The Unz Review: An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersiSteve Blog
National Merit Semifinalist Cutoff Scores by State
🔊 Listen RSS
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeLOLTroll
These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

Posting below on affirmative action by state in Nigeria, I was reminded that the cutoffs for being a National Merit Semifinalist in the U.S. vary by state. Semifinalists are the top 1% scorers on the PSAT. An anonymous iSteve commenter tracked down the data:

From here:

https://www.compassprep.com/national-merit-semifinalist-cutoffs/

I extracted the data from the table for the 50 states and D.C., and then sorted by the class of 2019 state score cutoffs, descending, with Asian percentage of population in parentheses:

223 (14.7%) California
223 (4.21%) District of Columbia
223 (6.53%) Maryland
223 (6.56%) Massachusetts
223 (9.70%) New Jersey
223 — Outside US
222 (4.63%) Connecticut
222 (3.91%) Delaware
222 (6.51%) Virginia
222 (8.41%) Washington
221 (3.24%) Colorado
221 (5.50%) Illinois
221 (8.76%) New York
221 (4.40%) Oregon
221 (4.73%) Texas
220 (3.39%) Arizona
220 (4.01%) Georgia
220 (37.3%) Hawaii
220 (4.88%) Minnesota
220 (2.83%) North Carolina
220 (3.40%) Pennsylvania
220 (3.59%) Rhode Island
219 (2.84%) Florida
219 (2.14%) Indiana
219 (3.00%) Michigan
219 (2.62%) New Hampshire
219 (2.11%) Ohio
219 (1.78%) Tennessee
218 (2.90%) Kansas
218 (1.45%) Kentucky
218 (8.54%) Nevada
217 (1.84%) Louisiana
217 (1.23%) Maine
217 (2.00%) Missouri
216 (1.38%) Alabama
216 (2.35%) Iowa
216 (2.35%) Nebraska
216 (1.59%) South Carolina
216 (1.64%) Vermont
216 (2.77%) Wisconsin
215 (6.26%) Alaska
215 (1.06%) Mississippi
215 (1.72%) New Mexico
215 (2.19%) Oklahoma
215 (1.38%) South Dakota
215 (2.49%) Utah
214 (1.58%) Arkansas
214 (1.49%) Idaho
214 (0.83%) Montana
212 (1.40%) North Dakota
212 (0.84%) West Virginia
212 (1.04%) Wyoming
212 — U.S. Territories

I don’t see any overwhelming patterns. It would be interesting to add data columns for state population, Non-Asian minority percentage, urban vs. rural, percentage of pickup truck owners, highest educational level obtained times number of tattoos factor, closeness to Canadian border, etc.

This is basically a measure of how smart the smartest kids are in each state, not the average student. The District of Columbia’s public school students, for example, have terrible median test scores on the federal NAEP test, but the best DC students, most of whom no doubt go to private schools, are quite smart. Something similar is true for California.

The top states on the list have or are next to big, rich cities, while the bottom states don’t have big rich cities.

Okay, anybody know anything about affirmative action by state and ethnicity in China? I hear the highest test scores are in Fujian province on the coast opposite Taiwan, between Shanghai and Hong Kong.

 
Hide 78 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
    []
  1. No surprises that I can see (though I wish DC weren’t always included with states) and I agree that additional columns would add interest. Surely we have a quant-sperg or two who could step up? That is what they’re calling them nowadays right?

  2. DC’s public schools are the worst. Lowest median SAT, lowest reading scores, 2nd-lowest math, highest dropout rate.

    Yeah, those St Albans and Cathedral kids are pretty smart.

    Georgetown Prep isn’t even top 10 in the DC area private schools, and our two most recent Supreme Court justices went there. During the Kavanaugh hearings I was scolded by a St Albans and Maret parent when I said G Prep kids sometimes attended parties with St Albans kids ( which they do). She said they were a different ‘tier.’

  3. Those cutoffs are strongly correlated to the % Asian by state. Someone else might want to do the work…http://worldpopulationreview.com/states/asian-population/

    • Replies: @Anonymous
    , @SFG
  4. Which states have the smartest dumb kids?

    • Replies: @Lot
  5. Lot says:
    @Reg Cæsar

    “Which states have the smartest dumb kids?”

    Probably New Hampshire, Vermont, Iowa, and North Dakota, in that order.

    • Replies: @Mr McKenna
    , @Reg Cæsar
  6. Anon[114] • Disclaimer says:

    Okay, anybody know anything about affirmative action by state and ethnicity in China? I hear the highest test scores are in Fujian province on the coast opposite Taiwan, between Shanghai and Hong Kong.

    I know something about this, though more along the lines of “what the system is” than “what are people scoring where”.

    The system as you generally see it described in the American press is as follows: high school students take the gaokao (高考), apply to colleges, and colleges admit whoever has the highest gaokao scores.

    There’s considerably more to it than that.

    Affirmative action by ethnicity:

    The minority ethnic groups (少数民族) are given a bonus to their gaokao score. I’m told the bonus is “significant”, but I don’t have any sense of exactly how much of a boost it is. If you’re a Mongol, your gaokao score will be higher than it would have been if you’d turned in the same test as a Han. After that, you’ll be admitted or not based on your (adjusted) gaokao score; your ethnicity won’t factor in further.

    Affirmative action by state:

    China is divided into several provinces (省), plus four province-level cities. Universities have per-major, per-province quotas, which they publish every year under the name 招生计划 (“student recruitment plan”). The plan might, for example, specify that the university will admit 20 math majors from Beijing, 10 math majors from Gansu, 15 math majors from Shandong… like that. Obviously, you need to apply to a specific major.

    Every university has relatively large quotas for its own local region, meaning the admittance threshold is notably lower for local students. China encourages you to go to school near your home. The top schools are in Beijing and Shanghai, so students from those cities get a prestige advantage through this system.

