Significance Tests
摘要
A hypothesis is an unverified assumption about a phenomenon, formulated and judged to be plausible by knowledgeable persons. Studies involving carefully planned and executed experiments play a leading role in the verification process. The methods used in the analysis of the data generated in such studies have been developed by statisticians and significance tests are a part of this branch of statistics. The idea at the core of this type of tests has a lot of appeal among experimental scientists, but has aroused some controversy in the recent statistical literature. The core idea consists in comparing the actual outcomes of the experiments with the outcomes to be expected if the hypothesis were false.