I am quickly rereading my post and notice all the grammatical mistakes like substituting their for there. My goodness. I really do need to sweep those cobwebs away. Anyway, please overlook.
Can add DD’s SI for the list: 1460/1520 220 SI (NV)
This is for @CA1543 and those who felt a data point from a smaller school without too many NMSF qualifiers historically would be helpful. I don’t have a lot of specifics yet – the GC who handles PSAT is out today and it may be a couple of days before I hear back from her since we have a snow day or two predicted.
In Iowa, the SI of 216 and TS of 1450 was “one of the highest” and a “very good score” this year. The GC I spoke to is in her third year at this school, and she remembers 1 qualifier last year and 0 the year before. She believes they had a couple NMF before she joined the staff at the school but no idea when.
I will provide more info as and if it becomes available.
Re post #2262. I should clarify that the GC comments about my son’s score were regarding his performance compared to classmates, not statewide. Sorry.
Given what else has gone on this year, I would be shocked if the release date of the state reports this year wasn’t the latest date in recorded history… 
So far I’ve heard 2/28 - so I’ll guess a release date of 3/15…
well take a look at when CB released the scores, a month after they said they would…so I wouldn’t be holding my breath for early Feb release…more like late Feb or Mar.
@IA Books – re post 2263 - Thank you - that does indicate a good score indeed (prior Iowa cut off was 208 & now predicted by Test Master to go to 212) and we look forward to any other info you learn,. Imagine the Iowa Caucus activity and barrage of presidential campaign efforts is consuming much attention out there all eye on Iowa today! Wonder if the GC will share if many are at or around 212. Thanks again for sharing any further info - the Mean score in Iowa & for the school are helpful too.
@PAMom21, thanks for the dates! Could be this week for the reports. But why do I have this feeling, they will be delayed???
@Plotinus thanks for your post - 2252. interesting points you shared. I have to say that the more i learn the more frustrated I am with the CB & dismayed about how much the PSAT & SAT “matter” in terms of students’ futures. I totally get what you are saying about the research study (and I wonder which students did better on the Frederick Douglass questions!), the lack of fair/well written challenging questions and how this whole thing should have less impact not more on college acceptances, scholarships, access to high quality summer programs etc. But I wonder if CB’s efforts to take market share from the ACT, to infiltrate state testing and get more students to take APs which inures to their financial benefit, will end up backfiring more than even some of its recent missteps. I hope a reputable investigative nonprofit takes on a study of all this & can perhaps enlighten the public, more school leaders, and ultimately adjust the thinking of colleges and the scholarship providers. We’ll see - hope it goes - certainly need more data is needed this year to really make meaningful predictions about NMSF cut offs.
Please visit the Thread & share your thoughts as time permits - you are an important voice here!
@Pickmen Thank you for straightening me out about the Compass Prep article probably not being correct about the magnitude of the effect of changing the percentile definition. I will have to go back and study some of the previous postings about this.
@CA1543, another thought (going to the other extreme) is that the higher mean vs research study group doesn’t affect the higher percentiles at all.
Remember Jed Routh’s article: https://www.■■■■■■■■■■■■■■/blog/category/jed-said/
If you look at his 2014 PSAT CURVE and 2015 PSAT CURVE charts,the peak is shifted to the right, but virtually no change at the high percentiles. If fact, he shows no “local peak”.
I think the charts were for National %, but SI could be similar.
https://www.■■■■■■■■■■■■■■/blog/wp-content/uploads/2016/01/Screen-Shot-2016-01-14-at-9.13.16-AM.png
@BunnyBlue #2269 no problem. I think the definition change is significant at the middle of the curve. What I posted earlier was probably not totally accurate wrt to the number in the 99 that could be affected. If there is a massive cluster at 1 decile in the 99% band, the definition could affect more testers than I suggested but we are still only talking 1 decile in the 99%, so it doesn’t really resolve the p11 2015 guide problem. @dboyle was hypothesizing that there is probably an aberrant cluster in the 99% or 99%+ band that shouldn’t normally be there under a normal distribution. I think he’s right. But that I see as causing a massive headache for NM Corp when trying to sort out SFs vs. commended. The existence of this cluster could cause more testers to qualify for SF than commended unless NM Corp finds a way to break the cluster up. Creating SI fractions? But maybe I’m off base here.
@CA1543 2268 Went to Google something and saw their icon of the day is the man himself, Frederick Douglass. I don’t think a lot testers will have good associations of him from this PSAT. My DD I think did OK on those passages b/c she had studied him in APUSH the year before. But I reviewed those questions and they were HARD. The language is baroque, Victorian era, and full of rhetorical flourishes. Contrary to some posters here, I am not sure this test was so easy.
I believe there is a problem with the Testmaster cut-off projections found at
http://collegeadmissions.testmasters.com/update-psat-scores-cut-national-merit-2016/
Looking at the historical cut-off scores below (from Mamelot’s post #1551), how do can they project that Connecticut will suddenly have the same cut-off as states that have consistently been 5 or more points below them?
State …2010…2011…2012…2013…2014…Testmaster Est.
AZ…213…212…214…213…215…215
CO…213…212…215…213…215…215
CT…220…218…221…220…220…215
FL…214…211…214…211…214…215
OH…;…214…212…215…213…215…215
So did the performance of the top 1% in Connecticut drop drastically this year or did the top 1% of the Florida scores increase to an all-time high? I do not see how FL and CT will have the same cut-off this year, no matter how much compression there is at the top.
It appears the Testmasters used the Concordance tables for their cut-off predictions. I think this abnormality shows that the Concordance tables do not accurately represent actual results, that or Testmasters just made an error.