    Beyond that, there are two other aspects of the system that might be considered “affirmative action by state”:

    1. Every province is notionally entitled to use its own gaokao. Two provinces might give two entirely different tests under the name gaokao. The Beijing gaokao is supposedly easier than a normal gaokao. In practice, most provinces use a shared gaokao. And, because admittance is done on a quota-by-province basis, I don’t see how different provinces giving tests of different difficulty would have any real effect on who gets admitted where — everyone you’re competing against took the same test you did.

    2. The quotas do not necessarily reflect the population of the provinces. Applying to college as a student from Shandong is legendarily competitive, because the quotas for Shandong are too small. Applying from Beijing is much easier, because quotas for Beijing are inflated.

    Other considerations:

    1. You have to apply to college before you learn what your gaokao score was. Colleges publish their admittance thresholds (分数线) from previous years; you work with a counselor to predict (1) what your score is likely to be and (2) how the thresholds this year are likely to have changed compared to last year, and you apply to a school based on that.

    2. You only get to apply to one college (again, without knowing what your gaokao score is!). More accurately, you can apply to more than one, but you have to list one as your first choice. When admissions for the year are decided, the university will fix a 分数线, admit all the students (up to quota) above that threshold who listed that university as their first choice, and only if their quota is still not filled will they then dip into the pool of second-choice listers, even if some of those students have higher scores than some who were admitted as first-choice listers. Unfortunately, I have no idea how the 分数线 is set; by adjusting it upward, it seems like a university could bias their student scores upward at the “expense” of admitting more students who didn’t have that particular school as their first choice.

    3. There are side channels. Universities may offer their own admission tests, which can substitute for the gaokao, or they may have special relationships with particular high schools. For example, the immensely prestigious Fudan University (复旦大学) in Shanghai is affiliated with the unimaginatively named High School Affiliated With Fudan University. If you go to one of those schools, a teacher recommendation can mean the affiliated university will admit you with a gaokao score significantly below the normal 分数线. When I surveyed four students at competitive schools (yes, small n), two of them told me they had been admitted based on raw gaokao prowess and two said they had come through a side channel.

    Historical interest:

    The south of China has been doing better than the north on standardized tests and in commerce for more than a thousand years. The pattern seems to have begun in the Tang dynasty, when the Han first settled the Yangtze River basin. The north bore the brunt of defending everyone against barbarians from the steppe; the south focused on economic production. There was an emperor in the Ming dynasty who, disgusted with the fact that all the people passing the highest level of the civil service examination were all Southerners, enacted some pro-Northerner affirmative action.

  7. @Lot

    Yep, that’s more likely to correlate with the Dreaded Whiteness.

  8. Anonymous[270] • Disclaimer says:

    Worth mentioning that these scores are not very different. Basically, 210 to 220. 1sigma for PSAT is something like 30.

    • Agree: lavoisier
    • Replies: @Anonymous
    , @J.Ross
  9. neir says:

    highest IQ province in china is Zhejiang, its average IQ is 118, it produces most of china’s billionires including Alibaba funder Jack Ma.
    It has the capital of the southern Song dynasty, which was when most of the northern chinese elites fled to Zhejiang during the mongol invasion.

    https://money.cnn.com/2016/09/27/technology/zhejiang-hangzhou-billionaires/index.html

    Zhejiang is china’s home of mathematics

    https://en.wikipedia.org/wiki/Wenzhou#Home_of_Mathematics_in_China

    • Replies: @EdwardM
  10. neir says:

    Zhejiang entire province’s average IQ is higher than Shanghai, Beijing or Hong Kong

    http://wcipeg.com/etc/brian/China_IQ_by_province.svg

    • Replies: @jon
    , @Desiderius
  11. Anonymous[251] • Disclaimer says:
    @International Jew

    Hawaii? That’s pretty Asian and it is in the middle of the list.

    Looks more like population density, though as a metric, population density is derived in a ham-fisted way if I understand it correctly (population/square miles.) In some remote places that works well because the rural dwellers are fairly evenly distributed. But what about places like Alaska or more extremely Greenland. Small population/ enormous acreage shows very low population density, yet most of the people are actually clustered in small cities and the sprawling countryside doesn’t amount to much of their daily lives.

    So maybe it’s more like population density times percent of urban dwellers.

    • Replies: @Wilkey
  12. Anon[396] • Disclaimer says:

    Slight correction: The post contains Class of 2017 data, not Class of 2019 data. I chose the wrong column. Here’s a new version with 2019 data. In addition I added the data for Asian percentage of state population, to two decimals, or one decimal for states with a two-digit percentage (to keep the data vertically alligned). Also, in response to the complaint about including D.C., … I have retained D.C. and added U.S. Territories and expat/overseas military kids. The former group ties with the worst states, the latter with the best states. The Asian percentages don’t precisely track PSAT scores, but comparing the top half of the results to the bottom half, the differnce is quite clear.

