3 ms·
It seems to me how the test subjects are selected likely matters. I presuming most studies advertise (in some form) for test subjects. That ad appeals to some b
by chiefalchemist 8y ago
It seems to me how the test subjects are selected likely matters. I presuming most studies advertise (in some form) for test subjects. That ad appeals to some but not to others, and that leaning could influence the results. In a way, the study starts there, not the lab.
If the study were a landing page, with a call to action, the conversions are influenced by the ad.
No doubt, 50% replication feels ugly. None the less, the fact is, the studies are slightly different. Maybe in theory that shouldn't matter, but perhaps the reality is it does. Again, per the PPC analogy.