Score inflation on the new SAT - Gifted Issues Discussion Forum

Gifted Issues Discussion homepage

I fear that the SAT (once the Scholastic Aptitude Test, before "aptitude" became non-PC) is becoming a progressively worse instrument to detect unusual verbal or mathematical ability. Here is a Concordance table.

*************************************************************

Why your new SAT score is not as strong as you think it is
By Nick Anderson
Washington Post
May 11, 2016

Quote

Many college-bound students across America are celebrating this week what appear to be impressive results from the revised SAT. But in general the scores are not as strong as they seem at first glance.

It turns out the new test comes with a degree of score inflation. Simply put: a 1300 on the SAT is not worth as much as it used to be.

Figuring out what the new SAT scores mean, and how they compare to old SAT scores or to ACT scores, is likely to be a major source of confusion for students and parents in the next couple of years following the debut in March of a major revision to the nation�s oldest college admissions test.

Charts the College Board released Monday show that for a vast swath of students, new SAT scores are comparable to results that would have been 60 to 80 points lower on corresponding sections of the old SAT.

For instance, if a student earned a score of 1100 on the new SAT, out of a maximum 1600, that would correspond to a score of 1020 on the math and critical reading sections of the old SAT. Same hypothetical student, but two different tests, with an 80-point bump on the newer one.

Why the bump?

�The scores have risen because of design decisions made by the College Board,� said Adam Ingersoll, a college admissions consultant in California. �Kids are not smarter. The test is not �easier.� The test has just changed. It�s a different test.�

Err, personally, I'm not convinced that a multiple choice test with questions that need to be answered in a minute each was ever an instrument for detecting unusual verbal or mathematical ability. A big vocabulary or an ability to solve a standard set of questions related to basic algebra and geometry? Sure. Fast processing speed? Sounds reasonable. Predict "Will this person do okay as a college freshman?" Sure. But can it measure talent for speaking/writing or proving new ideas in mathematics/solving thorny engineering questions? Well....I kind of doubt it.

That said, I recall that they decided to redo the test a few years ago after a big "scandal" over low SAT scores. There was a lot of worry about SAT scores showing that some large percentage of students weren't "college ready." Given that a very limited slice of the population (say, 25% if we're being nice and generous) will ever be college ready, the results at the time seemed high to me. Does anyone else remember that?

So the cynic in me is thinking, "Here at the College Board, we're believe that everyone can and should go to college.* See? Higher average SAT scores are great!

"**

--------------

*Or at least, everyone should take the SAT.

**And if you look at our conversion tables, which are taped to the bottom of a filing cabinet in the basement of our office on Mars, you'll see that the new scores correspond to a drop of 60-80 points over old scores.

Look. The important thing is that the scores are rising.

That means improvement, which will ultimately lead to a world of continued increased profitability and various other forms of important success.

If the scores were falling, we would need to panic because that would lead to economic stagnation, decline, and failure. This would lead to profound sadness and other negative things like extra sadness.

So, this is great news and we need to all be happy.

The end.

As usual, the article is meant to be provocative and a little misleading. While it is true that converting with the concordance table may lead to an estimate that is up to a maximum of 80 points lower on the combined score, it isn't true across the board. I reviewed all the tables fairly thoroughly since two of my kids took the March exam. For example, DS' score converts a total of 40 points lower on the combined score.

As far as the difficulty levels of the old versus the new SAT, both my kids find the new Math sections more challenging than those on the old SAT but somewhat easier than the two practice tests they tried. The new Language/Writing section was somewhat more challenging than the old SAT and definitely more difficult than the two practice tests. However, the Critical Reading section was easier due mainly to the change in the vocabulary portion. I am not really sure how accurate the Concordance tables really are. For that matter, how accurate the percentiles are at this early stage. I read that the percentiles are based on a December administration (presumably mostly seniors who were paid $50) to kids who have previously taken at least one old SAT during the Fall.

I am also still confused regarding the relationship between some of the concordance tables versus some of the percentiles. For example, DD received the exact same Math score on both the old and the new SAT. According to the concordance table the new score is about 20 points "inflated" so she arguably performed worse, but her national percentile (compared to mostly December seniors) jumped about five percentage points which suggests she performed better.

What I do like about the new SAT is that there is so much more information compared to the old SAT. Instead of just four measures, there are now eight measures plus more detailed breakdown within measures.

Originally Posted by Val

Err, personally, I'm not convinced that a multiple choice test with questions that need to be answered in a minute each was ever an instrument for detecting unusual verbal or mathematical ability.

The late Julian Stanley of Johns Hopkins, one of the most important figures in gifted education and the creator of an SAT-based talent search, would probably disagree.

How Hard Is the New SAT?
The score...me out�and they�re significantly higher.
JAMES S. MURPHY
The Atlantic
MAY 12, 2016

Quote

SAT scores were released on Tuesday, more than two months after the first administration of the revamped exam and years after the media started to speculate about the new test�s difficulty. And it turns out that, after years of stagnation, the scores have gone up�rather significantly.

Maybe students are smarter and more prepared for the new exam than their predecessors. But chances are the upward score drift is a product of the test�not the test takers. As in the past, each section of the SAT is scored from 200 to 800. The average combined score on the old, three-section test was right around 1500; that would indicate that the average score would be about a 1000 on the new, two-section test. But the new average is actually close to 1090. according to data released by the College Board, which administers the SAT.

So has the new test been dumbed down? Or has the revamped exam, which is supposed to be more closely aligned with real school work, created greater opportunities for students to shine?

Some of this score inflation was anticipated years ago, in March 2014, when the college board revealed plans to change the SAT: Major Changes in SAT Announced. The old thread from September 2013 on The College Board's false alarm on SAT scores may also be of interest. (Thank you, Bostonian, for consistently posting relevant news stories on the forum, throughout the years.)

The nature and purpose of the SAT has changed over the years, from being more of an aptitude test to more of an achievement test (more like the ACT). This is touched on a bit in a few of the posts on this old thread, A tale of two tests: SAT&ACT.

To address Quantum2003's questions on the accuracy of the concordance tables: the SAT has included score norming questions in its exams for years, as mentioned in a few posts on this old thread, beginning here.