Hello,
I am teaching a class on biostatistics this semester and want to introduce an exercise where students critically evaluate the methods section of scientific literature. So, I am looking for examples of published papers with problematic experimental design or statistics (e.g. non random sampling, pseudo-replication, violation of assumptions, type 1 error, etc.). If you are not comfortable calling out scientists in public, please feel free to send me suggestions in a direct message. I'll keep a list for anyone who would find this useful in the future. Thanks for your help!
-Claire
In this paper you can find some explanations and illustrations of how the arithmetic mean is sometimes not applied correctly when it is needed to use the geometric mean:
Article Measuring Forecasting Accuracy: The Case Of Judgmental Adjus...
You might be interested in this paper: Article Randomized, double-blind, placebo-controlled, linear dose, c...
Two of the 3 authors have disavowed the study, so I think that I can publicly recommend this as an example.
Usually the problem is pseudo-replication
Article Pseudoreplication and the Design of Ecological Field Experiments
In this paper you can find some explanations and illustrations of how the arithmetic mean is sometimes not applied correctly when it is needed to use the geometric mean:
Article Measuring Forecasting Accuracy: The Case Of Judgmental Adjus...
- Badges
- Science topic
- Similar topics
- Papers