5 ms·
100% agree There's a simple statistical argument behind it: the variance drops as you increase the number of tests. Let's say your true ability is A and you h
by klassik 3y ago
100% agree
There's a simple statistical argument behind it: the variance drops as you increase the number of tests.
Let's say your true ability is A and you have one measurement M1. The variance (or, let's put it in another way: the probability that your M1 is waaaay different from your A) goes down if you also get M2, M3, M4 and take an average.
In other words: your measurements become more precise and the stress goes down a lot
- standardUser 3y agoOne could say that grades are an even better metric! They measure all manner of academic topics over many years and across many different instructors.
- ken47 3y agoOnly if the grades are assigned in a standardized manner across all candidates, which they definitely are not.
- mr_toad 3y agoTaking four independent samples isn’t the same as taking one sample and cutting it into four.