IQSS Faculty and Student Question "Replicability Crisis"

March 7, 2016
IQSS Faculty and Student Question "Replicability Crisis"

A group of researchers claimed that half of all psychology studies could not be replicated. However, a team that includes IQSS director Gary King, as well as affiliates Daniel Gilbert and Stephen Pettigrew, is calling this claim into question.

A paper published by the Open Science Collaboration (OSC) in 2015 states that the OSC's attempt to replicate 100 published psychological studies resulted in a mere 36-47% success rate (depending on criteria used). However, in their recent technical comment in Science, Gilbert et al. explain that they found multiple flaws in the OSC's study--regarding statistical error, bias, and infidelity to original methods used. (A more expanded summary of the team's results can be found in a recent Harvard Gazette article.)

Additional comments have been published in a continued dialogue over this topic. The OSC's response can be seen here; Gilbert, King, Pettigrew, and Wilson have followed up with "A Response to the Reply to Our Technical Comment on 'Estimating the Reproducibility of Psychological Science'", followed by "More on Estimating the Reproducibility of Psychological Science'".

Gary King is the Albert J. Weatherhead III University Professor at Harvard and Director of IQSS. Daniel Gilbert is the Edgar Pierce Professor of Psychology at Harvard and an IQSS faculty affiliate. Stephen Pettigrew is a doctoral candidate in the Department of Government at Harvard and an IQSS affiliate. Timothy Wilson is the Sherrell J. Aston Professor of Psychology at the University of Virginia.