Looking @ scores from my S’s class it appears that CA cutoff will be 218 or above. One test prep site is predicting 219 for CA and seems accurate.
Finally got the email with the access codes! Didn’t work. Why do they taunt us? So cruel…
Hello. My friend got a 192 si. That is 97%national representative. Will he be commended?
@nillupa no, an SI of 192 is the 93% (look at the understanding the PSAT scores 2015 chart). To be commended you’d probably need a 199 (high 96 percentile) or 200 (low 97 percentile)
@miottawa LOL
Really not sure what to believe here- am I correct in thinking NJ and DC, MA, etc. will remain around 219-222 for cutoff and not around 213-215? I have a 221 in NJ, can I be confident in my chances? The stat work done by @Dave_N is making me suspect it will be lower than 217-218, but I’m really unsure
Do we know how many students took the PSAT?
*juniors
@miottawa How’d you get them? My friends and I still haven’t received our access codes…
I just randomly got the email that I was supposed to get last Thursday.
@Fambruhghini @Dave_N 's formula predicted 222 for NJ. Not that it’s reliable, just correcting your quote.
http://talk.collegeconfidential.com/discussion/comment/19181083/#Comment_19181083
But if they haven’t graded or uploaded the score reports, having the access code doesn’t necessarily mean that we will get our scores.
I read 220 for NJ/DC on one site, but I think 221-222 is far more likely. @Fambruhghini - you might just squeak by! In any event that’s a great score so congrats. I can’t imagine it would be as high as 223, but I really don’t imagine it will be below 219 at the very very lowest.
Thanks @payn4ward , but crap now I really am screwed :///
@plmdin this is a terrible question but the cutoff means that score and everything above go through, correct?
@Fambruhghini It’s frustrating - but you’ll just have to wait to see how it all plays out. I suspect we’ll all have a better idea when the revised concordance charts are released and we get to go through all this number crunching again.
@Fambruhghini - no, not a terrible question. The cut score means THAT score and all scores above it go through. So last year it was 225. Everyone in NJ with 225+ went through. If the cut is 221 or lower you’ll be fine and if it’s 222 - well, we won’t dwell on that.
Hmm… @plmdin, how would you explain the SI percentiles then? Theoretically, using the percentiles, the cutoff for Texasish states would be 214ish. I know that they don’t agree with the concordance scales, but using the percentiles, the cutoff in NJ would be a lot lower.
Assuming a normal distribution, for those people interested in estimating a percentile more accurately than the 99’s and 99+'s provided in https://collegereadiness.collegeboard.org/pdf/2015-psat-nmsqt-understanding-scores.pdf, you can calculate it as follows. First calculate the z-score (number of standard deviations away from the mean score):
z-score = (SI - 146) / SD
Where 146 is the estimated mean (it is actually the median score reported in the PDF).
@billchu2 showed in his curve fit that a SD of 28 matched the overall distribution fairly well. However, if 99+ is taken to mean 99.5% or better (which I think I’ve read), than for values in that range, a SD value of 26.4 matches the first such value provided by CB (i.e., 214).
The percentile can then be pulled using an online calculator like this one:
http://www.measuringu.com/pcalcz.php
(use the one-sided distribution).
Alternately, if you have Excel, you can use the “NORMSDIST()” function.
Some results:
score z-score percentile
228 3.1061 99.91%
227 3.0682 99.89%
226 3.0303 99.88%
225 2.9924 99.86%
224 2.9545 99.84%
223 2.9167 99.82%
222 2.8788 99.80%
221 2.8409 99.78%
220 2.8030 99.75%
219 2.7652 99.72%
218 2.7273 99.68%
217 2.6894 99.64%
216 2.6515 99.60%
215 2.6136 99.55%
214 2.5758 99.50% ← I used this to help set the SD for this area of the distribution
213 2.5379 99.44%
212 2.5000 99.38%
211 2.4621 99.31%
210 2.4242 99.23%
209 2.3864 99.15%
208 2.3485 99.06%
207 2.3106 98.96%
206 2.2727 98.85%
205 2.2348 98.73%
204 2.1970 98.60%
203 2.1591 98.46%
202 2.1212 98.30%
Well, I might be totally wrong. I’ve not done the same detailed statistical analysis other people have done (and done so well!) I’ve also only tried to estimate the very highest cut off (for NJ/DC). I have no guesses for any other state. I don’t have all my calculations here so this explanation might be a bit thin.
Basically, I looked at the % (for every section) from the 2014 PSAT (Understand your 2014 PSAT score booklet) and at all the concordances for the 2014/2015 scores. There’s a lot of leeway there so I decided to be as conservative as possible and used the lowest number when given a choice of equivalent scores. I focused on a 2015 (SI)score of 220 because that was the new estimate of cutoff scores on one of the sites linked to in the threads here. I worked out all the combinations on both tests that would get to various scores on the old test (SIs for the new) and looked at the % of those subscore combinations.
I could not make a 220 on the 2015 test equate to a 225 on the 2014 PSAT test so I concluded a 220 was too low (and a 223 was likely too high). I worry just that in some way I am comparing apples to oranges