in your round round.

      • The Census Bureau in 2000 labeled the question of whether you are Hispanic or Non-Hispanic as the Ethnicity question. The Census made a big deal about how ethnic Hispanics/Latinos could be any race.

        This is just federal government standard operating procedure. It’s not something evil horrible me made up.

    • Steve,

      You keep making references to these “racial/ethnic preference benefits.”

      Can you please be a bit more specific about the benefits to which you refer. It has variously seemed that they are governmental, non-governments and perhaps even personal or social.

      More recently, you cited minority business loans and minority government contracts. Are there others that are relevant to the census data? Did those two require a change in the census question?

      Does it matter what category a persons selects on the census form when it comes to qualifying for these benefits/preferences? Do the relevant laws, regulations, programs, rules (etc.) make reference to the categories as recognized/defined by the US Census Bureau?

      I am trying to understand the relevance of these programs (etc etc) to the issue Gelman raised about the census form, itself.

      • The federal government puts a lot of effort into lining up race and ethnicity categories in both data collection (e.g., the Census) and implementation (e.g., minority business development programs). The Office of Management and Budget plays a key role in coordinating government efforts.

        For example, in the 1990s, mixed race individuals persuaded the Census to allow them to check multiple racial boxes so they wouldn’t be, in effect, asked which parent they love more. This threatened to cause trouble in all sorts of minority preference programs, however, until in 2000 Bill Clinton announced that mixed race people would receive racial preferences under whatever nonwhite race they checked.

        The major exception where the usual OMB/Census categories often don’t apply remains crime rates, where Hispanics are frequently not broken out. This has both practical reasons and political ones.

        There’s a vast amount available on line on these topics and you might find it informative to read up on them.

      • There are a few cases where government preferences aren’t aligned with Census categories. The main anomaly I can think of is the Reagan Administration’s 1982 decision to extend preferences in government contracting and the like to whomever checks the Hasidic box. This came up in the 2008 Ephraim Diveroli arms-dealing scandal where a 22-year-old Miami Beach party animal won a $300 million dollar Pentagon contract to supply ammunition to our puppet regime in Afghanistan.

        http://isteve.blogspot.com/2008/03/how-did-diveroli-family-qualify-for.html

        Among other breaches, the fashionably stubbled Diveroli checked the box claiming to be Hasidic. (Diveroli’s uncle, Michael Jackson’s rabbi Shmuley Boteach, had converted to Hasidism for awhile, but there was zero evidence that Diveroli was Hasidic by either practice or recent ancestry.)

        But the broader implication of this odd 9-Days-Wonder of a story is that the little known Hasidic preference hasn’t really taken off since 1982. One reason might be that it doesn’t align with federal data collection standards, which ignore religion to concentrate on race and Hispanic ethnicity.

        • A correction: the Reagan Administration extended preferences to the Hasidic in 1984, not 1982 as I erroneously stated. From the New York Times:

          REAGAN GRANTS HASIDIM ‘DISADVANTAGED’ STATUS
          By RICHARD SEVERO
          Published: June 29, 1984

          They were talking about it in the Crown Heights section of Brooklyn yesterday. Bearded men in dark coats under a hot sun, men known for their deep spiritual values, their belief in education and hard work, their pride in self-reliance.

          They were all Hasidic Jews, and they were talking about President Reagan’s decision, announced Wednesday, to add them to a list of minority groups considered ”disadvantaged” by the Government.

          The list already includes Hispanic people, blacks, Indians and other groups that are considered by the Government to have encountered severe economic problems because of discrimination.

          The designation means the Hasidim are able to apply for Federal assistance in running businesses. They will also be eligible for programs that set aside work for minority-group businesses.

          The designation took effect immediately.

          Most of the Hasidim did not know the details of what the decision would ultimately mean for them. But there seemed to be accord, at least among the thousands of Lubavitcher Hasidim who live on Crown Heights, that the Government had done a good thing for them. Zeb Katz, who was rushing to close his Kingston Avenue textiles outlet at 2 P.M. so he could go to Manhattan to study the Talmud, said he hoped the designation would enable him to better compete with other business people and, perhaps, aid him in expanding customer list nationally.

          In the Bay Ridge section, Israel Szempow, who makes special items out of flexible vinyl at a business on 63d Street, said that the new policy would ”give us a great opportunity to move up in this country.”

