Reporting policy
Null results are published
A null result is a test that did not find the effect it was designed to detect. We publish them in full, with the same detail as any other outcome.
Unsuccessful tests are informative
A well-powered test that finds nothing narrows the space of plausible explanations. Withholding it makes the remaining literature look stronger than the evidence is.
Publication bias distorts the field
If only positive outcomes are reported, pooled estimates drift upward and replication becomes impossible to interpret. Preregistration plus full reporting is the only defence.
Neutrality has to be symmetrical
We commit to reporting the direction of every preregistered outcome before we know it. That commitment is meaningless unless negative outcomes are published as prominently as positive ones.
Failure modes are reusable
Exclusions, protocol deviations and instrument problems are documented so other groups can avoid or test them directly.
What we commit to
Every preregistered outcome is reported with its direction, effect estimate and interval, whether or not the threshold was met. Studies are not abandoned because they are heading towards a negative result, and a negative result is never re-described as an exploratory success.
Exclusions and protocol deviations are published with the report, so a reader can judge whether the null result reflects the hypothesis or the execution.
What a null result does not mean
A negative outcome rules out effects larger than the smallest effect the study was designed to detect. It does not demonstrate that no effect exists at any size, and it says nothing about the sincerity of the people who took part.