Disclaimer:
I am still a junior. My numbers were crunched two weeks ago, they were based on PSAT data tables (2012-2014). The numbers were expressed in a subjective, a speculative way, a biased angle or many assumptions and etc… Make it short, there are many flaws, missing points… do not take my numbers seriously or personally. I mean no harm at all. Just like the guessing games of NMF cut off based on calculated and logical reasoning.
The SI predictions being made here makes absolutely no sense when you look at the Cobb county data for Walton High School. On average, the top 100 students there scored 1453 TS, which when compared to the low SI predictions being made here (~ 209) would qualify all 100 students to make it to SF. Georgia has ~450 SF slots available, what makes you think all top 100 students from Walton will make it through? Can it happen? Sure, but it’s hard to believe a single school has the potential to contribute 22% of the state’s SF nominees. We won’t know until we see state wide data from CB. The concordance is only good for making commended SI cutoff.
@Mamelot – couldn’t agree more. That’s what makes the concordance tables so problematic. I’m looking at a SI209 that going through all the p3,4 prelim table permutations ranges from a 201-206. That’s quite a large spread and not very helpful. But it does tell me some things: (1) the reported 99% user and SI% my dd has are wrong; (2) she has a proverbial snowball’s chance making SF as we are in a mid/high cutoff state; (3) working through the permutations she has 70% shot at commended…
Another factor to consider is probably more kids prepared for SAT early this year becuase lot of paid counselors adviced kids to take old SAT. This may have boosted PSAT scores also.
Unless of course, following statement in the Cobb County pdf “Table 3. Mean PSAT scores for the top 100 scores for 11th graders at each school.” should really read “Table 3. Mean PSAT scores for the top 100 scores for 11th graders among all Cobb High schools.”!
@Mamelot: “But if everyone did that for his/her state and then submitted the answers to put in a table here, you’d probably get a pretty good set of predictions.”
I’ve done it for 214.
Range = 209-220
Combinations: 209 209 210 210 210 210 211 211 211 211 211 211 211 212 212 212 213 213 213 213 213 213 213 213 214 214 214 215 215 216 216 216 216 218 218 220
Median concorded score = 213
Mean concorded score = 212.7
Mode concorded score = 213
(These statistics would more useful if all combinations were equally likely.)
By the way, I wonder if Testmaster’s has a typo for Connecticut. It has Connecticut dropping from 220 to 215, while it has Georgia going from 218 to 216. Connecticut has had a higher cut for Georgia every year for at least the last 7 years.
Here is a little math exercise using New Jersey. They have about 540 SF slots allocated which is about 3.33% of the total available slots.
From previous postings, the number of perfect PSAT scores in a given year is around 100. Distributing those scores across the country, would give 3.33 perfect scores to NJ. But hey, Jersey kids are smart, so lets increase that to 5%. So 5 perfect scores in NJ.
To achieve a 225 score out of 228, students need a perfect score or can only miss 1-2 questions Total on the test. 2 missed questions in writing is actually a 224 and misses 225. But lets just say 1-2 questions over the 3 section areas. Lets call these the “nearly perfect”.
So 100 Perfect scores in a year. How many nearly perfect? Lets call that 400 (just my number for demonstration purposes). So a total of 500 students with Perfect or Nearly Perfect scores. NJ gets 5% of them. That is only 25 students. But they are allocated 540 slots.
How many Perfect and Nearly Perfect scores must there be nationwide for NJ to have 540? Approximately 10,800.
10800 * 0.05 = 540. So we need 10800 Perfect/Nearly Perfect scores this year in order to have a 225 SI score in NJ. A typical year maybe around 500 (again, my estimation for demonstration purposes). That means this year, there would need to be around 20 TIMES as many as previous years. That doesn’t pass the “smell test” for me.
Conclusion to me is that the high state cutoffs are coming down. 225 just isn’t gonna happen.
I think the real issue is the commended score. The SI% table indicates around 200-202. Testmasters is using 210, apparently based on the concordance tables. If 210 is correct, then some lower cutoff states will have cutoffs increasing. If 200-202 is correct, then nearly all states will see cutoffs decreasing.
Another musing on PSAT percentiles. DD took December SAT as she wanted 1 shot at the old test. She did ok , high 2100s, which I think is around 97/98%. Her PSAT was 1390 (last TS of 99%). A 1390 is 2080 on SAT 2400 scale which I think is around 95/96%. This is another reason why I think the percentiles are whacky on the PSAT 2015. But maybe you can’t compare PSAT to SAT…
Dumb question here. But might not CB have decided (either before or after the wacky SI chart) that the best way to save face for an overly easy exam, attract business, and not piss off students/parents would be to allow a lower commended score and lower qualifying scores and then leave the national merit folks dealing with the mess of how to reduce a larger number of candidates into national merit finalists?
@Speedy2019 - even testmasters is predicting an SI of 220 for NJ and not 225 as you report. Conceivably though what can happen is that the commended level cutoff may exceed the lowest SF cutoff for poor performing states (states with low participation rates). That is, if NMSC wants to preserve top 3% (test takers) or ~1.5% nationally to come up with the magic 50K total students. The way I see it, there is likely inflation of the 50K commended students if poor performing states are too far off the suggested SI cutoff.
The problem is @DoyleB, per the new definition of percentile (Definition B), you can’t have 100%ile at 1520 because that would be like saying you did better than yourself!
@OHToCollege wrote: “which when compared to the low SI predictions being made here (~ 209) would qualify all 100 students to make it to SF. Georgia has ~450 SF slots available, what makes you think all top 100 students from Walton will make it through?”
I couldn’t agree with you more. If the GA cutoff is 209, we’ll have 2000 NMSF. Which will pose a real problem with us only having 450 slots.
@OHToCollege, I used 225 in my NJ exercise for 3 reasons.
1). Another CCer predicted 225 at the high end yesterday. His prediction has as much weight as mine. However, I wanted to do some math that shows how difficult that would be.
Others were saying 2015 cutoffs would be similar to 2014. Another potential prediction worth as much as mine, but again, wanted to show how difficult that would be on the high end.
To show that the high cutoff states are decreasing. For me, the more interesting cases are the lower and mid states.
I keep harping on this, but per the suggested approach by @suzyQ7, the issue here is going to be how NMSC plans to handle the students at the lowest SI cutoff score for each state - how much borrowing are they going to do? Or will they make the SF to F criterion slightly more stringent to then award 7800 students with NMS titles.
@OHToCollege Actually the new definition is what allows 100%. You did as well as or better than 100% of the kids. All new tables should have a 100.0% at the top, given definition B. Definition A says “better than”, which is what prevents a 100% in the old tables.