- Education
-
Research
Current research
Talent
-
Collaboration
Businesses
Government agencies and institutions
Alumni
-
About AU
Organisation
Job at AU
RUNSAFE and collaborators now present “Seven Statistical Sins when Interpreting Statistics in Sports Injury Science”. This work summarizes seven key-messages from methodology-related editorials published in 2017.
2017.12.20 |
Committing a “statistical sin” is easy to do in sport injury research. In this editorial, we briefly explain how practitioners and readers of science can avoid falling victim to statistical misinterpretation when interpreting articles in the sports injury research context. RUNSAFE and collaborators hope you find this helicopter-perspective on pitfalls in research valuable and easy to interpret.
The sevens sins include:
Sin #1: Trusting coincidence
Sin #2: Getting causation backwards
Sin #3: Forgetting to consider multifactorial nature of sports injury development
Sin #4: Avoiding absolute measures of association
Sin #5: Relying on p values.
Sin #6: Deceptive graphs
Sin #7: Not considering sub-group differences
We encourage readers to access the full editorial by Nielsen et al. at the BJSM website: click here.
Reference:
Nielsen, R.O.; Chapman, C.M.; Louis, W.; Stovitz, S.D.; Mansournia, M.A.; Windt, J., Møller, M.; Parner, E.T.; Hulme, A.; Bertelsen, M.L.; Finch, C.F.; Casals, M.; Verhagen, E. Seven Sins when Interpreting Statistics in Sports Injury Science. Br J Sports Med, December 2017: In press.