Commentary to part V

Psychologists, like all humans, are subject to temptation. With regard to our data, we especially dislike having to admit to being wrong, because our reputations depend partly on our perspicacity, but also on our accuracy, integrity, and replicability of our data. We are especially tempted to ignore or deny errors if we are desperate and vulnerable, as the earlycareer examples presented in this Part suggest. But anyone can be tempted if the alternative-potential humiliation-is public enough and the sunk costs are large enough. The more invested we are, the more careful we must be. Lessons in temptation here come from discovering errors at various stages of the research process: midway through a research project if one has ill-advisedly been monitoring the trends, upon analyzing results that apparently undermine a pet theory, after discovering that another theory better accounts for one’s results, upon finding errors after the paper has been submitted for review, failing to replicate one’s own findings, and sharing data with the risk that someone else will fail to replicate your findings with alternative analyses. Contrary to human nature, we as scientists should welcome humiliation, because it shows that the science is working. The evidence accumulates no matter how the data will fall, human biases notwithstanding.