I’m not familiar with Houston schools - is HISD a lower scoring school district or one of those highly competitive districts, @TestRekt ?
@suzyQ7 Lower scoring as a whole, but as with most big districts, they also have 5-7 schools that dominate NMSF in the district. Good summary here http://www.houstonisd.org/Page/38527 (download the PSAT PDF) and
http://blogs.houstonisd.org/news/2015/09/10/62-hisd-students-named-national-merit-semifinalists/
http://blogs.houstonisd.org/news/2014/09/10/56-hisd-students-named-national-merit-semifinalists/
@suzyQ7 So here’s my reasoning: In the last known accurate list of state cutoffs (2015) Texas was the 9th state on the list. Inclusive of Texas, those nine states represented almost 40% of the US population (they included CA, TX, & NY). We know that “Commended” status is simply the top 50,000 scores across the nation. There are about 1.55 million Juniors who take the PSAT every year, so 50,000 represents 3.2% of the national pool. What that means in practice is that higher scoring states will have more than 3.2% of their students in the Commended pool and lower scoring states will have fewer. Based on where Texas generally falls in the list, they should have slightly more than 3.2% of their students in the Commended pool, probably around 3.4%. We know that this year the Commended pool included everyone with a 209 or higher. The Testmasters data only had 168 students out of 8551 fall within that 209-228 range. That is slightly less than 2%. That only comes to about 60% of the expected number of students you would find within that range in a truly representative sample of Texas. My assumption is, then, that their data, even at the highest end, is skewed lower than the Texas population.
Now, if their data is still a true sample of HISD, and we put the expected number of HISD SFs at 60, then it would still be a good guess to say that the cutoff will be 219, because there are 59 students in their sample who score between a 219-228.
So is the data useful? Yes. I just don’t think you should look at it and think that it is representative of a typical Texas sample or a typical national sample.
Has anyone seen the statewide Semi Finalist release for the state of Texas from last year? I’d like to know how many Semi Finalists the state had. There were 227,078 Texas juniors who took the exam in October of 2014. I’m trying to figure out what percentage made SF.
@candjsdad Per DoyleB at comment #2951:
02-09-2016 at 10:53 am
Last year 227000 juniors took the test in Texas.
(1353 / 227000) * 10000 = 59.6
60 is right on the money.
@Mamelot But there weren’t 10,000 in their sample. Testmasters mentioned that their dataset had “roughly 10,000” students. I’m guessing they used that number because they didn’t want it to be too obvious where they got the data from. But if you actually count the graph they provided (it has a very readable scale on the side) the number is much closer to 8500. My count was 8551. Someone else on this list did an independent count that was very close to that.
I found the list of how many SFs are in each state in the most current NMSC annual report (p23). That report is based on the October 2013 PSAT. That year there were 1,334 SFs in Texas. According to the Texas state report for that year, 217,323 juniors took the PSAT in Texas. That means that .6% of students made SF.
Now here’s the interesting part. The NMSC report also gives info on Commended students. That year there were 3,517 Commended students in Texas (non-inclusive of SFs). Add to that the 1334 SFs and you get 4851. That works out to just 2.2%. Nationally that year 3.4% fell into that category.
I don’t understand why the Texas Commended figure is only 2.2% of test takers, but since it apparently is, then the Testmasters data starts looking more like a representative sample of Texas. As a representative sample, .6% = 51 students, which would put the SF cutoff at 220.
It’s a sample, but not a representative sample since it came from one region only, and one that has never been representative of the entire state.
@candjsdad wrote:
Because Texas is very much a bifurcated state. There are a lot of top kids in the bigger cities, and thus Texas has one of the higher cutoffs among the states. At the same time, Texas has a low average score, and a low College readiness percentage, because there are a lot of poor kids and immigrants bringing down the averages. Since each state is considered individually for NMSF, every state will have approximately the same percentage of those. But since the 3% commended is across the entire country, every state can vary in that percentage.
@candjsdad that’s right! I had forgotten that Testmasters might have rounded that data set up a bit. Thanks for the correction.
You guys have done a great job trying to predict that Texas number. 219-220 seems about right when you look at where TX falls in the “cut-off spectrum” (of course who knows whether those relative positions will hold). Hard to get it more exact than that, especially this year. Art seems to think that it won’t go as high as 221.
I wish someone could pin down MN a bit more.
I think once we know California, we can see where the trend is going. If CA is 219/220, then Art’s predictions are pretty solid.
One more month…
Would a 208 score possibly be good in KY for NMSF?
@riskitall28 unfortunately a 208 is just below the commended cut-off of 209 so will not enable you to advance in the competition. However, it’s a signal that you probably did great on your SAT or even ACT so hopefully you cleaned up on that. Good luck to you!
Why is everyone concluding that the data of Testmasters is from HISD students? I have gone to Testmasters and many more students that go to Testmasters from KISD, CFISD, and FBISD, all of which are strong school districts.
@scorpion110 I believe we arrived at HISD based on the clues given by Testmasters and research into past HISD numbers.
What school(s) with nmsf scholarships are you hoping for if you or your DC makes it?
@nw2this Ole Miss, Michigan State, U of Oklahoma, U of Kentucky. Only Oklahoma will drop off the list if D doesn’t make NMSF as she’s competitive for large non-NMSF tscholarships at the others.
@nw2this TAMU, OU, UTD, South Carolina. All have rather large NM scholarships 
Though, they are scholarships for Finalist - not Semifinalist.
In decreasing order of generosity: University of Minnesota, University of Chicago, Northwestern.
@scorpion110 - Testmasters says the following: “Based on an analysis of nearly 10,000 students, Testmasters has updated our prediction for the National Merit Semifinalist cutoff score in Texas to be as high as 219.”
Also: “Using new data from roughly 10,000 Testmasters students and other students who took the PSAT in Texas, we reproduced this process. We know from historic data that of these 10,000 students, approximately 60 will become National Merit Semifinalists. Applying that number to our data, we have revised our estimate of the National Merit Semifinalist cutoff for Texas to be as high as 219.”
In the past HISD has had around 60 NMSF’s. @candjsdad, @dallaspiano and others who magnified the chart and did some careful counting-up have posted that there are approximately 8,500 data points - a bit less than 10,000 but very large nonetheless. They concluded it’s quite likely that the data came from HISD. Testmasters has explicitly said in its comments that it will not divulge the source so this continues to be a “best guess”.