QTM 100 Lecture 11: Hypothesis Testing
Document Summary
Gender discrimination experiment males = 21/24 = 0. 875 females = 14/24 = 0. 5833. Observed difference in proportions is simply due to chance. There is gender discrimination, observed difference in proportions not due to chance. Hypothesis testing framework: consists of null hypothesis and alternative hypothesis; conduct the test under the assumption that the null hypothesis is true. Null hypothesis (h0): hypothesis represents the status quo. Alternative hypothesis (ha): hypothesis that represents the research question and what is being tested for. Often simply is just alternative hypothesis null hypothesis. We conduct a hypothesis test under the assumption that the null hypothesis is true. Either via simulation or traditional methods based on the central limit theorem. Null hypothesis the data do not provide convincing evidence for the alternative hypothesis. Alterative hypothesis data provides convincing evidence for the alternative hypothesis. If they do, then we reject the null hypothesis.