Henry Collins
-
How to avoid mis‑interpreting non‑significant statistical results in your di...
Method Help
Non‑significant p‑values do not automatically prove there is zero effect. Misinterpretation of non‑significant findings remains very common in published manuscripts. What writing practices prevent you from over‑reading non‑significant statistical test outputs?
Chao Guo, Brandon Nguyen and 17 others4 CommentsView more comments-
Avoid phrasing like “proved there is no effect” when you observe non‑significant results.
2 -
Report confidence intervals alongside p‑values to show plausible effect‑size range.
2 - View more comments
-
-
How do you handle partial loss of experimental raw data after completing ana...
Method Help
Sometimes you finish analysis and produce figures, yet small segments of original raw data get accidentally lost. You still retain processed datasets and output results. What standards should you follow for manuscripts under such partial‑data‑loss scenarios?
View more comments-
Document exactly which portions of raw data are permanently missing in lab records.
2 -
Preserve every intermediate processed dataset derived from original raw inputs.
2 - View more comments
-
