SAT Scores Drop

August 30, 2006

Comments

America’s high school kids are getting infinitesimally dumber, as evidenced by the biggest SAT score drop in 31 years.

The high school class of 2006 got stuck with a new, longer version of the SAT and didn’t fare well on it. Average reading and math scores fell a total of seven points — the sharpest decline in 31 years.

Experts agreed the dip in combined math and critical reading scores on the college entrance exam was related to the new version of the test — but differed as to how. The updated exam, with a new writing section, also features more advanced math questions and replaces analogies with more reading comprehension.

Average reading scores fell from 508 to 503 and math scores fell from 520 to 518, the College Board announced Tuesday, with the changes hurting boys more than girls. Boys’ reading scores fell eight points, while girls’ dropped just three. And girls scored 11 points better than boys on the new writing section. Boys’ and girls’ math score fell two points each to 536 and 502, respectively.

The College Board, which owns the exam, downplayed the drop, saying it amounts to a fraction of one question per exam. The board’s explanation: about 3 percent fewer test-takers, out of 1.5 million, tried the exam a second time. Combined math and reading scores typically rise 30 points when a student retakes the test.

If it hasn’t already happened, we can start the countdown to a major Democratic leader blaming President Bush for this. Sure, it’s a totally different instrument and the difference is less than one question missed in the aggregate but the bottom line is that this happened on his watch. Perhaps if we had used the money spent trying to bring democracy to Iraq on test prep, this wouldn’t have happened. And Osama bin Laden is still on the loose, too.

15 responses to “SAT Scores Drop”

McGehee

August 30, 2006 at 9:25 am

And don’t forget HALLIBURTON!!! And, and DIEBOLD!!!!1!!
Herb

August 30, 2006 at 9:36 am

JJ: I am rather supprised that you advocated spending more money on “Test Prep”. While a prep for the test is proper, the schools around the country should be teaching a more rounded education rather that teaching for the test only as they have done in the past. The quality of the teaching skills by the highly paid teachers should be the main issue here. However, the teachers unions protect those teachers that have other int rests in mind over teaching.

When you combine low teaching quality and the palaces that are called schools with their extra curricular activities together, it is no supprise that scores are lower. The only reason scores were higher in the past is that schools and teachers were only teaching students for the test and not teaching their students a well rounded “Basic” education.

reading, writing and arithmetic. ( and don’t spare the “Hickory Stick”).
Mark

August 30, 2006 at 9:47 am

As far as “teaching the test” is concerned, isn’t that what NCLB resulted in for many school districts? Instead of teaching fundamentals, they are so concerned about getting a marked as a bad performer, and how a school is measured is by test scores.
legion

August 30, 2006 at 10:25 am

Mark,
I have a number of close friends who are teachers, and you are quite correct – that is one of the main reasons professional educators have pretty consistently reviled NCLB.

James, you say it sarcastically, but another one of Bush’s failed, politically-motivated, all-hat-and-no-cattle policies is a contributor to this problem.
James Joyner

August 30, 2006 at 10:31 am

legion: There’s zero evidence of a problem here. The scores are down literally less than one question’s worth since last year and it’s a totally different test. We have no baseline.

(I have no real opinion of NCLB, having not researched it much. Considering how poorly prepared the students I had when teaching college during the Clinton administration were, though, I find it hard to believe they’re any worse now.)
Anderson

August 30, 2006 at 10:43 am

If it hasn’t already happened, we can start the countdown to a major Democratic leader blaming President Bush for this

Right, because one has to search high and low to find something significant to blame on Bush … oh wait.
just me

August 30, 2006 at 11:50 am

I am not sure you can compare the previous test scores with the current ones, if the test is different. If the reading and math portions were the same, but with the added writing you could better compare.

That said, at the lower elementary level we don’t do much teaching to the test. As a matter of fact curriculum hasn’t really changed since NCLB. . Just what information is being left out if we are teaching to the test? Most of these tests cover general knowledge, I think it is mostly just a whine on the part of teachers who in general don’t care for standardized testing (although my caveat is I work in a lower elementary school, where the basics are the core curriculum-things may vary more at the middle and high school level).

