4 ms·

I worry about survey-based studies with small sample sets which reinforce conventional wisdom...

• dirtyaura 10y ago
  With a quick glance, I got an impression that it wasn't a survey, but this was an actual experiment. They sent 17000 randomized emails to 5000 early-stage investors through AngelList. Emails highlighted either the team, traction, existing investors or some combination of these. Researchers measured how many investors responded with "View" or "Get an Intro" buttons. I think that is pretty well designed experiment given the difficulty of researching such topic.