Searz wrote:
But is that really necessary?
It seems like a waste of time to me.
I would for example answer the three questions Jhoi put forth the same, with a maximum of one step of variance (1/10), so the variance seems pretty negligible.
Well, it's for control purposes. Although the variance seems negligible to you, because you perceive the questions to be the same (and admittedly, they are), it's to ensure that the answer isn't just spontaneous - so on the off-chance that you do get a high variance, you could consider the answers to be invalid. Of course, the only way to measure this is to do it repeatedly, with a minimum sample size of 3 questions.
Regardless of whether you think it's necessary or not, for research purposes, it's standard procedure. If he doesn't do it, some of his results or conclusions may be considered invalid for research purposes.
You need to log in before commenting.
<Veteran>
-
My Soraka Guide | My Review Service
Thanks a lot for the sig, jhoi! :)