Selective Hypothesis Reporting in Psychology: Comparing Preregistrations and Corresponding Publications
Van den Akker and colleagues compared four hundred and fifty-nine registered research plans in psychology against the papers that eventually reported them, looking specifically at the hypotheses. More than half of the studies omitted a hypothesis that had been registered, and a similar proportion added one that had not been. Around one in five reported a hypothesis whose predicted direction had changed. Taken together, three in five studies showed at least one of these discrepancies, which means that selective reporting of hypotheses is normal practice rather than an occasional lapse, even among researchers conscientious enough to register a plan. One expectation was not met: the added and altered hypotheses were not clearly more likely to be statistically significant than the faithfully reported ones, though the authors note they may lack the power to detect that. Replication studies showed the problem less often than original studies. Their recommendation falls on researchers, reviewers and editors alike. This vault cites it on what a registration does and does not guarantee.
Van den Akker and colleagues compared four hundred and fifty-nine registered research plans in psychology against the papers that eventually reported them, looking specifically at the hypotheses. More than half of the studies omitted a hypothesis that had been registered, and a similar proportion added one that had not been. Around one in five reported a hypothesis whose predicted direction had changed. Taken together, three in five studies showed at least one of these discrepancies, which means that selective reporting of hypotheses is normal practice rather than an occasional lapse, even among researchers conscientious enough to register a plan. One expectation was not met: the added and altered hypotheses were not clearly more likely to be statistically significant than the faithfully reported ones, though the authors note they may lack the power to detect that. Replication studies showed the problem less often than original studies. Their recommendation falls on researchers, reviewers and editors alike. This vault cites it on what a registration does and does not guarantee.
Written from the abstract, OpenAlex.
What our sources record
Reproduced as it was written when this source was checked, figures and all.
van den Akker, O. R., van Assen, M. A. L. M., Enting, M., et al. (2023). "Selective hypothesis reporting in psychology." AMPPS, 6(3). 459 preregistrations matched to publications: 52% omitted at least one preregistered hypothesis, 57% added one, with no sufficient evidence that added or altered hypotheses were more likely to be significant.
Where we use it
- What Forty-One Participants Buys, Why We Look
Reliability of this reference
The link below was matched against the publisher's record on title, first author and year, and all three agree.
The source
https://doi.org/10.1177/25152459231187988
DOI: 10.1177/25152459231187988