3 ms·
Not for an observational study. There is simy too much variation and too many confounding factors. There would be no statistical power and infinite questions on
by s1artibartfast 4y ago
Not for an observational study. There is simy too much variation and too many confounding factors. There would be no statistical power and infinite questions on the results.
Essentially everyone's condition and situation would be different.
Keep in mind that 1000 people is the size you might expect for a controlled trial where you change one single thing you expect to have a big impact like a cancer medication and even then you try to screen the patients to have as little variation as possible.