Ok, who would like to volunteer to put a CC SI % Table together using Definition A?
Seems to me, the cutoffs won’t match the concordance tables because the concordance tables predict high cutoffs to go down, mid range to stay same and low cutoffs to go up - vs - concept of cutoffs staying similar to last year.
Hmmm so much for being gone - the new article prompted a least a quick lunchtime note.
Changing the percentile definition just moves you up one row higher in the percentile table than previously. Not like the doubling from 99.0 to 99.5; it just bumps up by a single row - this doesn’t change the high part of the tables much at all, but could change the center of the tables a fair amount.
Example:
Old Table
73 99.3%
72 99.2%
71 99.1% Scored better than 99.1% of people
70 99.0%
69 98.9%
New Table
73 99.4%
72 99.3%
71 99.2% Scored as well as or better than 99.2% of the people
70 99.1%
69 99.0%
Your percentile value shown in the table has risen by 0.1%
At the high end, each row changes by less than 0.1%. So there’s percentile inflation, but it is very small.
However, in the center of the table, moving up one row could increase the apparent percentage by several points. It is much more of an concern there.
This isn’t much of an issue for us in trying to determine cutoffs.
The user vs national is more questionable.
@Speedy2019: Regarding any CC concordance tables, someone needs to be very cognizant of the change in definitions for the percentiles this year. Continuing my example from above, I now see that my D3’s current SI most likely DOES concord to previous SI’s on a percentile basis, whereas before they were off by up to 1%:
Current score of 220 this year concords to a 222-223 from last year.
220: At-or-above mid 99+ percentile in 2015
222-223: We don’t have last year’s percentiles but if I use 2013’s I get Above the high 99th %,.
Does this change Jed Applerouth’s conclusions? I don’t recall his discussing the change in percentile definition.
Applerouth never mentioned a percentile definition change.
Actually, what @DoyleB just posted is an interesting thought. At the high end, you just go up from say 99.1 to 99.2.
Let me ponder that.
@Speedy2019, @LivinProof, @likestowrite see my post #1618. Not as in-depth but you can see that the percentile tables are a lot closer now that we apply the CORRECT definition! =D>
Edit: not as in-depth as @DoyleB, that is.
So based on
Old Table
73 99.3%
72 99.2%
71 99.1% Scored better than 99.1% of people
70 99.0%
69 98.9%
Then we can interpret higher up of Old table
79 99.9%
78 99.8%
77 99.7%
76 99.6%
75 99.5%
74 99.4%
This way violates the basic concept - at higher scores the percentiles do not follow linear progression or evenly spread-out and up
So, IF the page 11 tables are representative of actual (and we don’t know that yet!) then Commended will be above the 97th percentile (because 50,000 / 1,700,000 = .029) which, on the page 11 table, is AT or ABOVE 98th percentile, which is 202. SALY.
@thshadow I’m beginning to believe more strongly that the cut-offs might not be all that different from previous years myself.
@dallaspiano wrote: “This way violates the basic concept - at higher scores the percentiles do not follow linear progression or evenly spread-out and up”
I understand that. It’s just an example. If you want to change the percentages for each row to:
99.34
99.27
99.19
99.10
99.00
or whatever makes you feel better, go right ahead. The point remains: the change in definition simply moves you up one row.
Looking at the commended level. Current 2015 SI% Table would indicate between 199 and 200.
Using Definition B, commended would be between 201 and 202 - similar to last year’s commended level.
Testmaster’s is predicting 210 using concordance tables. @Mamelot, still have a problem with the notion that the concordance tables are correct? Am I missing something?
Think testmasters may need to release new predicted state cutoffs - revision 3.
I like it now
@mamelot That is the conclusion I came to a couple of days ago. Based on all of the info being generated, it seems to me as if the SI scores are going to be fairly stable with historical numbers. When you think about it from NMSC’s perspective, that is probably the most logical path to take b/c it ultimately causes less confusion. The speculation that test scores would change might end up being the biggest source of confusion of all.
It also makes me laugh at myself b/c when I have talked to couple of people IRL about NM, they are clueless anything has changed and are going along with the premise of the normal avg cut-offs!
Credit to @DoyleB. Thank
%tile SI(2015)
99.54 213
99.51 212
99.46 211
99.40 210
99.34 209
99.27 208
99.19 207
99.10 206
99.00 205
Ahhhhhh! How can we 1.) move the percentiles up a notch 2.) state that national and user won’t be that different at the top part of the percentiles; and then 3.) decide that the new cutoffs will be a
That was supposed to say higher than before. Are we basing this off of 3 million test takers?
For those discussing whether it is likely Walton High School in Georgia could dramatically increase the number of NMSF over last year, just consider that, while last year (2016 program) they had only 16 NMSF, the year before (2015 program) they had a whopping 25, and the year before that (2014 program) they had only 13. So, in short, Walton’s total can swing wildly from one year to the next.
While I doubt they could double their number for the 2017 program, they could conceivably increase it by as many as 9 or 10, since they did that just two years ago.
Also, for those crunching Walton and Wheeler’s numbers, keep in mind that Wheeler is the STEM magnet in Cobb County, but also draw kids from their attendance zone, a lot of which come from less affluent and even poor neighborhoods. The two student populations at Wheeler are very different, thus Wheeler has some very high scorers (Magnet + gifted attendance zone students) and a lot of lower scorers (on-level attendance zone kids).
Walton (though a charter school) does not draw students from outside its attendance zones. These are all kids that live in the area zoned for Walton. They just happen to live in very affluent, highly educated neighborhoods. I would venture to guess that more than 60% of Walton students are classified as gifted. Walton is considered one of the very best public schools in Georgia and even ranks nationally. One year, US News and World Report ranked it in the top 10 in the nation.
@DoyleB & @dallaspiano – I am trying to follow the new percentile interpretations but am not sure of the implications - using the new info from the Compass Report -(thanks @Shelt29 ) can we project the cut offs for SI’s for GA/NY/TX/CT/CA/DC? Do we think that Test Masters looks reasonable or too low? thanks!
Repost #1480, I stand my ground (based on past PSAT years 2012 to 2014)
Testmaster published their estimate on Jan 8, 2016 - was considered HIGH by some CC posters in this thread
http://collegeadmissions.testmasters.com/update-psat-scores-cut-national-merit-2016/
Top Tier published their estimate on Jan 12, 2016 - was considered LOW by some CC posters in this thread
http://www.toptieradmissions.com/new-psat-scoring/
Seems to me, based on Definition B. that say 32,000 kids are being told they are 99% instead of 16,000 kids using SI% Table …
@CA1543 It just means that if we’re using the percentile tables to make estimates for cutoffs, we need to add 1 to our SI estimate.
Using my previous example again: Say we’ve looked at old cutoff data, and based on that we think the cutoff will be at the 99.1%. With the new definition, an equivalent point in the new table will be listed as the next rank higher up than that - 99.2% in my example…
2nd example: Yesterday we decided CA’s cutoff would be the 2nd 99 from the top. With the new definition, it would be one higher than that, or the highest of the 99s.
If that’s so, we would have 15000 11th score above TS 1480, and 15000 11th score above SI (2015) 221. I would say it’s possible and We, the US, number one smartest in the world :)) :(|)