Replicability and Questionable Research Practices in the Behavioral Sciences

Abstract

Replication is essential for scientific progress. It describes the extent to which results of an experiment are repeated under conditions identical to (direct replication) or different from (systematic replication) an original experiment. Successful replication enhances credibility of previous research and sets the occasion for future scientific discovery. The “replication crisis” has been publicized across various scientific fields. It is the failure of researchers to replicate findings of prior studies, along with the lack of published replication studies appearing in a body of scientific literature, generally. Challenges with replication compromise trust in science and erode the public’s confidence in the capacity of researchers to address socially important problems. Several viable solutions to the replication crisis have been proposed. One of the most well publicized is the open science movement, comprising practices such as study preregistration and open data sharing to enhance the visibility and accessibility of research. Additionally, improved research practices (IRPs) have been identified as alternatives to questionable research practices (QRPs) aimed at reducing researcher degrees of freedom that inject bias into research. Finally, editorial practices adopted by journals, reviewers, and editors can increase adoption of open science practices and IRPs to support transparency and replication.