National Merit Cutoff Predictions Class of 2017

@hcmom65 thanks for the super informative link!!

Based on that link, assuming the SI Percentile data in it is solid, then I’d estimate that 199 will be the cut off for Commended. If that’s the case, then the very lowest state’s cutoff for NMSF should be at or around 199. More competitive states will go up, up, up, to ??? My personal state is the lowest cut off state (WV) in recent years, so I’m not sweating the details for the higher cut off states. I’ll let someone else do that math, lol. If I had all the accurate data, I’d do it in a heartbeat, but obviously we’re dealing with a lot of guesswork on whether the released data/information is accurately reflecting the actual test this year . . .

My Commended cut off estimate is based on SI %ile cut off for Commended (%C) = #Commended (50,000) / # eligible students (1,500,000) = 3.333 %. So, top 3.33% or so of SI scores should be commended.

That means 97, 98, & 99% SI Scores should be Commended, plus possibly a small bit of the 96’ers, representing 96.66% and above. The chart on p. 10 of the above link shows 200 & 201 as the 97% scores. 199 and 198 are the 96% scores. So, depending on how big those chunks of students are at 199 and 198, I’d guess that they’d set the cut off at either 199 or maybe 200 to get their approximately 50,000 students for Commended. My bet is on 199, but 200 is likely as well.

So, I’d guess that the bottom state or two will have a cut off of 199 (or 200) for NMSF (i.e., bottom states with no Commended Scholars because they all make NMSF). It will scale up from there. Someone can play with the %iles to figure out where the other states end up.

Honestly, my best guess is that cut offs this year will be in the same general range as recent years, obviously the top possible cut off is 228, but I wouldn’t be completely shocked to see one or more states require a perfect 228 to qualify, although that seems crazy.

** This is all ASSUMING THE %iles in the SI ranking on p10 are right!! **

Thanks for the information!! @mmom99 I’m not sure why they would publish this if it might change significantly. I’ve said before percentages are percentages. So if 214 is 99+ that has got to be 99.5%. I keep using TX because that is where we are from. To my knowledge, we have NEVER been in the 99+ range. So, I’m SUPER excited because my son has a 215.

@Mamelot I stopped at 221. 220 is shown just 2 lines but has much more combinations. I got tired and CC doesn’t allow posting more lines in one post :wink:

So if we read this link:
https://collegereadiness.collegeboard.org/pdf/2015-psat-nmsqt-understanding-scores.pdf

then does it look like a 217 being that it is 99+ might work for California? Any one have thoughts…Thank you1

I’m glad 99% S.I came out to be ** 205 ** in the range I posted before otherwise we would have so much conflicting info!

http://talk.collegeconfidential.com/discussion/comment/19173991/#Comment_19173991

I can’t tell if this 205 is 99% “national” or “user” percentiles, LoL.

Looking good to 215+ scorers,

Here’s a thought. On page 6, it says " For the total scores and the section scores, percentile ranks are provided based on two different reference populations." It does NOT say this on page 10 regarding the SIs.

Does this new PSAT document from the CB clarify the total number of eligible juniors who took the test this past October? If so, is this total really bigger than in previous years? If those getting comnended or better is a fixed number, and the total eligibles is greater, then the percentile cutoffs have to be more stringent. Thoughts?

@mmom99

“My bet is on 199, but 200 is likely as well.”

“Honestly, my best guess is that cut offs this year will be in the same general range as recent years, obviously the top possible cut off is 228, but I wouldn’t be completely shocked to see one or more states require a perfect 228 to qualify, although that seems crazy.”

So you think commended could be 200 but top state cutoffs 228? Not sure if that makes sense. Why do you think such a high cutoff for the top states - is there something in the data that makes you think that?

@suzyQ7 NO way in he*l the SI cutoff for top state (DC or NJ) will be 228 this year. I say this because in my high school which produces 30+ NMSF per year “only” had several 228s. Unlike other methods, I am basing estimate cutoff SI score for CA on top 30 scores in my high school which produces around 30+ NMSFs per year. I would guess top 30 scores cutoff is 219, 220 or 221. Last year, the cutoff for CA was 223, so for top states, I would say the cutoff SI score will be SEVERAL points lower than the last year’s cutoff SI score.

@suzyQ7 I don’t see that any of the states are going to require a cut off of 228.

