Document worth reading: “Abandon Statistical Significance”
In science publishing and many areas of research, the established order is a lexicographic alternative rule whereby any consequence’s first required to have a p-value that surpasses the 0.05 threshold and solely then is consideration–usually scant–given to such elements as prior and related proof, plausibility of mechanism, analysis design and knowledge prime quality, precise world costs and benefits, novelty of discovering, and completely different elements that alter by evaluation space. There have been present proposals to change the p-value threshold, nonetheless instead we propose abandoning the null hypothesis significance testing paradigm absolutely, leaving p-values as merely thought-about one in every of many objects of data with no privileged place in scientific publication and selection making. We argue that this radical technique is every smart and clever. Abandon Statistical Significance