Report says "about 1.5 million test-takers meet requirements to enter NMSC’s competition each year . . . " that sounds like boilerplate of course but it’s what they are saying for this year.
@billchu2 You can link tinypic picture here. It’s not allowed but… or provide url with spaces between addresses
So then if I repeat my calculation for Illinois (Illinois cutoff was 2.39 stddevs from the mean), it becomes
In[66]: 146 + (2.39 * 28)
Out[66]: 212.92000000000002
This reasoning, tortured though it is, leads to a SI for Illinois for 2015 = 213, close to the other predictions.
Can anyone who is not self-taught in inferential statistics validate or refute this model?
Are the percentiles on the document blacked out for you guys too? Also, if 214 is the 99+, would a 211 have a chance in Illinois. Illinois was a 215, well below the 99+? Kind of wishing I had picked stats ove Econ as my elective this year!
I love all you smart stats people. Help me understand what you are seeing as the likely commended and likely top SI for NMF. Another thread was saying the top SI for some states could be 228 (!). Its not jiving with what I’m reading here.
@Jay12321 The numbers were blacked out on my mobile device but fine on my desktop
is there anyway u can view a local percentile on the score report? i got 1420 and that was 99 national
@Dave_N That is interesting.
Last year Illinois was 215, 2 point above 2013 (99%). This year you are predicting 8 point above 205 (99%). The curves must be very different between 2014 and 2015.
@a20171 If you have 215 that puts you 99+ %ile in SI. Looking pretty good for ny but no one can say definitely of course.
@payn4ward, indeed it is interesting.
I’m not sure I believe it. I just wanted to list out my work so others can critique it. I think this “distance from the median” has some validity but I am not sure.
I wouldn’t want anyone to get their hopes up or down based upon my doodling.
I think that “median absolute deviation from median” would be better, but I need a few more hours and access to different materials to check.
S just now received an email with his code to access his PSAT scores. Kind of funny because his scores are already up in his CB account. But for those waiting, codes are still rolling out.
Also, i got a 1480. Si of 220. Is that good enough for Te as?
This is what CB has stated on their “Understanding 2015 Scores” for concordance tables, as in why they are useful:
“To understand approximate NMSC Selection Index scores on the redesigned PSAT/NMSQT for students who took the PSAT/NMSQT in 2014 or earlier. (Concord NMSC Selection Index scores from the PSAT/NMSQT from 2014 and earlier to the redesigned PSAT/NMSQT [2015 and future].”
Thus, I believe that the concordance tables that CB has put on their website allows us to predict cutoffs for this year. Although they are preliminary, I doubt that they will change that much. I’m confident that I got NMSF in Ohio, as my 220 SI this year translates roughly to a 221, using the by section concordance. Plus, the percentiles that are posted for the selection indexes eased the amount of stress I had. Who knew the percentiles would be so generous? So, I think everyone can breathe a lot easier now.
^everyone in OH with a 220 you mean?
If you need to be in the 99th percentile for semifinal its status, why was the cutoff 207 in Iowa and 204 in Wyoming (last year) the percentiles for those scores were 98 and 97 respectively using the CB chart showing national si percentiles. Can someone explain?
@kyrieIrving2 It doesn’t need to be 99th % - someone misspoke. NMSF goes to top scorers in a state, based on number of high school seniors in the state. If a particular state has 300 allocated NMSF based on population, then the top 300 scorers will be NMSF. The lowest selection index of scorer number 300 will be the cutoff.
Last year every student with a 99+ SI percentile made semi-finalist in 49 out of 50 states. Only New Jersey and DC were exceptions. Translating that into this year, with 214 being the lowest SI for 99+, we should expect that the highest state cutoff for the 49 states would be around 213 (with NJ and DC being slightly higher).
That means California, Mass and similar states around 213.
Texas around 210.
Most states between 200-209.
It looks like prepscholar’s predictions are close to being accurate. They are probably too low by about 2-3 points.
Add 3 to prepscholar’s estimates and you will have a good number to use.
“National” Merit is, of course, a bit of a misnomer. State Merit is more like it as semifinalist status (unlike commended status) is actually based on state scores/population (roughly speaking). That’s why there’s such a big spread in the scores that qualify for NMSF but just one score for commended across the nation. Commended is the only one based on the nationwide scores. Last year I believe commended was around 202. The average for NMSF was around 213 or so but in the highest states, NJ and DC, a score of 225 was required. A few other states were not far behind.
It will be interesting to see what happens with the scores this coming year. I think the numbers in NJ and DC will remain quite high while others might drop some. I’ve seen estimates of 220 for NJ/DC but I think it’s more likely to be 221+ based on all the concordance charts and the national SI percentile chart.
I feel for guidance departments this year. The composite score isn’t what matters and they’ll have to re-educate parents about how to think about the scores and what they mean. Now kids with the same score will likely have different SIs; one might qualify and one might not. Someone with a lower composite might qualify over someone with a higher score. That’s going to cause stress/anxiety for many.