Questionable research practices
3 important questions on Questionable research practices
What is a questionable research practice regarding the power of your study?
optional stopping
- Adding more participants than originally planned because your study does not have a significant effect
- stop collecting data prematurely because the dataset as it is causes a significant difference
- again problematic because analyses are done multiple times, increasing risk of false positive
What is a questionable research practice regarding manipulation of existing data?
- Trying different outlier corrections to see if one correction causes a significant difference
- again multiple corrections increasing chance of false positives
What is a qrp regarding the statistical analysis?
- Using statistical method selection to your advantage to see if any significant results arise
- multiple t-tests instead of an anova
- bayesian stats with a strategically chosen prior
- some exotic non-parametric tests
- non paremtric tests are fine if your data doesn't satisfy assumptions of parametric tests. But don't chose non parametric tests because parametric tests give results you don't want
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
