^^^ I’d miss that one 
I’m just stunned how complicated this is. It shouldn’t take an entire instruction manual to figure out what your test score is.
"Even back in the pre-2005 version of the SAT that was scored out of 1600, the PSAT was scored out of 240. " Once upon a time, there was a verbal and a math PSAT score, exactly like a verbal and a math SAT score–only the numbers were 1/10 as high. Your “selection index” was the sum of your verbal and math scores. You could easily predict your SAT score by multiplying your PSAT score by 10. No complicated formulas or difference between your score and your “selection index”. I understand why they created two verbal scores but what purpose do the rest of these new complications serve?
To all those looking for a cut-off. I’m not sure how you can predict this numerically. It will depend on how they decided to construct the scaling charts. But the cutoffs will remain as they always were, something like the 99.5 percentile. However much the college board wants to scale and massage the numbers to give some artificially created score and “selection index”—that is what it is going to be. It would be far simpler for them to simply report percentiles and forget about all these artificial numbers and formulas.
For those of you confused about how the scaled score is calculated from the raw score, you need to refer to the scoring publication, particularly the chart on page 7 https://collegereadiness.collegeboard.org/pdf/scoring-psat-nmsqt-practice-test-1.pdf. You can see that the scoring is blatantly unfair–sometimes getting an extra question wrong gives no change in your scaled score and sometimes it does. Those numbers for verbal are then multiplied by two, magnifying this effect. Considering that just a few points makes the difference between winning large scholarships or not, this is inexcusable. The math scores seem to be divided by 10. Why the math and the verbals are scored on such different scales is also bizarre.
It does seem a little complicated, but I’m not sure it’s blatantly unfair. I remember in school if you got within a certain range (i.e., a set of numbers) you would get a certain score (e.g., A, A-, C…). Well, same thing here. If (referring to page 7) you get 12 or 13 answers correct in the Reading section, you would receive an overall Reading score of 19. It’s just not 1-to-1, certain counts roll up to the same final score. Very common.
And, since everyone is scored the same way I don’t see the unfairness. Multiplying by 2 doesn’t do anything but get the number to look relatively similar to years past.
At least this is my take on all this. I believe I got a C in statistics in UG though.
The one thing that I haven’t gotten past is that I still feel like the new format/scoring favors the Reading / Writing sections more than years past. I have nothing to base this one but gut feeling.
Look at the chart on page 7. Student A scores 47 reading 43 writing, missing only one question for a scaled score of 75. Student B scores 46 reading 42 writing, missing 3 questions for the same scaled score of 75. Student A scores 44 math, missing 4 questions, Student B scores 43 math, missing 5 questions. Both get the same score of 740. So both have the exact same score even though student A missed only 5 questions and student B missed 8. This has nothing to do with the difficulty of the questions, just where the number missed randomly happens to fall in the rounding the college board is doing when they scale these scores. That doesn’t seem unfair to you?
Now consider that in a high-cutoff state you are only allowed to miss about 7 questions on the entire exam if you want to qualify for NMSF. That number will vary depending on where your score falls with respect to the college board’s arbitrarily considering two scores to be the same. If you are lucky enough to have it rounded up like student B, your score is artificially inflated and you could win over other students with a higher total score whose numbers just happened to fall on the wrong side of those rounding cutoffs. That doesn’t seem unfair to you?
The first part does not seem unfair to me, no. Again, I think we’re used to this with current letter grades when you have tests that give you numerical results (e.g., 82-87/100 = B- ). Plus, it is the same for all, and I don’t consider it ‘rounding’ but Ok. I don’t think CB needs finer stats than this to differentiate? But, I give CB the benefit of the doubt here I guess.
The different state cutoffs is a different story, but seems unfair.
Is this ‘rounding’ a new issue? I would think its always existed, but I’m new to this.
The state cutoffs are unfair, but at least there is a good justification for them. The program is trying to find top students, which is not exactly the same thing as the highest achieving students, and it recognizes that top students in many areas don’t have access to outstanding educational opportunities and so won’t perform quite as well as those who do.
But there’s no good reason to muddy the scoring achieved by as much as 3 questions when you are drawing a very important line between 7 wrong and 8 wrong. There will be many cases where students who answered more questions correctly miss getting the scholarship while students who missed one or two more questions will make it, depending on whether they got lucky or unlucky with this score chart. Why should this be? Getting the necessary scores is already way too dependent on speed and accuracy rather than what the student knows. There’s no need to put additional arbitrary lines and luck into this process.
