movement-science-stats

To check your movement science manuscript for common statistical mistakes, run the movement-science-stats checklist on your results paragraph and analysis description to identify the ten common errors outlined by Makin & Orban de Xivry (2019).

Circular analysis in neuroscience data is a selection bias that this checklist identifies and corrects by validating your unit-of-analysis choices and ensuring independent test conditions align with predefined statistical QA criteria.

Yes, you can use this checklist to review kinematics time-series and bilateral trial-level data, as it flags timepoint-specific testing risks and recommends when to aggregate to participant level or use mixed models.

To prevent p-hacking and uncorrected multiple comparisons in movement science research, apply the checklist to your analysis code to systematically review reporting decisions and correct inflated units of analysis before publication.

The best way to interpret null results in motor control studies is to run the checklist to evaluate your statistical power and group comparisons, preventing the over-interpretation of null findings in your manuscript.

You should not use independent tests for each timepoint in kinematics data when it inflates the unit of analysis or creates uncorrected multiple comparisons, requiring aggregation to participant level or mixed models instead.