@CA1543 College Board set its PSAT college-ready “benchmarks” at a 75% probability of obtaining at least a C in a college course in the same area. So that “college-ready” basically means: “won’t flunk out of an average college.”
I agree that it is important to have a test whose main purpose is to let schools know which students are in need of remediation. But that does not mean we should use the same test to decide who should go to Harvard.
You should not lose your spot at Harvard because you made a mistake figuring 300/.06 without a calculator or confused “its” with “it’s”. There were things on the old test that were dumb, but these things are dumber.
In fact, I was just talking to a junior who is a student of mine. She took the old SAT in January, and probably scored around 2000 - not too bad for a junior, but not as high as she would like. Here is what she said about the experience of taking the PSAT in October: “While I was reading the questions, I was thinking, why are they asking me this? This is so stupid. The old test makes much more sense.”
I believe the validity study I mentioned above was not the same study as the concordance study or the percentile study. In the validity study, a correlation was established between scores of college freshman on the new SAT and their first year grades in college. I suppose more girls were studied than boys because in the past, the correlation for girls was worse. That is, the SAT scores of girls predict lower first year college grades than girls in fact obtain. Maybe it was important to CB to show that girls’ scores on the new SAT correlate better with their college grades, or that it is a “girl-friendlier” test. Just a guess.
@Speedy2019 – I think that could totally turn out to be right. His midpoint on his chart is about 145 - the actual SI mean - as corrected by the CB is 148. Jed’s upper tail - puts the 99%ile at about 205, right? Not sure what all data he had access to but yes, the shift in the curve did not convince him there was a real impact on the upper part of the tail, which on his chart’s is a shorter tail for 2015 than 2014. Thanks!
@pickman I agree the test was not easy or even easier for everyone, but I think it would be a lot easier for certain people. This is why I think there will be more very high scores.
Most students would have trouble with the Frederick Douglass passage because they are not familiar with this kind of language. But most students had trouble with at least one passage on the old test too. I don’t think students who read widely would have had many problems with the Frederick Douglass speech. The questions, even when “hard”, seemed to me easier than the “hard” CR questions. And there were no sentence completions.
Of course, these are just my subjective impressions. I don’t have any data to back them up. The data are coming, the data are coming…! 
@Plotinus – re the gender question in connection with the study etc - I had came across and shared this research:
http://files.eric.ed.gov/fulltext/EJ1011761.pdf – yes - would not be surprised if there was an effort to make it more girl friendly.
And there is recent research about the SAT not accurately predicting whether certain groups will succeed in college:
https://www.insidehighered.com/news/2016/01/26/new-research-suggests-sat-under-or-overpredicts-first-year-grades-hundreds-thousands
And it has been pointed out by Compass Prep that the benchmarks were lowered.: http://www.■■■■■■■■■■■■■■■/problems-with-new-psat-part-3-benchmark/
“An analysis of the new math and ELA college and career readiness benchmarks shows that while SAT Math standards were left largely untouched, the ELA benchmarks were lowered significantly. Below we explore five points that are at turns surprising, interesting, and troubling.
College and career readiness ELA benchmarks on the PSAT and SAT have been lowered.
The definition of “meets expectations” has changed.
The “meets expectations” rate for the new ELA benchmarks is far above those of comparable exams.
“Score pollution” at the low end of the EBRW scale impacts the validity of the 10th grade benchmark.
No research has been published explaining how and why the ELA benchmarks were lowered or if any watering down occurred.”
It is hard to have confidence (I really don’t) that the PSAT or the SAT can really measure all they purport too for as many populations as they claim to. Think rigor of one’s high school curriculum, grades and other school data and perhaps some APs & SAT subject tests are relevant more than the SAT. And now the SAT essay is being ignored by many colleges too.
My son focused on the ACT - did well & now has to manage 2 more SAT subject tests - frustrating that AP Physics1 1 does not align the Physics subject test, but that’s another story.
@CaucAsianDad, I like your view. Please let me use your table. Thank in advance
I will use 2013 PSAT data table compare to 2015 PSAT’s - based on assumptions, they behave accordingly due to
- 2013, 16 band slots in 99+% and 10 band slots for 99% ranges- 240 max
- 2015, 15 band slots in 99+% and 9 band slots for 99% ranges- 228 max
- Conservative Estimate (CE): data behaves normally by comparison and mapping
- Aggressive Estimate (AgE): if the test is easier or students get smarter (I doubt). So I shift up 2 bands
- Will use the @CaucAsianDad ‘s table
State …2010…2011…2012…2013…2014…Testmaster…CE…AgE
AZ…213…212…214…213…215…215…205…207
CO…213…212…215…213…215…215…205…207
CT…220…218…221…220…220…215…209…211
FL…214…211…214…211…214…215…204…204
OH…;…214…212…215…213…215…215…205…207
TX…220…217…209…211
CA…223…219…212…213
PS: true believers as @LadyMeowMeow categorized
@Plotinus – Yes, I think that’s what might be going on. For some in the top percentiles it was probably easy, perhaps more than usual, which is why we may have too may high scoring reports coming in. I agree too for kids who had some wide range reading in different eras (Shakespeare, EA Poe, Hemingway, Dickens etc.) the language in Douglass may not have been intimidating. Also, if the tester was not well read but very smart he/she might have been able to still knock out 2 of the four obvious wrong choices and then find the subtle difference between the remaining two close choices and pick the right one after grueling through the passages.
@SLparent * @dallaspiano – for the data chart
cgsarat71 - reported SI 218 / TC 1470 - Texas