 The probability of a research finding being true depends on several factors, such as study size, study design, and the ratio of true-to-no relationships among the relationships probed in a given field. When effect sizes are small, when there are more studies conducted in a field, when there is greater flexibility in designs, definitions, outcomes, and analytical methods, and when there is greater financial and other interest and prejudice, the likelihood of a research finding being true decreases. Furthermore, simulation suggests that for most study designs and settings, it is more likely for a research claim to be false than true. Additionally, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias. These issues have significant implications for the conduct and interpretation of research. This article was authored by John P. A. Ionidis.