“Everyone is entitled to his own opinion, but not his own facts.”
That’s just it – we don’t have many (any) facts. On the one hand we have CB documents. On the other hand we have anecdotes, reports, rumors, paraphrases, retellings, and summarized conversations from GCs (who may, um, not be as invested or attentive as we are). On either side, a lot can & maybe did go wrong in the reporting of the data.
It takes an act of faith to consider either CB or the anecdotes to be pure “fact,” so nobody who does so can claim to be entirely rational. Which is good, in my opinion, since it keeps things going nicely.
@azcpamom your anecdotal data supports a cut-off of 215 or higher. Four NMSF’s this year isn’t out of the realm of possibilities given the historical data and small size.
And small size, of course, but again check out the clumping. No 212-216? Weird.
We DO have facts, @LadyMeowMeow. Each of those anecdote is a fact if accurately represented. And what CB presents for the SI table is a fact. It may not be correct this year, but it’s a fact that page 11 is what they presented as the percentile table pertaining to the SI’s.
Our biggest problem is that the two sets of facts contradict one another.
I believe you said the spreadsheet had over 1000 entries, so maybe 500 sophomores and 500 juniors. That is a pretty good size school. Agree with @Mamelot, 215 is suggested which matches last year. On the positive side, no scores at 220+. I would say the school data does not support this being an easy PSAT test with so many high scorers.
But again, we have 2% of the students being in the top 1%.
@azcpamom sorry! I responded too fast before finished reading! I was excited with your real data! Can you flush out those sophomores? We don’t want them in our table.
@azcpamon, continuing that thought; 2% in top 1%. If we eliminated half the students and just went with the top 1%, (5-6 students) that could put the cutoff as low as 211.
@DoyleB the facts on the ground clearly are beginning to contradict the “reset” SI percentile table. But I simply can’t believe CB would release something that is so wrong. They. have. the. actual. data.
There’s a lot more to this story that we are missing and I hope it will unfold sooner rather than later.
Allow me to quote a bit of two paragraphs from the (truly excellent) compass article. I couldn’t agree with the article more.
“The data consolidation rationale does not exist for PSAT/NMSQT percentiles. As of October 28, 2015, every student who would ever take the 2015 PSAT/NMSQT had done so. Full results could have been tabulated and used for percentile calculation and reporting. Instead, College Board elected to use a sampling method that has not been disclosed and that is subject to the error inherent in any sampling.”
“The reasons behind the decision to change the percentile definition and the default reference group may be valid, but the fact that the changes tend to amplify the percentiles and include an opaque leap from test-taker group to a Nationally Representative sample creates a dubious impression. A productive solution would be to release the actual numbers for test-takers and publish all research study results. The new SAT debuts on March 5, 2016, and many of its components are being built on the same research studies and with the same methods used for the PSAT. It would seem prudent to establish credibility with PSAT data now rather than play catch-up after final SAT numbers are released.”
@Speedy2019 wrote: “But I do wish to hear more anecdotes from schools showing they expect SF numbers to drop. If some schools are increasing, other schools have to be decreasing in SF…”
I agree. To balance this out, at some point we should hear from a GC that says, “We typically have 8-10 NMSF per year. This year it looks like we won’t have any”.
One fundamental assumption being made just may be wrong…the assumption that the # of NMSF this year for a particular school will be in line with previous years. Drastically different test than in the past.
@DoyleB Of all the anectotes out there, have we seen even one that says they have less? Nope, I don’t think so.
@Mamelot “the ground clearly are beginning to contradict the “reset” SI percentile table” - did I miss a post? Did DB reset their Understanding scores???
@dallaspiano Are you posting from school? although many enjoy your detailed posts, we wouldn’t want your studies to suffer…truly. I find it odd that you have so much time to post on this thread during the school day. Please don’t let your grades suffer because of CC.
In another thread that I started, I generated a list of cutoffs per-state. The way I did it isn’t particularly defensible, but it (intentionally) yields cutoffs that are between those predicted by the percentile table and those predicted by the concordance table.
Shall the true believers fall back, and regroup at the following? (It’s still not supported by many of the anecdotes, but it’s at least not quite as optimistic as the straight percentiles. Plus my daughter will (just barely) get in!!! )
)
)