The 99+ percentile is only about 7500 kids (.005 * 1.5 mil) - throughout the entire US (and DC and outside US). Even assuming an even distribution from 214 (bottom of the 99+) to 228 (the top) that would only leave 500 or so kids at 228 - and I’d argue that is an overestimation. The likelier scenario is that those kids are bunched between 214 and 223 or something and only a few are around 228. Looking at the NMSC annual report from 2013, the # of NM’s in the highest scoring states were CA (2027), NJ (540), MA (328), DC (57), MD (318) and VA (390). With the exception of DC, each of those had over 300 kids qualifying for National Merit. The number of qualifiers isn’t going to be decreasing between 2013 and now. So you would need a cut-off low enough to allow the appropriate number of students to qualify. 228 is just too high - there are likely too few students who hit that score - let alone bunched within one state.

In the collegeboards new publication of understanding score 2015, The SI score of 214 to 228 , all are marked 99+ percentile. Now that is 15 “slots” altogether, and SI score of 221 is right in middle.Considering that the qualification Semifinalist are to be in top 1/2 %. Now considering that no of students in SI 222 to SI 228 is LIKELY to be less than those in SI 214 to SI 220 slots . So students with 221 and above are in top 1/2% in nation. Now CA being high SI cut off, but it also have more students as it share of scholarships, due to top 1/2% a larger state’s larger student population. From there can we infer that high performing AND large state like CA cutoff is likely to be somewhat below 221. Thinking like that points to what testmaster predicts 219 for CA or what @websensation says 219-221.

It is not accurate that all NMSF are in the top 1/2%.

1/2% or higher is accurate for CA.

@replyback CA was not the highest cutoff (last year). CA and MA were 223 and DC, NJ, and “other selection groups” like Internationals and boarding schools were 225 (!) There were no 224s for some reason.

SO, what do you guys think will be the HIGHEST cutoff based on the new data?

I think it’s stated that 16,000 get NMSF, which would be slightly more than 1% of 1.5M. But it will obviously vary by state. So theoretically in an “average” state, 99th percentile would match up with the NMSF cutoff.

In 2014, 99+ was 224. And the 2014 CA cutoff was 223.

So if you just use these percentiles, given that 214 is now 99+, that would predict a cutoff of CA of say 214 (or maybe 213).

Note that the concordance tables (which is what that publication recommends using) projects a 2014 223 maps to a 2015 1470. If you guess that the average 1470 had 750M / 720V, then that yields a cutoff of 219. (Which is where I think many of these sites are guessing 219 for CA.)

Thinking further on similar line, the cutoff last year was 223, now without even considering other things, it appears students “improve” or for whatever reason, the cutoff goes UP over the year? In that case can CA cutoff this year expected be bit more than 219 at least, not “same” as last year. We are in CA , keeping finger crossed!

@suzyQ7, no I don’t think top states will have a 228 cut off. I think that’s crazy. :wink:

However, I do imagine it could be possible. The way I’d imagine it being possible is IF the college board created this test without really focusing on differentiating among tip top scorers. If they eliminated some of the hardest material and/or changed the curve so that you’d have a big clump of top scorers at 228 or close to 228. I.e., if there are a big heap of perfect scores. I don’t know if that’s the case, and I don’t think we’ll know for some months unless more data is put out. If that is the case (that there are huge numbers of perfect scores), then it is possible that a top state cut off could be a perfect score, or maybe a couple points short of perfect, etc.

Having top score states with near-perfect-score cut offs could be an unintended consequence of “making the test easier” by eliminating the hardest questions and/or reducing time pressure issues. (Again, I’m not sure this is the case. Some folks have made a case for this, but I don’t really believe it myself.) For instance, if the college board wanted to make this test more attractive to educators (and students) with more typical in-the-middle abilities/scores.

Historically, it seemed to me that PSAT was really just mostly interesting to NM contenders. In our region, many/most students weren’t being encouraged to take PSAT unless they were both a top student (NM contender) and their advisors/educators were on their game and aware of NM possibilities. Up until very recently, most students in our region (top high schools in the state, but that isn’t saying a lot), weren’t taking the PSAT because, I believe, of these issues. (My kids took it because I expected that they’d be NM contenders, and we used SAT testing as prep for PSAT . . . I wouldn’t have considered using the PSAT as practice for the SAT . . . Why not just take the SAT itself??) Anyway, there has been a shift in recent years towards having all the kids take the PSAT, in multiple grades, etc. Before a year or two ago, the ONLY people I knew who signed up their pre-11th grade kids for PSAT were parents like me who expected and hoped for high scores. We used 9th, 10th grade PSAT as a chance to hone skills for the all-critical JR year PSAT. Now, at our public schools, vast numbers of early grade kids are taking the PSAT and all the JR year kids took it, on a school day, excused from class. Similarly, SAT prep has become more widely available and is free or highly subsidized in the schools. (I’m not 100% sure on these public school details, as we homeschool, but we use the schools for testing, so I do know I am right about the fact that all the JR kids took it during a school day, as my kid took it with them.) Obviously, a company in the business of selling tests and test prep is going to be more successful selling their products if the products are useful to a wide range of people as opposed to only being interesting/attractive to top students.