    223 (14.7%) California
    223 (4.21%) District of Columbia
    223 (6.53%) Maryland
    223 (6.56%) Massachusetts
    223 (9.70%) New Jersey
    223 — Outside US
    222 (4.63%) Connecticut
    222 (3.91%) Delaware
    222 (6.51%) Virginia
    222 (8.41%) Washington
    221 (3.24%) Colorado
    221 (5.50%) Illinois
    221 (8.76%) New York
    221 (4.40%) Oregon
    221 (4.73%) Texas
    220 (3.39%) Arizona
    220 (4.01%) Georgia
    220 (37.3%) Hawaii
    220 (4.88%) Minnesota
    220 (2.83%) North Carolina
    220 (3.40%) Pennsylvania
    220 (3.59%) Rhode Island
    219 (2.84%) Florida
    219 (2.14%) Indiana
    219 (3.00%) Michigan
    219 (2.62%) New Hampshire
    219 (2.11%) Ohio
    219 (1.78%) Tennessee
    218 (2.90%) Kansas
    218 (1.45%) Kentucky
    218 (8.54%) Nevada
    217 (1.84%) Louisiana
    217 (1.23%) Maine
    217 (2.00%) Missouri
    216 (1.38%) Alabama
    216 (2.35%) Iowa
    216 (2.35%) Nebraska
    216 (1.59%) South Carolina
    216 (1.64%) Vermont
    216 (2.77%) Wisconsin
    215 (6.26%) Alaska
    215 (1.06%) Mississippi
    215 (1.72%) New Mexico
    215 (2.19%) Oklahoma
    215 (1.38%) South Dakota
    215 (2.49%) Utah
    214 (1.58%) Arkansas
    214 (1.49%) Idaho
    214 (0.83%) Montana
    212 (1.40%) North Dakota
    212 (0.84%) West Virginia
    212 (1.04%) Wyoming
    212 — U.S. Territories

    • Replies: @Anon
    , @Almost Missouri
  13. @Ghost of Bull Moose

    “Georgetown Prep isn’t even top 10 in the DC area private schools”.

    Academically? That would have surprised me once, but I have little doubt that the neo-Catholic commitment to “diversity” will have begun to show its baleful effects by now.

    On the other hand I suspect your parent was referring to class, not brains. And, proud grandson of a Georgetown Prep graduate (1908) though I may be, I can only agree with her. I have no doubt that from the old-fashioned Social Register point of view (which I share), St Albans was and remains well ahead of Georgetown or any other Catholic private school.
    In my own extended family, one couple sent their two girls to Madeira; their brother went to Georgetown. Another couple has a boy at Landon; yet another have twins, a boy and a girl, both now at Holy Trinity, the Jesuit parish school in Georgetown, where they live. It will be interesting to see what they decide on for secondary level. The husband is not Catholic, so I can well imagine that somewhere like StAlbans or Landon will be on their list. Or perhaps Sidwell Friends, as they are painfully PC Obama lovers.

    • Replies: @Tim
    , @Tim
  14. jon says:
    @neir

    Only 77.3 IQ for Tibet?

  15. How about grape soda consumption per capita? Though DC would mess it up.

  16. per my research over the last 20 years, the number of national merit scholars is now going…down. that seems to be a recent trend over the last few years, not the trend over 2 decades. i’ll check again with the 2019 numbers. but year to year the total seems to be going down.

    there’s about 7000 per year, but it’s declining now. just like SAT performance. which started going down a few years ago. except for ‘asians’. which i assume is about smart people from china and india moving to the US.

    if it really is based on some percentage of total participants in a state, then it might simply mean less teenagers are taking PSAT. maybe they are skipping it or maybe they are only taking ACT. but in my opinion, there are probably just plain less national merit scholar level students in the high schools now, regardless.

    so the smart fraction is declining.

    NMS is about equivalent to 140 on a wechsler. or it was. maybe it’s lower now.

    • Replies: @res
    , @S. Anonyia
  17. Anon[396] • Disclaimer says:
    @Anon

    Nevada and Alaska are the high-Asian outliers: relatively low scores. Nevada has a lot of Samoans, and the low scoring U.S. Territories includes Samoa. So if Asians includes Pacific Islanders, that could be an explanation. As for Alaska, about half of the Asians are Filipino, I believe.

    https://en.wikipedia.org/wiki/Filipinos_in_Alaska

  18. SFG says:
    @International Jew

    ‘Asian’ is a blunt instrument.

    What you’re probably seeing is the spillover of the highly skilled immigration from India and China starting after Hart-Celler–their kids started showing up in the 80s or so. I’d correlate with that, and probably toss in ‘South Korean’, not so much Japanese–ironically, their country already being rich meant they had a lot less need to migrate.

  19. here’s an old calculation i did on the 2013 data set.

    column 1: national merit scholars in the state that year
    column 2: the state
    column 3: number of people in the state that year, in thousands
    column 4: number of people per national merit scholar, in thousands

    vermont did the best. 1 scholar per 14,900 people.

    nevada did the worst. 1 scholar per 30,200 people.

    i used to run this on the data set every year. over time, you could see states slowly getting less smart, or vice versa. it wasn’t a huge change, but trends were emerging after 10 years of analysis.

    should be easy to run the report on the 2019 data set. things won’t change much.

    sorry, no idea at all why that image is so huge. please shrink appropriately, web master.

    • Replies: @Almost Missouri
    , @res
  20. “Nevada and Alaska are the high-Asian outliers: relatively low scores.”

    nevada results have nothing to do with asian low performance.

    nevada has the worst high schools in america. the high school dropout rate is 50%. why?

    because you can drop out of high school and start working at a casino, doing unskilled labor, and make 50 grand a year in only a few years. with your food, clothes, and parking paid for by the casinos. for 30 hours a week of work.

    i lived there for 4 years. saw tons of high school dropout level casino workers who made 60, 70 grand a year after 10 years dealing cards. even the valet guys make 40 a year parking cars. for zero skill work.

    table worker pay seemed to top out around 85k per year at the bellagio and wynn back then, it might be higher now.

    what’s the point of even taking the SAT and fighting your way into a career for less money at 40 to 50 hours a week, when you can do the casino instead?

  21. wo_fat says:

    so the two legged biped who did this was a “jounralist”?
    no, the proper way to do this, and omg, this pains us, is to to have what is called a “hypothesis”, such as racists whites are better at psat tests vs. poor oppresed coloured people.. well, am sure one could use raw data to support their pet hypothesis, but this “person” does not evey try to come up with a “hypothesis”. Please look that up, young man.

