I’m hoping 221 will make it in DC/NJ! Do we have a California homeschooler that has a 219/220/221?
I’m thinking that the cutoffs will compress. Previously, California was 2 points (1 point in some years) lower than DC/NJ. With the compression in these new scores, I think the difference will not be more than a point. They may actually end up with the same cutoff. So, if California is a 219 or 220, then DC/NJ could be 219/220 or at the most 221. That is what I’m hoping!
I’m hoping 221 will make it in CA . . . . Anyone have any sense whether this will be high enough?!
@PicoLA one set of estimates is over here http://www.■■■■■■■■■■■■■■■/national-merit-semifinalist-cutoffs/
@PicoLA there’s no guarantee until we hear results in September, but Compass Prep predicts a 220 for California and the possible predicted range is 219-222. You’re right in there - rooting for you!!
My best guess, based on everything that we’ve seen since last October is that DC/NJ will be 222 or 223. I don’t think there is enough room above that for it to go higher, but it is possible that it could hit 224. There will be far fewer scores at 223 than at 222 or 224, so I don’t know what exactly to make of that. I will be quite surprised if the DC/NJ cutoff is 220. I think Compass Prep’s estimate of 221-224 is exactly right.
@candjsdad Did you see testmasters chart with 10,000 actual results from Texas? There is a huge drop in the 223 numbers. My guess is 222, but I’m hopeful for a 221.
http://collegeadmissions.testmasters.com/wp-content/uploads/2016/02/Picture3.png
@suzyQ7…No…no…no. We don’t need Texas to be over 221!! Art at Compass shows range up to 220 for Texas and Testmasters shows 219.
@Tgirlfriend, @suzyQ7 was just describing what the score distribution looks like, not what Texas will end up with for a cut-off. They are actually talking about DC/NJ. Rest easy #:-S
@Mamelot …so they aren’t talking about Texas being at SI 222 then? I guess I read and then went into panic mode. I am telling you September needs to get here already! I say nothing to my S however I am worried sick for him. He has had so many let downs his Junior year I really Senior year is better. Getting NMSF would be a feather in his cap. Panic mode OVER!!
@Tgirlfriend I hear ya. Hopefully, too, you’ll feel better knowing that Texas over 221 just doesn’t make any sense whatsoever. As weird as things have been this year, they aren’t quite in the realm of Crazy. At least not yet 8-}
@Tgirlfriend I was talking about the top score NJ/DC - using the Texas data as predictors for that. I think Texas will be 219, CA 220, and NJ/DC 221 or 222. Hoping for NJ/DC at 221. It would not shock me if CA was also 221.
I really believe Testmasters has it nailed for Texas. They have actual results, for goodness sakes. I wish we had someone from NJ with actual results.
@suzyQ7 I did see the data. In fact, I was the one who counted all of the bars on the graph and converted it into a good estimated numerical distribution. I had one significant concern about those numbers. We know that Commended is 209 and that about 3% of students nationally should be commended. Based on their historical NMSF cutoff, I think it is a fair estimate to say that Texas should have an average number of students in that Commended range – so 3% of students should be between 209-228. However, when I ran the numbers from that specific data set there were too few students in that range. Only 1.9% of the students in that sample scored a 209 or above. That suggests to me that at the highest score range, that sample set skews low.
@candjsdad It’s not all that surprising that the data skews low, and it shouldn’t negate its value. What’s far more important is consistency. If we assume that the HISD pool didn’t change much year-over-year and assume that the new SAT did not significantly advantage or disadvantage those students, then we’d expect a similar number of Semifinalists. The good news is that 78.6% of HISD students took the PSAT/NMSQT in both 2014 and 2015. The mix of students also didn’t change much. As you point out, Houston students have mean scores lower than the national scores (445 EBRW and 455 M versus 507 and 502 nationally). Again, that difference doesn’t matter much as long as the number of NMSF students is relatively consistent from year-to-year at HISD. It’s not going to be perfectly accurate – the difference of a few more or few less students reaching NMSF status does mean that we can’t assume the Testmasters data predicts things to an exact point value. It should be close if they have used a properly similar pool (i.e. they didn’t throw in a lot of their own students that were prepped in one year but not prepped in another.).
@candjsdad But couldn’t that be because the test was ‘dumbed down’ so that more kids in the middle scored higher, but the very highest level there was not as much ‘dumbing down’? I am not a statistician (far from it), but I think that is what the analysis on this thread and others info has provided… that the bell curve has shifted to the right and lower cutoffs will go up (sometimes substantially) but the highest cutoffs will come down. So rather than thinking that the Texas actual sample we have skews low at the highest ranges, its that at the highest levels the scores top off close to where they have historically topped off. That’s what I’m hoping, anyway.
@TestRekt Testmasters never (to my knowledge) stated where they got this data, and in fact seemed to suggest that it came from multiple sources. Of course everyone assumes that most of the data came from HISD, but if some of it didn’t, that makes interpretation harder. If it is just HISD, then I don’t think I ever saw a number for the average number of NMSFs from HISD. What I was saying is that, especially at the higher end, the curve shouldn’t be taken as representative of Texas as a whole. It is skewed lower than what Texas (or the national) curve would be.
@suzyQ7 That doesn’t quite work since we have a true point of fact now about this year’s test. Nationally, approximately 3% of students scored at or above a 209. Since we know that we can compare the Texas data to the national data and state that the dataset is skewed lower than the national data. To whatever extent the curve changed from previous years, that won’t affect the comparison of the Texas data to this year’s Commended score.
@candjsdad That has been my concern as well regarding that data. It may not be representative sample of the entire state, which is not surprising. I just hope it all balances out.
Okay…everyone is making me nervous again. I counted the bars also on the testmasters graph. Not that I could make heads or tails out of the numbers. Math is not really my thing. @candjsdad …so you think testmasters scale is not representative of most kids in Texas? If anything I would think that testmasters used scores from the kids that they are helping take the tests and then some from the public and private schools in the area. @suzyQ7 …I sure hope you are correct with your numbers. Our Principal text me the other day asking if I had heard anything about NMSF. I think he is more excited about my S being a NMSF than we are. He has known my S since birth and asked that if he could hand him is diploma. Next year he will retire and having a NMF in his school after 20+ years would be a feather in is hat. Our fingers…toes and everything else are crossed until then.
Texas dudes, here is the link to the earlier conversation about the Testmasters Texas data (beginning with comment #2941). There is some helpful info. there that can hopefully answer some of these questions:
http://talk.collegeconfidential.com/discussion/comment/19269953/#Comment_19269953
This is some thread.
Why do you think its at the highest score range that the sample set skews low? @candjsdad
This is just pure July, geeky, speculation fun - since the thread has been so quiet, LOL
Testmasters has been a bit coy about the exact source, but the evidence seems fairly strong. HISD has had 63, 56, and 60 NMSF’s in the last 3 years, I believe. They had 9,194 11th graders take the PSAT in 2015. Ultimately it’s not that useful – whatever the source – unless the approximation of 60 students from the pool is correct. Agreed that the students are not representative of Texas as a whole.