Personally, I haven’t tried to work out the numbers for the top scoring states, and I have an easier task figuring these guesses out with only really needing to guess at the Commended cut off, as that is essentially where our home state lies for NMSF cut off. (0 Commended Scholars last year in WV, an artifact of being the very worst scoring state last year. Makes me sad for our state. I hope for the day when my youngest, last, next kid takes the PSAT in 4 years for reals, and I hope she has more competition that year than my first two kids have had . . . All that said, I love our state, and the NM kids I know who’ve headed off into the big world have all been exceptionally successful at their universities, including Brown, Princeton, Olin, etc . . . Kids who rise to the top in a state like ours with limited resources can and do perform perfectly excellently alongside those other kids who rose to the top of their own more competitive states. . . )

If I were going to try to figure out the other states’ cut offs, I’d put them in order of SI index cut off from prior years, then I’d try to correlate those cut offs with national %iles in those years, then I’d try to align that with the NM data on SI %iles for the current year. Since there are a huge number of SI scores that are in the top 1% (and even the 99+% whatever that means), it may be very hard to figure out where any cut off would line up within those brackets. I.e., if NY typically has a SI cut off of 99.5%, then I’d want to figure out where the current year’s 99.5% cut off is, which is easy to do if we had the raw data, but impossible with just the big range of 99 and 99+ SI scores. Someone could give it a good guess, I am sure, but it’s not gonna’ be me, lol.

All that said, my best guess at the moment is that the cut offs might be close to prior years. The NMSC and College Board obviously collaborated on the PSAT scoring system, and it would be appealing in some ways to tweak the curving/scoring to somewhat approximate prior year cut offs, so maybe that’s what they did. Who the heck knows?

http://talk.collegeconfidential.com/discussion/comment/19170635/#Comment_19170635
In looking at this chart (it might not be the most recent), we can see that math & reading had fairly generous curves – should be based on the difficulty of the test & actual test takers, right? If the Oct. 2015 PSAT were too easy for many and there were LOTS of perfect/near perfect scores, the curve should have been less generous, right? And taking into account the new CB report – https://collegereadiness.collegeboard.org/pdf/2015-psat-nmsqt-understanding-scores.pdf
I think that our hotshots stats folks here, Testmasters & others are coming up with some good estimates for cut-offs - I have not seen a SI concordance table measured against 2014 : https://kleinhs.kleinisd.net/users/0012/docs/14-15_NewsInfo/PSAT_UnderstandingScores2014.pdf) is that posted on a thread?) but each of us can pretty much compare our state’s cutoff and percentiles from the past and get a sense of what is lately to happen this year or at least a potential range. Hopefully the percentiles in the 2015 report are “actual” and not based on a small sample or last year’s scores. :slight_smile:

The new CB document claims that “about” 1.5 million eligible kids took the exam, which is odd, because they said that so many more kids took it this year and last year’s number was 1.5 million – were all the new kids sophomores? Seems unlikely. I think we may need to be a little conservative with the percentiles in case the estimate of eligible kids is low.

But with respect to the high cutoffs – you’ve got 205 - 213 as 99th, which ought to put 214 as the very bottom of 99.5%. The problem is that the very high cutoff categories probably need to be in the top 99.5% of their state/category, rather than the national percentile and we don’t really know what that is (other than the fact that the reason the high cutoff states fall into that category is because they have so many high testing kids). The Testmasters 220 cutoff estimate for D.C. looks reasonable, but the percentile spread makes me think that kids with lower scores should not give up hope.

@mmom99 I was never very good at math, but I don’t understand how you guys are looking at the new CB publication, figuring that the number of test takers is about the same or only slightly more than last year, and then coming to the conclusion that the cut-off scores are going to be about the same as last year. That just doesn’t make sense to me. But again, I’m not that great at math.