‘A Big Win’: Dubious Statistical Results Are Becoming Less Common In Psychology

More than a decade after psychologists began to seriously wrestle with questions about the rigor and reliability of their work, there are signals that the field is improving. A recent analysis of 240,355 psychology papers reports that “fragile” statistical results—a potential marker of poor research practices or samples that are too small—decreased substantially between 2004 and 2024.

It’s a “thorough and impressive piece of work,” says Mark Rubin, a psychologist at Durham University who was not involved in the study, published in Advances in Methods and Practices in Psychological Science in April. He adds that the finding is consistent with other recent signs of improvement in psychology, such as an increase in sample sizes reported in an April preprint.

Read the whole story (subscription may be required): Science