I actually like the fact that they have included a writing portion on the test. Writing is one skill we seem to have gotten away from teaching.
Brendan McManus

August 30, 2006 at 12:25 pm

_If it hasn’t already happened, we can start the countdown to a major Democratic leader blaming President Bush for this._

When GWB ran in 2000, he cited educational decline as a reason not to vote for Gore. It is neither surprising nor unfair that Bush will be blamed for it. It is just as fair as himselfhaving done it and it is accurate. Federal education policy under Bush has been anti-education, anti-child, and anti-science. The only thing it has been pro- is “business-style, lets make up some benchmarks so we can say we’re accountable” assessment regimes that cost vast amounts of time and money. Bush is a moron and he’s doing all he can to make sure every child meets that standard.
Michael

August 30, 2006 at 12:37 pm

NPR report on this yesterday also noted that ACT scores this year where the highest they’ve been in 20 years. This is a non-story, as should have been evident by the story itself noting that the SAT this year was longer, and had fewer people take it a second time (people generally get better scores the second time).
yetanotherjohn

August 30, 2006 at 1:36 pm

I ran a quick calculation on the impact of the 3% not re-taking the test.

.97 * (503 + 518) = 990.37
.03 * (1021 + 30) = 31.53

990.37 + 31.53 = 1021.9, so only a small portion of the 7 points would be explained by lack of re-takers. Now as a general rule, you should put your best explination forward first. So if they cite something that would have less than 1% impact out of 7 points gone missing, then they better hav a raft of other explinations.

When I took the LSAT, it was a year of change for the test. New sections in, old sections out, general revision, etc. I got my test scores and was a little disappointed (if I remember correctly, I got top 5%). Then I received a “revised” test score. The number remained the same, but the national percentage changed from top 5% to top 0.2%. They had initially released the test scores with what they thought were going to be the national average results and then had to go back after they got the actual results tabulated.

Michael,

Can you link to the NPR report? My quick search found 2005 was unchanged from 2004.

Further as recently as 2001, ACT scores were higher than 2004/2005.

So I’m not sure what is up with the NPR report, since it doesn’t seem to match the ACT news releases.

As far as spin goes, the 2004 rise in ACT scores marked the first rise in the scores since 1997. So obviously anyone who is not partisan can see that the NCLB is starting to rectify the educational quagmire bestowed on the nation by Clinton.
Michael

August 30, 2006 at 2:55 pm

yetanotherjohn,
Sorry, but I have no link to the NPR report. I heard the report on the radio (that being the ‘R’ in NPR) yesterday evening on my drive home. However, I did find a CNN article which claims the same thing (Note this is about 2006 scores, so it does not conflict with your link about 2005 scores).
http://www.cnn.com/2006/EDUCATION/08/15/act.scores.ap/index.html

I was also able to find on the ACT site confirmation of both reports:
http://www.act.org/news/releases/2006/ndr.html

If you were looking for confirmation of the claim that ACT scores have risen, there you go. If you wanted confirmation that NPR said it, sorry.
Tano

August 30, 2006 at 3:08 pm

From what I have heard, the ACT is more focused on testing factual knowledge, and the SAT more on thinking skills. That ACT scores are up, and SAT scores down certainly seems consistent with the spread of NCLB-effects; i.e. less critical thinking, more teaching to the test.

Damn Bush….
yetanotherjohn

August 30, 2006 at 3:44 pm

Michael,

Thanks. I wasn’t so worried whether NPR or whom ever had reported it as I was the data. When I saw the August 17 date on the 2005 data, I did a brain fart and didn’t check the year.

So the largest increase in 20 years and the highest scores in 15. That kind of reinforces the “coming out of the Clinton educational quagmire” meme.

Tano,

Do you have anything to back up your claim of a difference between the two? My quick search couldn’t find it. Also, would the comparison be to the old SAT or this new SAT?

Difference Between SAT and ACT
SAT vs ACT, the main difference between the two tests is that the ACT has a Science section. ACT English is similar to SAT Writing (not including the essay); ACT Mathematics is similar to SAT Math; ACT Reading is similar to SAT Critical Reading; and ACT Writing is similar to the SAT essay.

also

Skills Heavily Tested SAT Vocabulary and Reading; ACT Math Grammar and Reading; Math

I also found this interesting. The 21.1 ACT score would correspond to just over 1500 on the old SAT score. Either the comparison is totally bogus or there is an even bigger discrepancy.