# Methods
## Survey
The analysis was restricted to current volumes.
For all articles published by *B&L* from 2011 to the 3rd issue of 2013, three raters (not blinded to the purpose of the experiment) investigated all published experimental papers (excluding reviews, simulation studies, editorials etc.).
For each experiment reported in a study, the stimulus/materials sections were investigated for descriptive and inferential statistics derived from populations that were exhaustively sampled without error.
If a descriptive and/or inferential statistic (such as mean and standard deviation) were reported, the study was coded as one where the researchers were interested in a known quantity, otherwise it was discarded.
If an inferential statistic (such as a *p*-value) was reported, the study was coded as one where researchers answered that interest with an erroneous parameter estimate, otherwise as one where researchers did not commit the error.
If a statement of the form that groups were thought equivalent regarding the parameter was made, such as claims that they were “matched”, “equal” or “did not differ”, and this statement was backed up by a *p*-value greater than 0.05, the study was coded as "accepting the null".
In cases of rater disagreement, the majority vote was registered.
Representative statements from studies committing an error are available online at https://github.com/jona-sassenhagen/statfail.