datacolada.org
[18] MTurk vs. The Lab: Either Way We Need Big Samples
Back in May 2012, we were interested in the question of how many participants a typical between-subjects psychology study needs to have an 80% chance to detect a true effect. To answer this, you ne…