Publication Bias and P-hacking

How to evaluate a study design? Dr. Martijn Schuemie shows the result of this study, isotretinoin on a specific outcome, inflammatory bowel disease in g a graph. He explains that he wants to evaluate this design so he did a replicate similar experiment to examine the result.

He also explains the concept of P-hacking and publication bias. P-hacking is usually done through manipulation for a certain purpose. The manipulation could include the timing to stop collecting data, decide whether collected data will be transformed or not, which statistical tests and parameters will be used, etc.

If you want to learn more about P-hacking, click here.

You can click on the related links to see the research results in the final slide, Interpreting observational studies: why empirical calibration is needed to correct p‐values. We will continue to hear from Dr. Martijn Schuemie to explain step 2 for 21st-century development.