  22. Hodag says:

    Not being Chinese I have no inside knowledge but there is a stereotype of Fujian dwellers as notorious swindlers and cheats.

    • Replies: @neir
  23. Nathan says:

    I think boomers tend to overrate the National Merit Scholarship and the PSAT. I can tell you that nobody at a public high school in my home state takes it. Other than bragging rights- which you can get out of the plain old SAT- there’s no reason to. Only a handful of colleges offer any kind of financial award for National Merit Scholars, and they aren’t the schools national merit kids want to go to:

    https://www.college-kickstart.com/blog/item/colleges-with-great-scholarships-for-national-merit-finalists

    http://www.thecollegesolution.com/realities-national-merit-scholarships/

    • Replies: @slumber_j
  24. neir says:
    @Hodag

    That’s because Fujian used to be a poor province, most of the interior Chinese provinces like Henan are stereotyped to be full of swindlers and cheats and xinjiang is stereotyped to be full of thieves. Some people say CCP has purposely kept Fujian poor and backward to prevent any collaboration between Fujianese and Taiwanese.

  25. MSP says:

    The D.C. score isn’t calculated; it’s simply set at the highest value from any state.

    • Replies: @res
  26. @Anon

    I suspect that the Asian correlation is mostly the side effect of Asians favoring the same high-density super-cities that produce the high PSAT scores. Though no doubt Asians are over represented among high scorers.

    • Replies: @stillCARealist
  27. @prime noticer

    Yeah, no surprises in this list. Except maybe how badly things tail off in Nevada.

    It would be interesting to have same list but for only white+Asian population, to remove some of the racial confounds.

  28. @Ghost of Bull Moose

    The District of Columbia’s public school students, for example, have terrible median test scores on the federal NAEP test, but the best DC students, most of whom no doubt go to private schools, are quite smart.

    I’d guess that a good proportion – perhaps a majority – of the best DC students in those private schools are resident in affluent surrounding areas in MD and VA, skewing the results of all three.

  29. TP says:

    @primenoticer

    The number of National Merit Semi-Finalists is always 1% of the top scorers on the junior year PSAT in each state. This year there were around 16,000 NMSF.

    To become a National Merit Finalist, you have to complete an application and submit a confirming SAT score, usually 1500+. This year there were approximately 15,000 NMF.

    To beomce a National Merit Scholar, you also need to receive a scholarship from one of three entities: the National Merit Corporation, Participating Companies or Participating Colleges. This is usually around 7,500 students.

    If you are only looking at National Merit Scholars, you are missing half of the top scoring students.

    • Replies: @prime noticer
  30. This range of cutoffs, from 212 to 223, seems surprisingly narrow to me, only 5% difference between the highest and the lowest. To a pretty good approximation, there is no difference. The flatness of the distribution is more worth theorizing than why California and Maryland are higher on the very gentle slope than Wyoming and North Dakota.

    • Replies: @res
  31. TP says:

    The calculation for the PSAT/NMSQT to determine the cutoff for each state is:

    (ERW+ERW+MATH)/100

    I’m not sure why language arts gets double-weighted but that is the calculation.

    The top 50,000 scorers nationwide are awarded National Merit Commended status. The Commended cutoff in 2009 was 208, in 2019 it was 2012 so the delta was +4. I also compared the state cutoffs from 2009 to 2019 and the changes are found below. Along with the Commended cutoff, almost all of the state cutoffs have gone up as well. Only Alaska and Vermont have gone down.

    State DELTA
    Commended 4

    Alabama 2
    Alaska -1
    Arizona 6
    Arkansas 3
    California 4
    Colorado 4
    Connecticut 3
    Delaware 2
    District of Columbia 2
    Florida 3
    Georgia 2
    Hawaii 2
    Idaho 0
    Illinois 4
    Indiana 2
    Iowa 1
    Kansas 2
    Kentucky 4
    Louisiana 3
    Maine 1
    Maryland 3
    Massachusetts 2
    Michigan 5
    Minnesota 3
    Mississippi 7
    Missouri 0
    Montana 0
    Nebraska 4
    Nevada 6
    New Hampshire 3
    New Jersey 3
    New Mexico 1
    New York 3
    North Carolina 2
    North Dakota 4
    Ohio 2
    Oklahoma 1
    Oregon 4
    Pennsylvania 3
    Rhode Island 3
    South Carolina 0
    South Dakota 3
    Tennessee 2
    Texas 3
    Utah 5
    Vermont -1
    Virginia 2
    Washington 3
    West Virginia 2
    Wisconsin 1
    Wyoming 4

  32. @Lot

    I wonder if Hawaii would join them.

    • Replies: @Anonymous
  33. Anonymous[264] • Disclaimer says:
    @Anonymous

    “Worth mentioning that these scores are not very different. Basically, 210 to 220. 1sigma for PSAT is something like 30.”

    That’s what I was thinking.

  34. Anon[427] • Disclaimer says:

    First: these are not the official cutoffs. They are only estimates.

    Second: The “Asian percentage” is extremely misleading as it takes the Asian pop. as a share of the total population. The accurate measure should be for that age group. I think the Asian share under 18 is much larger than the Asian share in the general population. In Western WA for example, Asians make up 8% of the total population, but anywhere from 25% – 75% of the student body in any given school.

    • Replies: @res
  35. Anon[427] • Disclaimer says:

    Another strange thing about National Merit Scholar: the verbal portion counts twice as much as the math portion, this gives an edge to whites esp. Jews vs. Asians.

    Strangely though, in our local schools, practically all the NMS finalists are still Asian.

  36. Anonymous[232] • Disclaimer says:
    @Reg Cæsar

    No it would not. Hawaii is absolutely filled with pigin-speaking morons.