          He said that over the years he has been in business, there had been numerous occasions when he had bid for jobs but did not receive them, even when his bid was lower, ”because when I made the bid I was wearing a yarmulke.”

          http://www.nytimes.com/1984/06/29/nyregion/reagan-grants-hasidim-disadvantaged-status.html

  6. @ceolaf

    1) Even if I had done it as Anonymous, why would that matter? If the facts I reported were true — and they were — then why would it matter who I was? Sure, if would make it harder for someone to engage in an ad hominem attack against me, but obviously that’s not your point. Honestly, I do not see what about that particular paragraph makes it out of bounds for an anonymous post.

    Me: Evidently, you don’t consider Anonymous postings out of order and they are allowed here. But that’s not a universal view. You’re handing ammunition to those who want to disagree with you. They could say “Here is someone hiding behind Anonymous and adding innuendo. They can’t be taken seriously if they don’t have the guts and self-confidence to say who they are while attacking someone identifiable.” If Anonymous is your handle, you need to be utterly sparing on side rhetoric and totally focused on logic and evidence. That’s my view on good practice in debate.

    2) OK, here are the two paragraphs that you read differently.

    “I was wondering who this man is, so I went to google to look him up. I typed in “Steven Sailer.” Google’s first suggested autocompletion was ‘Steve Sailer racist,’ then ‘Steve Sailer vdare’ and ‘Steve Sailer twitter.’ As I this is my first encounter with anyone names ‘Steve Sailer,’ I cannot claim to know whether any of that is apt. I do not even know if that Steve Sailer is this Steve Sailer.”

    “I reported that it was out there. I cited it because google seems to indicate that there are a lot of people who think that he — or someone else by that name, as I acknowledged — is racist. As I said, I am not in a position to judge what is in his heart or his mind. But I think we can consider this wider view, though we should not be ruled by it.”

    Neither paragraph endorses a judgment of Sailer. And both have disclaimers. The only difference I see that makes the latter softer is the explicit statement “we should not be ruled by it.” On the other hand, it is also more explicit about accusations of Sailer being a racist, and explicitly says that can consider that. (And obviously, it makes reference to the source of this information, rather than actually citing what I found.)

    I am curious why you read the latter as notably milder and more reasonable? Is it the difference between “I cannot claim to know whether any of that is apt” and “I am not in a position to judge what is his heart or his mind”?

    I truly am curious. To me — the author — they read very similarly in tone and and level of accusation.

    Me: Repeating my view won’t strengthen it your eyes, but the second does seem milder and more temperate to me. Minute differences in wording can make a big difference.

    • @Nick Cox:

      You said:

      ” My objection was, as explicitly stated, to your last paragraph when posting as “Anonymous”, which gives complete asymmetry of identifiability”

      There is an asymmetry yes, but it isn’t a liability for someone like Steve Sailer. When he chooses to advertise who he is in blog comments it is because he wants the name to give him credentials.

      So it isn’t as if Anon versus Steve Sailer gives Anon any degree of unfair advantage.

      I still think your objection to @ceolaf’s anonymous comment was uncalled for.

      • I guess I agree with you on all important principles, so I have to assume that you have read the entire exchange, including posts later than that you respond to, yet I am just not being clear enough.

        @ceolaf posted as Anonymous; he has explained that was a mistake, but I was responding to that original posting. I was also objecting only to his last paragraph — because I think its style weakens his argument, which with minor exceptions appears cogent and correct so far as I can see. As you know, in similar debates, the main aim is not to convince the person you think is very wrong of the error of their ways, which usually would be futile; it is persuading as many of the other people as possible that your antagonist is wrong. Cheap tricks give excuses to sceptics or the opposition to discount you or disregard you.

        Like almost anybody else, I object to personal insinuations or ad hominem comments being introduced into debates like this. That principle still leaves enormous scope for disagreement about precisely what is fair comment or allowed as rhetoric and what isn’t. “Who is this guy?” and “What else has he said?” are different questions.

        It was incorrect of me to guess that @ceolaf was trying to hide, but really, that was his mistake in posting as Anonymous; that is now a minor detail, but it’s the context for my first posting. @ceolaf has assured us that he was not intending to be ad hominem and I am very pleased to hear that. I think my first posting was fair comment on what it replied to.

        I have absolutely no interest in defending Steve Sailer’s views, by the way.

  7. The problem I see it that Hispanic people will see this question as a way to hide or ignore them, to reduce their power as a group, so they’ll asnwer this question as “some other race” – “Hispanic” and then maybe tick “white” or whatever. And then the “maybe tick white or whatever” will cause huge headaches for being able to match back to previous censuses.

    It seems odd to care about the distinction between Pacific peoples (e.g. Samoan/Tongan) who are relatively few in number but to not care about counting a large group such as Hispanic people.

    • Hispanics want the advantages of racial/ethnic preferences without having to declare themselves racially anything other than blue-blooded Castillians. You can be the co-star of the biggest TV show of the 1950s (Desi Arnaz), you can be President Nixon’s best friend (Bebe Rebozo) and still be eligible for racial/ethnic preferences.