@mathyone, @chengallen - Perhaps the score chart they are showing in the practice test is not how they actually grade it? I’m sure it all computerized so maybe when they grade it, they use more precise individual calculations and round the final score to a whole number…
Does anyone know what the cut offs would be for the state of Colorado?
I’m wondering if anything we are seeing is going to be accurate. My dd is finding these tests way easier than any others she has taken. We suspect they are going to have a very high number of students scoring extremely well. Even though by all predictions she should be way above the margin for NM, we are not counting on it.
@Mom2aphysicsgeek Same thing is happening with my son. The practice PSAT was really easy. I’m going to have him take as many new practice SATs this weekend as possible to see if its just a fluke with the 1 practice PSAT we have. We are going to be using the CB free “new SAT” practice tests here:
https://collegereadiness.collegeboard.org/sat/practice/full-length-practice-tests
I am worried about that as well. If the test is difficult, they can simply adjust their scaled score conversion chart to compensate for that. But if the test is too easy there is no way to distinguish the students. Even the current situation is pretty bad in this respect because the test is easy enough that in high cut-off states you can only afford to miss a few questions–simply mis-timing one section can easily knock you out of a scholarship. My daughter didn’t finish the last section–she’s not a fast reader and she was getting tired. 3 questions not answered and that probably would have knocked her out of contention. (Can’t stand hearing the disability accommodations crowd insist that the time limit is not a problem for non-disabled students). Even confusion about one or two ambiguous or poorly written questions could also make the difference. The more I see of these tests, the more apparent it is that a difference of a few questions is pretty meaningless. I really wish they were a lot harder with enough time that the students don’t have to race through and that would make it easier to tell the best students, rather than judging students on whether they misunderstood a question or are more error-prone when rushing under extreme pressure or needed an extra 2 minutes to finish a section.
It’s silly to worry about what the cut off scores will be because the score conversion chart will vary depending on exactly how they scale it. They can multiply all the numbers by 796 and subtract pi, and you will still need to be approximately in the 99.5 percentile.
My dd has taken 2 of the SAT practice tests and only missed a very small # on both. She is worried by how easy they are.
@suzyQ7 @mathyone @Mom2aphysicsgeek
My personal belief is that College Board will avoid a situation where they make the test too easy and compress the top end of the scoring range (ie too many people get very high raw scores). This would be quite unusual and they usually have a good amount of statistical research on questions to prevent this.
It’s a good idea to calibrate your scoring using the 4 new SAT practice tests released by the CB. If your children are scoring near 1600 on these tests, then it may just be possible that they’re performing at a very high level. On the other hand if they miss more questions than on the practice PSAT, they can work on the revealed weaknesses.
This is an unfortunate case where we’re in a transitional period so there may be some growing pains. I remember when I was in high school and the 2400 SAT came out as I graduated. There was a lot of fear and confusion but that quickly settled down once the new test was better understood.
I assume they tried to calibrate these questions because they have been testing them for the past year or so. However, the students they tested them on were used to the old format, maybe haven’t even read the instructions for the new questions. And the style is different enough that some students may have known it was an experimental section and not tried very hard. If students who actually practiced the new style of questions can do a little better (this would be expected with a little familiarity), then yes there could be a problem.
Oops. I was trying to delete draft, not post.
How long do they think it will take to release scoring for the first new PSAT? Has CB said?
About two months after the test
https://collegereadiness.collegeboard.org/psat-nmsqt-psat-10/scores/student-score-reports
@3scoutsmom
Sorry, but I do not understand. You say that the highest selection index will be 228.
However, if I, say, got all the problems in PSAT correct. No OMIT, No wrong questions.
Then, My Selection index would be (47+44+17+31)*2=278.
(47=the number of Reading section question)
(44=the number of Writing section question)
(17=the number of Non-calculator Math question)
(31=the number of Calculator permitted Math question)
What do you mean by “228”, then?
@MadAstro you have to convert your raw score using the chart the college board provides
https://collegereadiness.collegeboard.org/pdf/scoring-psat-nmsqt-practice-test-1.pdf
see page 7
What that scoring sheet didn’t describe is converting your raw math score to the number used in calculating the selection index (math sections combine to a single raw score). You need to divide by 20. Then, add the three numbers up and multiply by 2