    Asking, “which states have the smartest dumb kids?” is basically asking which states are almost all white. It would in theory work to ask “which states are almost all East Asian?” but of course there aren’t any.

  37. EdwardM says:
    @neir

    People from Wenzhou are also known as the “Jews of China.”

  38. res says:
    @prime noticer

    per my research over the last 20 years, the number of national merit scholars is now going…down. that seems to be a recent trend over the last few years, not the trend over 2 decades. i’ll check again with the 2019 numbers. but year to year the total seems to be going down.

    I’d be interested in seeing that research since NMSF is supposed to be the top 1% of students taking the PSAT. It might be better to look at changes in the qualifying scores if you are trying to see changes in student ability. Is it possible the number of students taking the PSAT is changing in a way that does not match the population change?

    Here is a table of the estimated US population by age from 2010-2017:
    https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk
    I see age 16 population varying ~2.5% over that time.

    but it’s declining now. just like SAT performance.

    Well, in 2017 average SAT scores went up significantly (about 50 points). I think it is hard to figure out what is really going on given the score shenanigans.

  39. res says:
    @prime noticer

    I think you need to use the number of test takers rather than the total population for each state.

    https://en.wikipedia.org/wiki/National_Merit_Scholarship_Program

    About 1.6 million students in some 22,000 high schools enter the National Merit Scholarship competition annually when they take the Preliminary SAT/National Merit Scholarship Qualifying Test (PSAT/NMSQT).[5] This serves to screen program entrants, measuring critical reading ability, mathematics problem-solving ability, and writing ability, rather than existing knowledge.[6] Semifinalists are designated on a state representational basis, contingent on the total number of entrants and in proportion to each state’s percentage of the nation’s high school graduating seniors.[7] Semifinalists are the highest-scoring program entrants in each state and represent the top 0.5% percent of the state’s senior students.[8]

    I suspect your analysis is more a look at age demographics for the states.

    • Agree: International Jew
  40. @Almost Missouri

    You should see the last names for NMS kids in CA. “Over represented”? Try totally dominant.

    I helped proctor the AP Calc exam the other day at Folsom High and since I had 3 hours to observe a set of their smart kids, I noted their demographics. Out of 83 students taking the test, only 27 were “white”. The rest were some sort of Asian, including the Middle East. No blacks or Hispanics that I could tell. The school itself is 62% white (very unusual by CA standards) but only 32% of AP Calc exam takers are.

    In 2017 my kid missed the national merit by one point. Too bad we didn’t live in AZ or NV!

    For 2018 I see that my locality had no NMS whatsoever, despite having three quite large high schools. Lame.

    • Replies: @International Jew
  41. Anon[300] • Disclaimer says:

    “I don’t see any overwhelming patterns”. But I do.

    I see that people from overseas are held to the same standard as the most stringent US states standards.

    I see purest-white states scraping the bottom. It’s not a party issue, even Vermont is near the bottom.

    I see DC (supposedly filled with incompetent bn) at the top.

    I see Porto Rico (presumably the bulk of US territories) and New Mexico near the bottom. Send this to Unz.

  42. @prime noticer

    Younger generations, including whites and Asians, are getting dumber. Movie dialogue is dumbed down compared to a decade ago. Long books don’t sell well anymore, fewer people are reading period. Also even in private schools there is a shift towards the bullshit that is “project-based learning”….aka make school fun and there’s no pressure to memorize hardly anything.

    Also likely reflects a shift towards the ACT since it’s becoming more popular. ACT is a little bit easier to game with tutoring than the SAT.

  43. res says:

    One thing that is annoying about trying to interpret the NMSF threshold data is that the selection index used is different from the PSAT scores which are generally reported. Some information resources for working with this data.

    National Merit Scholarship overview: https://en.wikipedia.org/wiki/National_Merit_Scholarship_Program

    2018 College Board report for their tests:
    https://reports.collegeboard.org/pdf/2018-total-group-sat-suite-assessments-annual-report.pdf
    Notice that they focus on PSAT scores of the form:
    PSAT/NMSQT and PSAT10 sections: 160–760
    and the combined scores with a range of 320-1520

    Looking at page 9 we see that the PSAT actually has three tests: Writing and Language, Math, Reading
    Each test is scored 8-38. The Math section score is simply the Math test score times 20. I am guessing the ERW section score is a weighted average of the Reading and “Writing and Language” test scores times 20, but I do not know that for certain.
    (I find their “test” and “section” terminology confusing, I hope the above was clear)

    Here is how the Selection Index (SI) is computed:
    https://blog.collegevine.com/what-psat-score-do-you-need-to-qualify-for-national-merit/

    After you take the PSAT, you’ll receive a score report listing your test score out of a possible 1520 points. However, that’s not the scoring system that matters for the purposes of National Merit eligibility. Instead, National Merit takes your Math, Reading, and Writing subscores from the PSAT, adds them up, and multiplies them by 2. The resulting number, which falls between 48 and 228, is known as your Selection Index (SI).

    More detailed information on the PSAT:
    https://collegereadiness.collegeboard.org/pdf/psat-nmsqt-understanding-scores.pdf
    Page 12 has a raw score to test score conversion table which I found interesting.

    One interesting thing about all of this is that the PSAT composite weights math and verbal equally while the National Merit selection index double weights verbal (note this is especially relevant in the context of the following paper about “gender inequality”).

    Does anyone have statistical data (mean and SD) for the Selection Index? This is the best reference I found: https://files.eric.ed.gov/fulltext/EJ1011761.pdf
    but I think it raises as many questions as it answers.

    They give mean/SD for the individual tests by sex in 2010-2011
    Section Male Female
    Reading 47.2/11.9 47.4/11.1
    Mathematics 50.3/12.2 47.7/11.0
    Writing 44.7/11.8 46.1/11.3

    Unfortunately they don’t give data for the combined scores. The means are 142.2 and 141.2, but we can’t calculate the SDs because the variables are correlated. It is worth noting that the male SDs are 5-10% larger which will result in more males in the far right tail.

