Whenever life gets too tough on my ego, I comfort myself by thinking back on my glory days--i.e., 1965--when I got two 800s on my SATs. Yeah, I know: pathetic. But it's been a slow 20 years. Now this friend of mine has made things worse by insisting that he knows someone who scored 810 on one of his SATs. They must have changed the scale, I said. No, countered my friend (hah!), we all know that you don't need to get all the answers right to score 800. OK, I conceded. Well, this guy, my friend continued, got everything right, so they had to give him more than 800, and besides, I saw his score report and a letter from the College Entrance Examination Board confirming that there had been no mistake, so there! Gee, I don't want to call my friend a liar, but I maintain that it is impossible to score 810 on a scale of 800. Was the scale changed, is this whole report bogus, or would the CEEB actually do such a thing?
Illustration by Slug Signorino
This seems like a pretty trivial thing to get worked up about, A., but I suppose after two decades of non-achievement your sense of proportion starts to get a bit frayed. Be that as it may, you’ll be happy to know that there is no such thing as an 810 score on an individual SAT test, 800 being the scholastic equivalent of the speed of light. I am assured on this point by an official of the admissions testing program of the Educational Testing Service of Princeton, New Jersey, which administers the Scholastic Aptitude Test. He concedes that a typographical error might possibly have resulted in someone getting an 810, but says that under no circumstances would a letter have gone out “confirming” the error. You might want to invite your friend, therefore, to cough up his alleged evidence.
All this is not to say that somebody might not have earned a score above 800–quite the contrary. Unadjusted scaled scores in some years ranged as high as 840, due to a process ETS calls “equating,” which we’ll discuss anon. The 800 maximum is an arbitrary limit that ETS imposes mainly to keep neurotic geniuses like yourself from complaining that they got ripped off. The idea behind the present SAT scoring system is that scaled scores should be equivalent from one year to the next–in other words, so that getting a 650 in the verbal section requires the same amount of brainpower in 1997 as it took in 1965. The problem, obviously, is that the difficulty of the tests fluctuates from year to year, as does the caliber of the test-taking population. (For many years, as I’m sure you’re aware, it appeared America’s college-bound seniors were getting progressively dumber each year). So every year ETS has some students take a special SAT section consisting of questions from tests given in previous years. The results don’t affect the students’ scores, but they do give ETS a statistical basis for establishing the relative difficulty of a given test, which is then used to “equate” this year’s scoring scale with those of earlier years.
The one drawback to this otherwise admirable system is that a perfect score on an “easy” test might earn you only 801, whereas a perfect score on a more difficult test might be worth 840. If such scores were reported without adjustment, the people who got only 801 might complain that they got lower scores through no fault of their own. Now, a rational person might object that it doesn’t make any conceivable difference, since either score is enough to guarantee you admission to virtually any college in the country. Unfortunately, as you can surely attest, A., high school prodigies are not rational. Hence the 800 maximum.
Finally, I should point out that even if somebody did get a score ten points higher than yours, the difference is statistically meaningless. ETS states that only scoring differences in excess of 65 points indicate a genuine difference in ability.
Send questions to Cecil via firstname.lastname@example.org.