4 ms·
> there would be no significant difference between the students in the study I didn't say this. I say that you can't generalise to students NOT in the study,
by silvestrov 8y ago
> there would be no significant difference between the students in the study
I didn't say this. I say that you can't generalise to students NOT in the study, i.e. students with parents who didn't care to participate in the lottery.
- croon 8y agoOf course you can. You can control for the factor/variable you said was the "single most important", which is done through the design of the study, by your own admission. And then look at results from the other factors. Whether your factor is important or not doesn't even enter the picture.