    For rough estimate purposes let’s assume the national selection index average of 210 is the 0.5% level. That would be about 2.6 SD so taking (210 – 141.7) / 2.6 = 26 is probably a pretty good estimate for the SD (close to the 30 mentioned in an earlier comment above)

    They are looking at a doubled version of the 8-38 test scores above. But they claim a max of 80 for each test giving a max for the Selection Index as a whole of 240. Compared to the 6*38 = 228 I would expect from the earlier links above. Does anyone know what is going on here?

    One thing I found very interesting is in Table 4 they compare three states (NY, CA, and TX). There is an almost 3x difference between TX and CA for the percentage predicted at or above state cutoff score (roughly 0.5% vs 1.4%). Given that they are supposed to be taking the top 0.5% by state, this is very odd. Two possible explanations: unusual distributions invalidating the assumption of normality, or a real difference in percentage selected.

    I wish I had access to the dataset used to write that gender report.

  44. res says:
    @MSP

    The D.C. score isn’t calculated; it’s simply set at the highest value from any state.

    What do you base this on?

  45. res says:
    @John Mansfield

    Better to look at it in terms of standard deviations. With an SD of 26 (estimated in earlier comment) the 11 point SI cutoff range is about 0.4 SD. Not that much in absolute terms, but it can make a big difference in population frequency. Since we are looking at about 2.6 SD (0.5%) the difference between that and 3.0 SD (0.13%) is substantial (3x) in terms of population frequency.

    • Replies: @John Mansfield
  46. res says:
    @Anon

    First: these are not the official cutoffs. They are only estimates.

    If you follow the link the class of 2019 cutoffs (given above) are actual, the class of 2020 cutoffs are estimates.

    The “Asian percentage” is extremely misleading as it takes the Asian pop. as a share of the total population.

    Good point.

  47. @res

    Thanks for the explanation of the numbers. So, as far as distributions go, it would be analogous to a 145 IQ cutoff versus 139.

    • Agree: res
  48. Wilkey says:
    @Anonymous

    “Hawaii? That’s pretty Asian and it is in the middle of the list.”

    Hawaiian Asians are not regular Asians. Many of them are Pacific-Islander. Many or most of the rest are descended from Japanese coolies who came over to work the sugar plantations. Not any sort of Tiger Mom ethic there. I wouldn’t be shocked if Hawaii is one state where whites do better than Asians on SAT scores, PSAT, etc.

    • Replies: @Triumph104
    , @GermanReader2
  49. Tim says:
    @Old Palo Altan

    “In my own extended family, one couple sent their two girls to Madeira; their brother went to Georgetown. Another couple has a boy at Landon; yet another have twins, a boy and a girl, both now at Holy Trinity, the Jesuit parish school in Georgetown, where they live. It will be interesting to see what they decide on for secondary level. The husband is not Catholic, so I can well imagine that somewhere like StAlbans or Landon will be on their list. Or perhaps Sidwell Friends, as they are painfully PC Obama lovers.”

    Man, so much there. I feel like I know all these people. But sending your girls to Madeira is a terrible idea. They will learn how to ride a horse, but they rarely go to good colleges.

    Landon is good no doubt.

    Holy Trinity–Oy! The education is excellent, but those Jesuits are insufferable. I went to the 5:30 Mass one time and all the Priest did was explain how Jesus didn’t say this or that, and this was slipped in later.

    The schools no one has mentioned are some of the best–Cathedral School, Stone Ridge (my Mom went there), Visitation, and Gonzaga (my Dad went there). I don’t think any of those schools are less than $25,000 a year, I was told by a girl who goes to Visitation (Visi) is if you play on the girls La Crosse team, you WILL GO to Yale.

    Last comment, I don’t know much about Sidwell Friends, accept it is THE SCHOOL for non-Catholics to get into. I did know one teacher there. He was a black guy who is exactly like you’d think a black teacher at Sidwell Friends would be like–affable. He was like Theo Huxtible, but, you know, without the rapiness. . . as far as I know.

    • Replies: @Old Palo Altan
  50. Tim says:
    @Old Palo Altan

    “In my own extended family, one couple sent their two girls to Madeira; their brother went to Georgetown. Another couple has a boy at Landon; yet another have twins, a boy and a girl, both now at Holy Trinity, the Jesuit parish school in Georgetown, where they live. It will be interesting to see what they decide on for secondary level. The husband is not Catholic, so I can well imagine that somewhere like StAlbans or Landon will be on their list. Or perhaps Sidwell Friends, as they are painfully PC Obama lovers.”

    Man, so much there. I feel like I know all these people. But sending your girls to Madeira is a terrible idea. They will learn how to ride a horse, but they rarely go to good colleges.

    Landon is good no doubt.

    Holy Trinity–Oy! The education is excellent, but those Jesuits are insufferable. I went to the 5:30 Mass one time and all the Priest did was explain how Jesus didn’t say this or that, and this was slipped in later.

    The schools no one has mentioned are some of the best–Cathedral School, Stone Ridge (my Mom went there), Visitation, and Gonzaga (my Dad went there). I don’t think any of those schools are less than $25,000 a year, I was told by a girl who goes to Visitation (Visi) is if you play on the girls La Crosse team, you WILL GO to Yale.

  51. @stillCARealist

    Heh, and Folsom is still pretty white. You gotta see Cupertino and Sunnyvale.

  52. “I think boomers tend to overrate the National Merit Scholarship and the PSAT.”

    no way. it was the standard for sure. i can post extensively on how important this metric actually is. it’s probably the single most important statistic we have on national brainpower. it might be declining in value today though. that’s true.

    i’m about to make a major post about national merit scholars at various universities.

