Not-So-National Merit

Last December, thousands of high school sophomores and juniors learned the results of the 2013 Preliminary SAT (PSAT) test.  The juniors’ test scores will be used to determine whether they qualify as semifinalists for the prestigious National Merit Scholarship, which in turn makes them eligible for a host of automatic college scholarships.  (Sophomores take the test just as practice.)

The juniors will have to wait to find out for sure if they qualify until September, just before they begin submitting applications to colleges across the country.  But it is fairly straightforward to predict, based on their scores and last year’s cutoffs, whether they will qualify as semifinalists.

Many students would be surprised to learn that qualification depends not only on how high they score, but also on where they go to school.   The National Merit Scholarship Corporation (NMSC) sets different qualifying cutoffs for each state to “ensure that academically talented young people from all parts of the United States are included in this talent pool.”  They have not disclosed any specific criteria for setting the state cutoffs.

A high school student’s chances of receiving the award can depend crucially on his or her state of residence.  Last year, students in West Virginia needed only a 203 to qualify as a semifinalist (scores range from 60-240), while students from Texas needed a 219 and students from Washington, D.C. a 224.  Nationally, the West Virginia score was in the 97thpercentile of scores, while the Washington DC score was at the 99.5th percentile based on a mean score of 143 and a standard deviation of 31.

I’ve crudely estimated that because of this state cutoff discrimination, approximately 15% of students (about 2,400 students a year) who are awarded semifinalist status have lower scores than other students who were not semifinalists merely due to their geographic location.  Troublesomely, I also found that states with larger minority populations tend to have higher cutoffs.

Instead of just complaining, I have partnered with an extraordinary high-school sophomore from New Jersey named India Unger-Harquail to try to do something about it.

We’ve just launched a new website,  You can go to site, enter a score, and it will quickly tell you the states where your score would have qualified you as an NMSC semifinalist.

But wait, there’s more.  The site also offers to certify qualified students based on a national standard of merit.  If you represent and warrant to us that you received a PSAT score meeting the minimum cutoff in at least one state (and you give us the opportunity to try to verify the accuracy of your score with NMSC), we’ll give you the right to describe yourself as an “Acadium Scholar.”  We’ve separately applied to the USPTO to registrar that phrase as a certification mark (in parallel fashion to my earlier “fair employment mark”).

Instead of the yes-or-no signal offered by the NMSC, we’ll also certify students based on the number of states in which they would have qualified as semifinalists.  For example, a student who scored a 211 could be certified to describe herself as a “19-state Acadium Scholar.”

Our certification allows:

·         A student from a strong cutoff-state, like Texas, who scores a 218 (just missing the Lone Star qualifying cutoff of 219) to say nonetheless that he’s a 41-state Acadium Scholar.

·         A student from a weak cutoff state, like North Dakota, who scores an extraordinary 235 on the exam to say that she is a 50-state Acadium Scholar.

We’re even letting sophomores use their scores to certify so that all the pressure isn’t on junior year.  There are also some sophomores who may have scored ten points better in their sophomore than their junior year.  Now those students can certify as Acadium Scholars based on their higher scores.

Neither India nor I think that this location-based discrimination is the most momentous issue facing our nation.  And we don’t think there’s a single correct definition of merit. We’re just offering another way to measure merit—one that ignores the geographic boundaries within our country. Many people are surprised to learn that the “National Merit Scholarship” is really a state merit award, and not so “National” at all.

If receives more than a few hundred visits we’ll be surprised.  But we’d like to throw it back to you, Freakonomics nation, to guess how interesting this idea will be to Americans.  I’ll send a signed copy of Super Crunchers to whomever posts the best guess about how many people will post a score in the home-page widget before July 4th in the next week. Happy guessing!

[Don’t be confused.  The Acadium Scholar site is not affiliated in any way with the National Merit Scholarship Corporation or the PSAT/NMSQT.]


Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. m.m says:

    Hidden due to low comment rating. Click here to see.

