Or the cut offs will be higher than expected. I don’t know what I want to believe.
@pbleigh, here are SAT scores for John Cooper:
Mean SAT Scores:
2014-15: 1965
2013-14: 1948
2012-13: 2020
2011-12: 1981
2009-10: 1978
Converted to PSAT:
2014-15: 196
2013-14: 194
2012-13: 202
2011-12: 198
2009-10: 197
@pbleigh - thanks!
Any chance you could ask the GC what the average was last year? That would be somewhat interesting.
Also - if you were willing to ask - what would be great to know was “what was the 7th highest score this year”?
How to become finalist? The higher SI or a combination of all your other qualifications, essay, etc?
Has anyone read through this validity study for the new SAT? https://collegereadiness.collegeboard.org/pdf/redesigned-sat-pilot-predictive-validity-study-first-look.pdf
I wonder if CB will publish a similar report about the PSAT. The validity study was run in Fall 2014, the report seems to be published just now, January 2016.
PREDICTION:
The key is to compare the percentile numbers of 2014 to the percentile numbers of 2015 at the College Board sites. Percentiles are the most logical way to look at this because percentiles are percentiles no matter what the scale:
In 2014, the 99%+ percentiles began at an SI of 224.
In 2015, however, the 99%+ percentiles begin at a raw score of 1430. When you convert that 1430 to the new SI you get 215 (±).
Thus, a 215 this year (being the first 99%+ percentile), is equal to 224 from last year, which was also the first 99%+ percentile. That is a drop of 9 points. Therefore it is reasonable to assume that the 2015 cutoff scores will drop around 9 points to be consistent with 2014 scores, using the 99%+ as the marker.
I extrapolated the numbers from the following College Board documents:
John Cooper not “blowing the test out of the water” is a VERY GOOD sign for TX cutoff being lower.
Very good school.
Big credit to @VandyAlum93, msg #2094
“With so many bands at the 99% level, the (% of test takers who scored the same as you) will be very small (well under 0.1%). Your percentile is not going to change much”
Back up with real data from 2015 SAT:
https://secure-media.collegeboard.org/digitalServices/pdf/sat/sat-percentile-ranks-composite-crit-reading-math-writing-2015.pdf
At highest end of 99 range (associate with score of 2300), We have 1410 students with score 2300.
And we had 1,698,521 test takers, one is ahead a bunch plus 1410 then percentile change 0.083%
With old definition : score 2300 associate with 99.46%tile
With new definition: score 2300 associate with 99,54%tile (99.46 + 0.083)
@liketowrite, not change that much
@mantua wrote: “Percentiles are the most logical way to look at this because percentiles are percentiles no matter what the scale”
Apparently that’s not true this year. At least not in the bizarro college board world we are living in. In this world we have percentiles of test takers, percentiles of phantom non test takers, percentiles of real kids plus phantoms, percentiles of research groups, and percentiles of weighted research groups. We have percentiles of kids who scored “above”, and percentiles of kids who scored “at or above”. We have 3 percent of kids scoring in the top 1 percent. We have higher percentages of sophomores with high scores than percentages of juniors with the same high score. We have a certain percentage of parents signing their kids up to take the SAT because they scored high on the PSAT.
Indeed it’s a strange new world.
@thshadow wrote:
“Yes, I really didn’t want to be rude. @likestowrite had explicitly asked me to respond in I think 3 different posts. I ignored many of them because I couldn’t think of anything helpful to say. I finally did respond, with “apologies in advance” and smiley faces. Regardless, my response was bad because I knew it was rude, but I honestly couldn’t figure out a way to phrase what I wanted to say without being obnoxious. So in summary - I hate being rude - I knew I was being rude - I did it to try to communicate - I hoped he would take it in a light-hearted way. And again, I’m sorry, I was definitely hoping that no one’s feelings would be hurt.”
I thought that this was a venue where posters could share information and theories and speculation. I had a speculation about how to convert the SI chart, via a definition change, into a chart that made sense and correlated well to past charts. I took it not from a math background, but from a philosophy background. I put it forward and asked for your help, @thshadow, bc you had weighed in on the SI % chart the issue earlier and done such a good job. I really wanted to hear your opinion not only of my method but of the results. It was honest inquiry of a board member, a mother of a junior who took the PSAT, reaching out to someone who had taken a leading role.
I thought I might receive, “The results look good but I have to say the method is at odds with statistical analysis principles x and y. Thus, I just don’t think it holds.”
Instead it get, “I honestly couldn’t figure out a way to phrase what I wanted to say without being obnoxious. So in summary - I hate being rude - I knew I was being rude…”
I have had a WONDERFUL experience on this college confidential board with the exception of your admittedly rude, and avoidably so, response.
