Erica Meltzer Reading And Writing

0 views

Skip to first unread message

Laurene Arrison

unread,

Aug 3, 2024, 11:42:44 AM8/3/24

to consmulcondsa

.3. The test is ADAPTIVE by section: Each section will have two modules. The first module has easy, medium and hard questions. Depending on how you do on the first module you will then see an easier second module (with more easy and medium questions) or a harder second module (with more medium and hard questions) You need to answer about 2/3 of the first module correctly in order to get placed into the second harder module

-*not that you would ever spend exactly x min per question - you would obviously spend more time on harder questions and less time on easier questions, but it useful to get a sense of the timing on the new test

.5. Much SHORTER READING PASSAGES (3-5 sentences) with only one question per passage. This is the biggest change of the test. You also have the ability to highlight, mark out wrong answers, and flag questions to come back to (within each section, but not across sections)

.6. The test app has a DESMOS calculator built-in and will be available during all math questions (so no more no-calculator section) You'll want to explicitly teach your kids the DESMOS calculator (it has a lot more functionality) so they can use it on the test. Kids can also bring their own calculator. Feedback so far has been that some kids use the DESMOS calculator exclusively and some kids use their own calculator for quick calculations and then use the DESMOS calculator for its graphing and function capabilities.

.9. The feedback from students has been amazing: They LOVE it. (The digital test went live overseas in March 2023) They say it FEELS a lot easier. (My students hated the long reading passages on the old test)

.11. If anyone remembers the debacle from the last time the SAT switched tests back on 2016 with a new test design and new concordance tables, etc, folks are saying that this switch is much easier. And so far the College Board is administering it pretty flawlessly. It looks like they learned from their mistakes back in 2016.

.12. Test Prep Material: You can access 4 tests now on the College board's app called the Bluebook. They will also be partnering with Khan Academy. College Board will be releasing two more tests in the fall along with a question bank. Third party vendors are creating mock test question banks and a similar online interface to fill the gap for students who want more practice

.13. Score reporting - much less helpful. They won't release old tests. Each test question is worth an unknown, varying, amount of points so they won't tell you how many questions you got correct. I'm not sure if you'll even be able to see your old test questions (like you can now with the paper test).

Originally, I was going to advise students next year just to take the ACT and let the SAT work the bugs out of the digital test. But now, hearing how much students love it and seeing how well the College Board is handling the transition, I have changed my mind and think everyone should at least look at it.

Great question - The PSAT used to be half the length of the SAT. Now the PSAT will be the same as the SAT, but it will use an easier and different scoring table. The SAT goes from 400-1600. The PSAT goes from 320-1520. And the PSAT is scored against other 10th graders while the SAT is scored against other 11th graders.

I am struggling to understand the "adaptive" part. If the student doesn't place into the "harder" questions, will her score potential be reduced? (I guess I don't understand how two students can get equivalent scores if they are ultimately taking different tests?)

Thanks for all the info! I have a rising senior, so I don't have to worry about this for a few more years, but it's really interesting to hear all of this. The College Board paid my college freshman son to take a digital SAT this past year as part of the testing phase, and he LOVED the desmos calculator.

Yes, there is a ceiling to your score if you end up in the easier second section. I don't know how accurate this is since the College Board isn't releasing this info, but tutoring folks have estimated that scoring range for the easier second module is 200-650 and the range for the harder second module is 400-800. So quite a lot of overlap. Folks seemed to indicate that statistically the methodology is pretty solid - meaning that, even for slow starters, kids should end up in the correct second module.

Yes, my understanding is that PSAT will have the same format as the SAT. We don't quite know that for sure yet, since College Board hasn't yet released a digital practice PSAT - but they should be doing that this summer.

With this adaptive test, with questions being pulled from a question bank, each kid is going to have their own test. But the College Board is confident they can come up with an accurate score that's comparable across kids and tests.

Yes - if you muck up the first section, the test will then give you the easier second section which will have a ceiling on your scores. (At the moment we think the range of scores possible for the easier second section is 200-650 and the range for the harder section is 400-800)

I was also skeptical. It's based on Item Response Theory which is a mathematically valid way to compose tests. Right now I have to plow through all the easy, medium, and hard questions on the existing paper SAT. But I am a high scorer - and the easy questions really aren't useful. I consistently get all of them correct. To really gauge my ability, the test should give me more medium and hard questions. And the reverse is true for the lower scoring students - giving them hard questions is a waste of time becuase they will miss most of them.

By giving students the right sort of questions, the test can more quickly figure out where they are going to score. So they can cut out 1/3 of the questions b/c high scoring students don't need easy questions and lower scoring students don't need hard questions.

As to taking different tests - it could be as simple as changing the numbers on the exact same question. Or they could give entirely different questions that statistically test the same. I have many, many beefs with the College Board. But the one they thing are brilliant at is statistical analysis - they have millions of data points and can very accurately predict student scores.

Okay, awesome! That makes me feel much better about things, since my kids are likely to score great on the Reading/Writing section, but a bit more middle-of-the-road on Math. (Didn't want the Math section to ruin the whole test for them.)

Agreed. I also hope there's more time on a standard (vs. extended with exceptions) basis. harder questions can also take more time for some students. Their ability to solve them, accurately, is what matters. NWEA/MAP has been doing adaptive tests for years. I MUCH preferred it over old Stanford versions and it proved very accurate for DD. Remains to be seen for DS.

Agreed. I also hope there's more time on a standard (vs. extended with exceptions) basis. harder questions can also take more time for some students. Their ability to solve them, accurately, is what matters. NWEA/MAP has been doing adaptive tests for years. I MUCH preferred it over old Stanford versions and it proved very accurate for DD. Remains to be seen for DS.

Apparently there is 20% more time. I totally agree with you about the time thing. Personally, I'd prefer no time limit for everyone and have it adaptive down to each question (the way the MAP does--though the MAP loses its utility for high performing students in mid-elementary). And it would be nice to have harder problems that are just harder and not taken from coursework beyond Algebra 2.