@MZJBajViQgdJqOB そうですね。二重検定の問題と、正規性検定の精度の問題からノンパラメトリック検定を選択した場合のαエラーが増えることが指摘されています。 https://t.co/ex13LfSEV2
2,243 followers
2,722 followers
e.g., I think Rochon et al. 2012 "To test or not to test: Preliminary assessment of normality when comparing two independent samples" is a great starting point for understanding the utility of simulations and is quite easy to reconstruct. https://t.co/PsDO
88 followers
RT @mueblesfeos: We really need to do a better job teaching people the difference between the distribution of a mean estimator and the dist…
792 followers
RT @statsepi: Do you really need that normality test? DO YOU!? https://t.co/cyx9fQTZkI