      Heck, look how Marco Rubio is considered Presidential Timber despite being only a first term Senator. Why? Because he’s Hispanic!

      • Steve,

        1) Michelle Bachman. Donald Trump. Herman Cain.

        Who considers Marco Rubio to be “Presidential Timber”? (Heck, what does “presidential timber” even mean?)

        Years before the the first GOP primary, people are talking about Rubio as having a shot. But far closer to the GOP primaries of 2012, people spoke about various people as leading candidates. Those were, at times, front runners by various measures.

        2) Rubio is taken seriously as a potential seeker of the presidential nomination. And his ethnicity is relevant to that analysis. But that has absolutely nothing to do with government preferences, programs or set asides. It has nothing to do with census questions, either. So, it’s not clear what he has to do with issues at hand.

        3) In fact, HIspanics are our fastest growing minority group and our largest minority group. This would be true even if there was no Ethnicity question on the census and there were no government benefits or programs for minorities. A leading HIspanic politician from the (slight) minority party that has traditionally had enormous difficulty getting Hispanic votes is taken seriously as a potential presidential candidate because such a person offers a plausible theory for a route to the White House.

        But note that my next door neighbor is not considered “Presidential Timber,” although he is Hispanic.

        Note that George Lopez is not considered “Presidential Timber,” although he is Hispanic.

        Note that Bob Menendez is not considered “Presidential Timber,” although he is Hispanic.

        4) The passive voice in your construction is critical. Who is it who actually seriously considers Rubio to be “presidential timber”? Who are you talking about? Do you even know? Are you mistakenly seeing sincerity among the political pundit class?

        • The mindset among non-Hispanic white elites that sees Marco Rubio as a natural leader of the vast Hispanic minority is largely a social construct tracing back to the Nixon Administration’s decision to lump Mexicans, Puerto Ricans, and Cubans together under the rubric of Hispanic ethnicity.

          It seems like a Law of Nature today because that’s how the feds categorize people, but it’s easy to imagine alternative classification systems (e.g., each nationality gets its own quotas, or all these Latin American groups are just lumped under the broad white tent, the way that Arab-Americans are today).

          Even under the current system, you’ll note that this notion of pan-Hispanic loyalty is mostly found among elites. How much polling evidence is there that the white Cuban Rubio appeals to, say, the masses of Mexican-Americans? A lot of white Republican elites assume they can easily sell Rubio to Mexican-Americans, but where’s the data?

        • To continue, most of the data suggests that Mexican-Americans vote heavily Democratic for perfectly rational reasons of self-interest. They like traditional Democratic tax and spend policies because they tend to come out ahead on them. (Plus, the Democrats vociferously defend ethnic preferences for Hispanic/Latinos. While the Republican elites don’t do much about them, clearly the GOP base is less enthusiastic about affirmative action than Democrats.)

          The widespread Establishment notion that Mexican-American, Puerto Rican-American, or Central American-American voters, will massively convert to voting for a Tea Party white Cuban Republican out of sheer ethnic solidarity with Marco Rubio’s Hispanicness is speculative to say the least. In fact, cynics might be forgiven that the idea of President Rubio is being pushed so hard by Democratic and liberal leaders to outwit poor dumb Republicans.

        • Again, trying to stay relevant to the issues of the census questions….

          It seems that you and I agree that Rubio is NOT a serious candidate for president, right? Or, at least not for the “Oh, he’s HIspanic!” theory, right? I think we are on agreement about the political significance of this first-term senator.

          But you cited him as being a current “influential Hispanic,” the modern equivalent to the “influential HIspanics” who got the ethnicity question added to the census, right? You have since pointed out that Nixon’s Hispanic best friend was the sort of guy who fit into the “influential Hispanic” category back when the ethnicity question was added.

          I can totally see how the president’s best friend would be a kind of influential guy, even if he had no pull in any Hispanic communities. That makes sense to me. But I don’t see how first-termer Rubio — who is neither influential among many (any?) HIspanic communities, nor influential in the White House — is similar to him, for the purposes of this discussion. Sure, they’re both Hispanic. I get that. But what do you mean by “influential”?

          What kind of “influential” are you talking about, that can have such impacts on census questions?

      • That may be true but still it is an issue for only a few of those who happen to be able to classify themselves as hispanics. Not for most of the brown-poor hispanics.

      • Also, the need for them (white hispanics or whatever you call) to show their white credential might itself be a function of US racism. They fact that in US history white americans so picky about how can be considered white. Recall that in the past Irish were among those who were prouder of being white possibly just because they have it challenged.