    Disliked! Like or Dislike: Thumb up 14 Thumb down 24
    • plusECON says:

      Not what you’d be saying if you scored a 218 in Texas…

      Ian, my guess is 3300. Good luck – I think it’s a noble effort.

      Thumb up 5 Thumb down 3
  2. Matt says:

    I’m somewhat curious about why you take issue with the geographic affirmative action that NMSC apparently practices, particularly since one of the reasons you take issue with it is that it you claim that it has a disparate impact on minority students. Why set the bar lower for students from an inner city, but not a desperately poor area of rural Appalachia?

    I’m extremely curious about exactly whom you think is harmed here and why. Being commended or a semi-finalist doesn’t do much for you, but someone who would be a finalist (based on scores, resume, and transcript) if they had managed to qualify as a semi-finalist (based solely on score) could potentially lose a scholarship. But only about 15% of commended students win scholarships; based on your own estimate, this effort would help no more than 350 students per year (probably less, as marginal PSAT scores make them less likely to get a scholarship). In reality, it would be significantly less–all of those students are already in the top 3% of their class nationwide, so many of those students would get scholarships elsewhere.

    The college admissions process as a whole is pretty broken; this seems like a pretty marginal problem for a professor at Yale Law to be tackling.

    Thumb up 9 Thumb down 5
    • plusECON says:

      Addressing a small problem like this can provide insights and ideas on how to address the big problems of the world. So I think this is exactly the sort of thing we want a professor at Yale Law to be tackling and talking about. As he makes progress on this, he may inspire others working on other problems, or become inspired himself.

      Well-loved. Like or Dislike: Thumb up 17 Thumb down 2
    • NewE says:

      The process to become a semi-finalist is well documented and fairly algorithmic given the PSAT score. One may not agree with the algorithm, but one cannot fault the transparency of the process. Since most semifinalists become finalists, the process up to becoming a finalist is reasonably objective. There is some subjectivity involved in the choice of the Merit Scholar from among the finalists – but not by much since 50% of the finalists are awarded scholarships. Contrast this with the college admission process itself which is completely opaque and subjective. If some energy is to be expended, then it should be on the college admissions process.

      Thumb up 0 Thumb down 0
  3. Harry Bucher says:

    Hidden due to low comment rating. Click here to see.

    Disliked! Like or Dislike: Thumb up 0 Thumb down 17
    • Matt says:

      He means states where a larger percentage of the population belongs to a minority group (black, Hispanic, etc). Minorities tend to do worse relative to their skill on standardized tests, so if a state with more minorities has a higher cutoff, it makes it increasingly difficult for a minority to become a NMS finalist.

      Well-loved. Like or Dislike: Thumb up 12 Thumb down 0
      • Matt says:

        Oh, and I’ll guess you get 1400 scores submitted. :)

        Thumb up 0 Thumb down 3
      • Harry Bucher says:

        I don’t see why it would be more “troublesome” if black, Hispanic, etc did relatively worse. If the whole point of the blog post is that some people get a raw deal based on what state they’re from, why does it matter what the skin color is? Would it be better if the ‘losers’ in this system were ‘majority’ not minority?
        Are we worried about deserving people in general who don’t become an NMS finalist, or are we especially concerned about a socio-racial subset?

        Well-loved. Like or Dislike: Thumb up 10 Thumb down 0
      • Enter your name... says:

        Harry, it’s like this:

        Washington, D.C. is 50% black and 18% poor, If you live there, you need to score 224 to win one of these scholarships.

        West Virginia is 94% white and 17.5% poor. If you live there, then you need to score only 203.

        The effect is that kids in WV don’t have to perform nearly as well as kids in DC to get a scholarship. It “just happens” that the kids in WV and DC are almost exactly as likely to be poor. The difference is that the kids in DC are fourteen times as likely to be black.

        This is the question they’re asking: Why should an impoverished black student have to score better than an impoverished white kid, to win the same college scholarship?