    “Yeah, no surprises in this list.”

    over 10 years of running the analysis, my impression was that illionis and california were slowly slipping down the list. but i would need to run the data year by year to make sure that wasn’t statistical noise.

    Florida Man, however, is definitely a real thing. the people in florida are dumb.

    “Younger generations, including whites and Asians, are getting dumber.”

    the data overwhelmingly says this. except, new, smart east asians keep coming from china, to escape the dump country they’ve created. and there’s a billion of them, so the mean scores for east asians could keep slowly going up for a while, maybe towards wechlser 115.

  53. “I’d be interested in seeing that research since NMSF is supposed to be the top 1% of students taking the PSAT.”

    again, i’ll clarify. i’m not talking about semi-finalists. i’m talking about finalists.

    like i said, i want to run it every year for 10 years or more, to make sure this isn’t statistical noise. in the 00s it wasn’t changing much, but in the 10s, it looked like the numbers started slipping a bit.

    this could be because teenagers are giving up on PSAT. deliberately not taking it, or just taking ACT. my SAT research says teenagers ARE slowly giving up SAT. i assume they are moving towards ACT. however, ACT scores are declining slightly too. so we know the general population of teenagers is getting less intelligent. that was my only unanswered question in the past. are the smarter kids giving up SAT for ACT. no. there’s just a general movement towards ACT, but it’s not intelligence based. they’re all getting less smart.

    “Well, in 2017 average SAT scores went up significantly”

    if that happened, it’s probably just david coleman bullshit. i don’t remember seeing that, but i’ll go in my tables and look again. i think we just posted on steve’s blog a year ago about how SAT scores were definitely now steadily going down, after a plateau in the 00s and early 10s.

    • Replies: @Redneck farmer
  54. anon[247] • Disclaimer says:
    @Ghost of Bull Moose

    DC’s public schools are the worst. Lowest median SAT, lowest reading scores, 2nd-lowest math, highest dropout rate.

    probably because of high % of “immigrants” in addition to the blegs

    when the globalists bring in their third world pets they usually place them in the capitol cities first

  55. @TP

    TP, thank you, but i was a national merit scholar, so i know how this works.

    semi-finalists don’t mean anything. a standard of 1%, or 1 in 100 test takers, is below the intelligence threshhold for the finalists. that is somewhere in the 130s on a wechsler. there is a review board which looks at each test taker and puts them into groups. to even get any scholarship money, you have to be about wechsler 140 or higher. and if we accept the data that the teenagers are getting less intelligent, then this wechsler number is going down every few years.

    not all the scholarships are the same. the national merit corporation gives out about 2000 per year itself, to the highest scorers. probably people doing 150 wechsler or higher. steve hsu has also talked about this, because he got one of those.

    the rest of the awards are sponsored by a corporation or a university, and those are less difficult to get, for the wechsler 140 range people like me. i was always stuck in the 140s on wechsler tests and was never able to get to 150. same with the SAT. i suck at math and the best i ever did was 710 math on the old SAT. i have answered every question correctly on the verbal, but i was only able to do that one time. so again, into the 1400s, but never to 1500 on a single test. i’m about to make a post about how many national merit scholars are at all the universities. the top universities do not sponsor any scholarships, so the NMS at those places are the smartest teenagers on average. the NMS at the other universities which do sponsor scholarships are not as smart on average, since it’s easier to get those. so even with NMS scholars, there is a hierarchy.

    on the old SAT, about 700 people per year were able to answer every question correctly on the math section. so basically, unless you were able to get to 800 math, you weren’t good at math, in absolute terms. that’s 700 per year, so over a decade, there’s 7000 people able to do this. the ceiling of the math test is below what lots of people can do.

    as always, it’s easy to get perspective on what this means if you think in terms of sports. if you’re a scout and you have 700 prospects per year who can all max out a test, then you’re nothing special at all if you’re the number 200 or 300 prospect. there’s 200 guys better than you, just in your class. and another 200 in the next, and so on. how excited would you be if your university signed the number 200 prospect in his class? nobody would care. which is exactly how math works too. if you’re below 800 math, you’re even worse than that. you suck at math. and that’s a fair description when we’re talking about real intelligence. this is a blog where people casually talk about gauss, feynman, newton, einstein, galton, von neumann, euler, like it’s no big thing. those guys would be once in a generation players if it was sports. the best player ever out of millions and millions.

    everybody knows what it feels like to get clobbered in sports, they have perspective and no illusion they’re actually good. i can tell you what it’s like to not exactly be dumb, but be getting crushed on the math olympiad and calculus BC exam by the wechsler 160, SAT 1550 guys, and then yes, when you experience that, you will definitely feel stupid. because you are. i’m definitely not smart.

    • Replies: @Desiderius
    , @TP
  56. eah says:

    She wouldn’t have made the cut in any state.

    • Replies: @J.Ross
  57. @Anon

    Another strange thing about National Merit Scholar: the verbal portion counts twice as much as the math portion, this gives an edge to whites esp. Jews vs. Asians.

    I think this is because it is usually much harder to get a 100% score in the verbal/language part of a general test than in a science or maths test, whereas anyone with a specialism in math or computer science would very easily ace a math test for the general population. So very likely the entire top 10% in math would have the same score.

  58. slumber_j says:
    @Nathan

    Only a handful of colleges offer any kind of financial award for National Merit Scholars

    Well, there’s the National Merit Scholarships themselves. I got $1,500 or something like that, sponsored by the Polaroid Corporation. Or maybe it was that much annually? I can’t remember.

    Anyway, compared with the cost of college that was paltry even in the mid-Eighties, but it was still ample reward for having taken the PSAT.

