“…Several studies from across the social sciences have documented a variety of issues in existing research, including publication bias (Sterling (1959), Rosenthal (1979), Franco, Malhotra, and Simonovits (2014)); specification search (Simmons, Nelson, and Simonsohn (2011), Humphreys, de la Sierra, and van der Windt (2013), Brodeur et al (2016); bias and failures to replicate (Open Science Collaboration (2015), Chang and Li (2015), Camerer et al (2016), Ioannidis, Stanley, and Doucouliagos (2017)); and outright fraud (Simonsohn (2013) and Broockman, Kalla, and Aronow (2015)). 1 This research raises concerns about the validity of empirical work in economics and in social science more generally.…”