Crowdsourcing hypothesis tests: Making transparent how design choices shape research results

Maura Pozzi, Justin F. Landy, Miaolei Liam Jia, Isabel L. Ding, Domenico Viganola, Warren Tierney, Anna Dreber, Magnus Johannesson, Thomas Pfeiffer, Charles R. Ebersole, Quentin F. Gronau, Alexander Ly, Don Van Den Bergh, Maarten Marsman, Koen Derks, Eric-Jan Wagenmakers, Andrew Proctor, Daniel M. Bartels, Christopher W. Bauman, William J. BradyFelix Cheung, Andrei Cimpian, Simone Dohle, M. Brent Donnellan, Adam Hahn, Michael Hall, William Jiménez-Leal, David J. Johnson, Richard E. Lucas, Benoît Monin, Andres Montealegre, Elizabeth Mullen, Jun Pang, Jennifer Ray, Diego A. Reinero, Jesse Reynolds, Walter Sowden, Daniel Storage, Runkun Su, Christina M. Tworek, Jay J. Van Bavel, Daniel Walco, Julian Wills, Xiaobing Xu, Kai Chi Yam, Xiaoyu Yang, William A. Cunningham, Martin Schweinsberg, Molly Urwitz, Eric L. Uhlmann

Research output: Contribution to journalArticle

23 Citations (Scopus)


To what extent are the results of research investigations influenced by subjective decisions that scientists make as they design studies? Fifteen research teams independently designed studies to answer five original research questions related to moral judgments, negotiations, and implicit cognition. Participants from two separate, large samples (total N > 15,000) were then randomly assigned to complete one version of each study. Effect sizes varied dramatically across different sets of materials designed to test the same hypothesis: materials from different teams rendered significant effects in opposite directions for four out of five hypotheses, with the narrowest range in estimates being d = -0.37 to 0.26. Meta-analysis indicated a lack of overall support for two original hypotheses, mixed support for one hypothesis, and significant support for two hypotheses. Overall, none of the variability in effect sizes was attributable to the skill of the research team in designing materials, while some variability was attributable to the hypothesis being tested. In a forecasting survey, predictions of other scientists were strongly correlated with study results, and average predictions were similar to observed outcomes. Crowdsourced testing of research hypotheses helps reveal the true consistency of empirical support for a scientific claim.
Original languageEnglish
Pages (from-to)451-479
Number of pages29
JournalPsychological Bulletin
Publication statusPublished - 2020


  • research robustness
  • scientific transparency


Dive into the research topics of 'Crowdsourcing hypothesis tests: Making transparent how design choices shape research results'. Together they form a unique fingerprint.

Cite this