alt Hacker NewsI mean... usually those tests check the correctness of answers too, so you're comparing students under the same circumstances, evaluating how much (writing, calculations.... whatever) they're able to do, correctly of course, within an alotted time period. If someone can correctly solve 17 math problems in that time and someone else can do 21, the second one is "better" than the first, since they're both faster and their answers are still correct.
They could extend the test time for everyone, but in reality, you won't get many time extensions in real life, where speed is indeed a factor.
If someone can do 21 correct answers in an hour and someone else needed two hours to do the same, due to a faked disability, it's unfair both to the 1-hour student and an actually disabled student who might be missing a hand and needing more time to write/type with a prosthetic.
Replies
If I'm paying a professional by the hour, yes, it matters if he can do it in one hour rather than two.
I once hired a civil engineer to do a job for me, and he started billing me for time spent learning how to do it. I refused to pay him. (There was nothing unusual about the job, it was a simple repair task.)
But where is that level of speed distinction important? I just don't know anywhere where being 10% faster translates into much actual real value. If you can write a function in five minutes and it takes this other person 5.5 minutes -- do you really view that as the key difference in ability? Even in time constrained situations, compute/processing speed is almost never the issue.