To all those posters who have been so helpful, thank you. I still have not heard what is wrong with the end product I have reached and who knows, perhaps y’all should take a look at it bc it is the closest end solution I have found to making sense of this SI% chart. I know sometimes math solutions can be reached that are close to correct but the method is wrong. I have no Idea what I will find to be the case when this process ends and September reveals results. However, I won’t be following along because if I do, I will want to participate. However, @thshadow has made me feel that participation, without certainty of ones efforts, could result in such a rude response. That is not the game I was in for. Best of luck to you students and the students of the parents on this board.
@dallaspiano I am sorry but I won’t be engaging in the 99+ end speculation. That end of the chart was difficult to figure out with my methodology which relied upon moving the percentiles up one percent. Since there was no 99++, I did not know where to start the 99+s. I chose 222. It is my daughter’s score and I knew it concorded with a 224 in 2014 and that was the first spot to become a 99+ then. Figuring out the new top level, the numbers where 99% becomes 99+, I think will be difficult. I know you are in TX and are interested in those numbers, so I am sorry about that. Best of luck to you and I hope that you make NMSF!
@liketowrite, glad you clarify.
How about Q2 and Q3, they are not about 99+%tile at all
Yes, I like to make it to NMF but it’s not my choice (my score will be compared with 220,000+ TX test takers and 1,724, 416 others all around US
Anyway, thank you for saying “luck”, I probably need a lot of it along with my growing up life ahead
<<@Mamelot I know how to check individual scores for concordance but I don’t know a good, systematic procedure to test this chart against the concordance charts. Any ideas?>>
@likestowrite from #1900 I wish I had some good ideas on that!!! Everyone’s been running sophisticated models or writing programs to crank out concordances . . . I’m just grinding through a few here and there (usually on a post it note . . . ). If you can convert current SI’s to a range of total scores you can concord those pretty quickly from that first set of tables (I haven’t really used them much so not sure how accurate they are). Start at 200 and advance every five or so SI points and see what you end up with. In the perfect world you will concord to scores that are consistent percentiles from where you started. I would think that would verify the accuracy of your table. Good luck!
@likestowrite - I’m sincerely sorry for making you upset. That was not my intention.
I should have just said “Your predictions seem fine. But note that justifying a full percent shift in percentiles due to the change in the percentile definition is not correct.”
@thshadow, thank you for your lucid explanation of how to transform scores when whole groups of SIs have the same percentile values. I actually understand it now and am persuaded by your reasoning. But I think shifting by one score may nonetheless be insufficient because of other inflationary factors, i.e., those having nothing to do with the change in the definition of percentile. There are still too many kids in the 99th.
@pbleigh – re Post 2096 – thanks for the info about John Cooper’s (Houston) 109 test takers!! In looking at the Understanding You Scores Report a 1283 is about 93%ile & the SI range is 190-192. So the data you are reporting seems consistent with the past - though we down know if there is a large number of stops at the top and also towards the bottom. Wonder if they would share any info about numbers of scorers at the higher end? thanks again!!
I did a random FB search for for “psat” - a woman in Atlanta posted yesterday: “Wow. I like this new PSAT. Of my juniors who took it, 40% had a selection index of 99%.”
So, maybe only 10 juniors took it? Or 20? But maybe 200 did. Who knows. It seems to be an unusual phenomenon to her though…
2115 Edit: "selection index in the 99th percentile"
My son missed 4 total on reading and 3 out of the 4 were in a row on the Douglass passage. He was not happy with that passage.
Here is my email message to daughter’s tutor and her response. Judge it how you will. We are in Texas.
Email:
“How would you characterize the scores you are seeing?
Many 220+ scores?
Many 210-219 scores?
Many 200-209 scores?:”
Response:
“A few above 220, and a greater amount from 210 to 219. Of course, I certainly have a number of students who scored below 210, but the majority of them did little or no work specifically in preparation for the new PSAT, instead choosing to focus on the ACT or the “old” SAT”
My thoughts:
I’ve always thought the high cutoff states would come down. Testmasters shows this. SI% table shows this. Question to me is by how much. Tutor’s response seems to indicate last years 220 is coming down. A reasonable guess is between 210-219. Testmasters is at 217. I’m predicting 212. Tutor’s students tend to be from affluent area, high achieving students.
Absolutely, there are many people who don’t think the percentile tables are completely accurate (even after adjusting for the changed definition). @DoyleB had a long post about it. He said something like he can’t come up with any intellectually honest way to reconcile the differences between the percentile tables and the concordance tables (and the anecdotes). I agree with that statement. I’ve made a set of guesses where I kindof split the difference - which really doesn’t make any sense at all, as it disagrees with everything… 