        Well-loved. Like or Dislike: Thumb up 30 Thumb down 2
  4. J says:

    My guess: 6235 posts

    Thumb up 0 Thumb down 1
    • Jane D says:

      Maybe in states with higher minority populations that have already left school by the junior year the number of low performing students taking the exam goes down which drives up the score necessary to be top whatever %. Texas also has an auto admission process for top 7% to the top public universities and we know folks whose kids class rank AS A PERCENTILE drops dramatically as the size of the class drops dramatically when large numbers of kids drop out of school. Unintended consequences can be ugly.

      Thumb up 0 Thumb down 0
  5. Jim Tincher says:

    Great post! As the father of a high school junior who scored highly, I’m very interested in this topic.
    I’m optimistic – I think you’ll get 245 scores.

    Thumb up 1 Thumb down 1
  6. Voice of Reason says:

    Maybe just index the scores to the parents average AGI over the past five years? And maybe say that students with parents having a net worth of $50 million or high are automatically disqualified?

    Maybe if handicapping it too harsh, have five brackets of income where the top percentage of students of each income bracket qualify.

    Thumb up 1 Thumb down 1
  7. Jack T says:

    Whenever affirmative action is discussed, Asians are seldom mentioned as if we are non-existent, or we are the majority. I wonder how many readers here are aware of California SCA-5 which attempts to reverse Prop 209 which bans using race in UC admission.

    Media (with liberal bias) simply under-report or do not report this at all (e.g., NYTimes).

    After three Asian state senators withdrew support for the bill, the Latino/black caucus issued a statement:

    “… the major reason this measure has been delayed is due to a malicious disinformation campaign being waged by disingenuous ultra-conservative partisans intent on denying equal opportunity for all Californians.”

    which is ironic because about 80% of Asians vote Democrat.

    For me location based discrimination is as bad as race based discrimination. However when Asians are discriminated, liberals just pretend they don’t see it at all.

    Well-loved. Like or Dislike: Thumb up 14 Thumb down 1
    • John Vidale says:

      My impression from long ago when I was in school is that the cut-offs were designed to set a constant proportion of high schoolers as semi-finalists in each state. Similar sorting applies to valedictorians, the top 5% of their class, etc.. Such a quota would not be unusual, nor grounds for strong protest. Is that demonstrably not the case now?

      Also, my impression is that the scores themselves are more critical in admissions than the semi-finalist status. Certainly, the scores and their percentiles are available to admissions officers, so their derived products would be discounted by effective admission programs, unless they were specifically aiming to boast about their count of semi-finalists or use them as criteria for their scholarships.

      Finally, given that the cut-offs are available, it is not hard to test whether rich or poor states systematically benefit. If cut-offs are designed to assign a constant proportion, poor states should tend to BENEFIT from this discrimination, not suffer, as their scores are likely generally lower. So do poor states benefit or suffer?

      Thumb up 3 Thumb down 0
      • Ian Ayres says:

        John’s right that the state-specific cutoffs are similar to valedictorians. But there is important difference in characterization. When you tell me you were valedictorian of your high school, I know you had the highest gpa of THAT school. The “national” merit semi-finalists for Kansas are really just exhibiting “Kansas” merit.

        Thumb up 6 Thumb down 2
    • TedS says:

      @Jack T. I agree with most of your points. I would point out, though, that liberal support for affirmative action is not universal. A significant minority of liberals (myself included) oppose affirmative action for reasons such as the discrimination against Asians in affirmative action policies that you mention.

      Thumb up 0 Thumb down 0
    • Julie Wong says:

      Represent! Great to hear another invisible minority member trying to make the facts visible to the wider community!

      Thumb up 0 Thumb down 0
  8. Marquis Hotrum says:

    I discovered your blog site on google and check a few of your early posts. Continue to keep up the very good operate. I just additional up your RSS feed to my MSN News Reader. Seeking forward to reading more from you later on!

    Thumb up 0 Thumb down 0