        It is a like a new rich attitude of being proud of their money, mostly because they just got that, while old rich are much more cynical and blasé about that.

        • “Also, the need for them (white hispanics or whatever you call) to show their white credential might itself be a function of US racism”

          You should read up on white racism in Latin America. It’s not something Latin Americans needed to learn from the U.S.

          Try looking at pictures of, say, the last ten presidents of Mexico or pictures of leaders of the Cuban-American community.

          They have a somewhat different system of racism, based on looks rather than the one drop rule, but it’s been in place for 500 years and isn’t disappearing. For example, Jorge Castaneda wrote in the Atlantic in the mid-1990s that Mexican elites have been getting progressively whiter over his lifetime. (Castandeda, the foreign minister of the 6′-5″ tall Vicente Fox, has blue eyes and looks rather like Don Johnson.)

          Consider where more skin-whitening cream is sold: the U.S. or Latin America? (Or India?)

        • “Consider where more skin-whitening cream is sold: the U.S. or Latin America? (Or India?)” or korea, where my colleagues are rather obsessed with their skin color. But that is not a good proxy for anything. For example, in USA whether you’re are white or not is orthogonal to your tanned skin – see, say, poor whites in texas. You need to look at other stuff people might be adopting, such as “anglo” names. Also in places like Brazil, while is good to be white, it is often ugly to be pale – but, on the other hand, people might be concerned about their curly hair.

          For me, as someone that was raised in Brazil, who has family in Europe and have been living in USA for a few years now, here (in USA) seems rather obsessed with race, however that means. They never have laws like you guys had here or stuff like on drop rule in Brazil, Argentina or Europe recently – perhaps Nazi Germany. Discrimination is rather different in these other places. Anyway, my point is that you USA people challenge the white status of the, say, cuban immigrants and, in doing so provoke a reaction on them.

        • My point is that while discrimination exists everywhere – though at different degrees and against different people – USA is vey much obsessed with racial classification, in putting someone is a particular box, more than any country I know of. People ask you race even when you enroll in a gym – and this is pretty unique, I think.

        • A major reason why contemporary America obsessed with racial classifications is because America awards money and prizes to people who put themselves in protected categories. Brazil, in contrast, had no affirmative action programs until this century, so there had been less need for bright line distinctions.

          Racial categorization in the U.S. is based on self-identification. This works good enough for government work for blacks due to the historic one-drop rule and relatively lows rates of intermarriage (although as Gates and Guinier pointed out, it’s breaking down at Harvard, where most of the black beneficiaries are more like Barack than like Michelle Obama.) For others, its becoming more dubious (e.g., Senator Elizabeth Warren being celebrated at Harvard for being an American Indian professor).

          When Brazil passed an affirmative action law for college admissions a number of years ago, the lack of a tradition of a one drop rule raised suspicions that self-identification would lead to massive cheating. So, Brazil uses visual inspection to classify college applicants as to whether they qualify for affirmative action. In one notorious case, two identical twins were separately inspected and put into different categories.

  8. Is Steve Sailor a racist? This is a strong statement and I don’t want to make this accusation. And, for sure, this is not the point of this blog – or of any blog I read. That said, a quick look at Sailor’s blog does show a private crusade against hispanics. It is different than the usual ones. He attacks the white hispanic elites as oppose to the brown people that we often associate with hispanics. But it is there: just check.

  9. I don’t know about race. I do wish, however, that Congress would lift the ban on the Census asking about religion. Not that the Constitution really asks for it or anything, but I’d love a census tract or blockgroup breakdown by religion.

    • Right. And since the Census doesn’t ask about religion, it could not ask about race or (especially) ethnicity as well. Except that by awarding money and prizes based on race and Hispanic ethnicity for the last four decades, the federal government has by now created powerful established interests that fight hard to protect their legal privileges.

  10. Another comment thread derailed by the “pretty normal” (ha!) Steve Sailer. Those of you that have engaged with that delusional and self-aggrandizing lunatic have the patience of a saint.

    Cheers.

      • Actually, he hasn’t.

        He has repeatedly gone back to his resentment of ethnic elites’ attempts to gain access to the kind of privilege that more established groups get through their social networks. He has repeatedly gone back to his simplified conclusion-driven version of events. He has repeatedly gone back to insulting those who seek to engage with him (primarily me) by mistaking engagement for ignorance.

        But more directly to your point, has has ducked or ignored most of the questions I have posed. Particularly, he has avoided answering questions intended to scaffold his linking his rants back to the topic of Gelman’s blog post.

        So, not reasonable. And not really answering.

Comments are closed.