    • Replies: @Steve Sailer
    , @Desiderius
  59. J.Ross says: • Website
    @Anonymous

    The cutoffs for Nigeria were much more diverse.

  60. RickinJax says:

    18 years ago my daughter was a NMS (from NC) and got what amounted to free ride offers from more than 100 schools including many flag ship state schools.
    3 yrs under-grad and 3 yrs Law (Univ of Fl) cost us under $ 60,000 net.
    Maybe things have changed.

  61. J.Ross says: • Website
    @eah

    Bob Marley is such a white person answer too, one of the most irritating things a white person can do is affect or profess to dig things Jamic and not know who Peter Tosh or Jimmy Cliff are.

    • Replies: @Jonathan Mason
  62. @Tim

    You are right – neither of the Madeira girls went to good colleges. But they did find good husbands.

    Now I agree with you about the Jebbies – but, as I said, my cousin and her husband are Obama lovers and Trump haters, so I am sure they feel right at home.

    In any case, great to have the views of someone who knows the area as well as you do. Thanks, and how would you compare Georgetown Prep to Gonzaga these days?

  63. @prime noticer

    Scores are going down because people who used to be told you’re not college material are now expected to. So you get dumber people taking the test who really shouldn’t.

  64. Steve-O says:
    @Anon

    Wouldn’t doubling the verbal give a boost to females over males? I know we had a lot (5% or so) or semifinalists at my school and only 10% of those were female. Among my peers whose SAT (not PSAT) scores I knew, I only recall one male who scored high on verbal than math and no females who scored higher on math than verbal. The group was probably 12-16 people, and roughly 50/50 M/F.

  65. @slumber_j

    I got $1,000 bucks from Lockheed for being a National Merit Semifinalist. I vaguely recall that Rice’s tuition for 1976-77 was $2300, so that was about, maybe 9% of 4 years at Rice, which wasn’t bad at all.

  66. @neir

    Add another candidate to bug-out list.

  67. @slumber_j

    Pretty sure mine (scholarship winner) was $1,000 annually from P&G, where my mom worked. On top of full scholarship and co-op I think I used it to pay car insurance.

    • Replies: @slumber_j
  68. @prime noticer

    It’s hard to be super smart when you’re spending so much mental energy noticing. Good for wisdom tho.

  69. @Anon

    Wasn’t the extra 800 for the essay which has now been eliminated? When I took it in the 80’s top score was 1600.

    • Replies: @res
  70. slumber_j says:
    @Desiderius

    Yeah, maybe it was $1,000 annually. Anyway, I think tuition, room & board was like $18,000 even then where I went, so it wasn’t much of a dent. Still, nice to have.

  71. Marty T says:

    One clear pattern is Trump’s best states clustered at the bottom, while all the top states were Hillary. I guess not that surprising considering Trump’s rural and blue collar popularity. Though Wyoming and North Dakota are really no better than West Virginia?

  72. res says:
    @Desiderius

    The PSAT Selection Index for National Merit has always double weighted verbal. The nominal max (see my longer comments above) is 240 so we usually see SIs around 210-220.

    Why the verbal double weight is an interesting question. Any ideas?

  73. @Wilkey

    Whites score higher than Asians on the ACT in Hawaii. According to Table 1.5 scores for most ethnicities rose from 2014 to 2018. Whites went from 21.1 to 21.6, while Asians went from 18.9 to 20.0.

    http://www.act.org/content/dam/act/unsecured/documents/cccr2018/P_12_129999_S_S_N00_ACT-GCPR_Hawaii.pdf

    Profile reports for other states:
    http://www.act.org/content/act/en/research/reports/act-publications/condition-of-college-and-career-readiness-2018.html

    • Replies: @res
  74. The Laboratory of the States* Google Calc Spreadsheet is available for those wishing to pursue ecological hypotheses at the State level.

    *This is most of the data that used to be at laboratoryofthestates.com. I’ve a much larger collection of data at the county level in the Laboratory of the Counties on github.

  75. @J.Ross

    Bob Marley was mixed race like Kamala, but I am sure that Kamala’s favorite Marley song, (and mine), is One Drop.

  76. TP says:
    @prime noticer

    Okay, that makes more sense.

    I think the numbers break down approximately like this:

    16,000 – National Merit Semi-Finalists
    15,000 – National Merit Finalists
    7,000 – National Merit Scholarships (Corporations/Universities/National Merit Corporation)
    2,000 – National Merit Scholarships awarded by the National Merit Corporation

    If I understand correctly, your analysis is only looking at the students that received a scholarship from the National Merit corporation, 2,000 out of 1.5 million test takers. And just to confirm some of your assumptions for IQ, my daughter sits in the 130-135 range. She is a NMF and was awarded a full scholarship to UA, but she did not receive anything from the National Merit Corporation. She had a 1570 SAT (790 EWR/780 MATH), so I assume there were many students with perfect scores.

    Interestingly, I saw a recent report where the number of perfect ACT scores had greatly increased over the last ten years. Increasing popularity of the ACT since it is much easier, plus increased test prep were two of the reasons the author credited for the big increase.

  77. @Wilkey

    Hawaii is for Japanese what West Virginia is for whites.

Current Commenter
says:

Leave a Reply - Comments are moderated by iSteve, at whim.


 Remember My InformationWhy?
 Email Replies to my Comment
Submitted comments become the property of The Unz Review and may be republished elsewhere at the sole discretion of the latter
Subscribe to This Comment Thread via RSS Subscribe to All Steve Sailer Comments via RSS
PastClassics
Which superpower is more threatened by its “extractive elites”?
The “war hero” candidate buried information about POWs left behind in Vietnam.
What Was John McCain's True Wartime Record in Vietnam?
Are elite university admissions based on meritocracy and diversity as claimed?