Authors : Thomas Gall (University of Southampton) & Zacharias Maniadis (University of Southampton)
Paper No. 1504
Abstract
The `credibility crisis' in science has led to various suggestions for
reform by researchers in several scientific disciplines. We use game
theory and general equilibrium arguments to evaluate recent proposals
to increase transparency by strengthening disclosure requirements
when an empirical study is submitted for publication. We nd that
a policy that prohibitively increases the cost of less severe questionable
research practices, such as selective reporting, tends to decrease the
overall rate of researcher misconduct, because the rate of `felonies',
such as fabrication, also tends to decrease. Accordingly, proposals that
aim to prevent lying by omission (Simmons et al., 2011, Landis et al.,
2012, Fanelli, 2013) are likely to be effective in reducing researcher
misconduct. Blunt measures such as government audits, which can be
used to counteract pure fraud, do not seem equally effective.