COMM 162 Lecture Notes - Lecture 12: Null Hypothesis, Statistical Hypothesis Testing, Test Statistic

• School: Queen's University
• Department: Commerce
• Course: COMM 162
• Professor: Tatsiana Levina

Document Summary

Hypothesis testing: allows us to determine whether enough statistical evidence exists to conclude that a belief about a parameter is supported by sample data. Research hypothesis: a statement of what a researcher believes will be the outcome of an experiment or study. Statistical hypotheses: relate to the formal process of conducting a statistical test of a research hypothesis. Composed of both null (h0) and alternative/research (h1) hypotheses that jointly exhaust all possible outcomes. Alternative: captured the research hypothesis (what we could like the data to show us) Statistical hypotheses use symbols, rather than words, to represent population parameters. H0: u = 5, h1: u =/= 5. H0: uty = uly, h1: uty =/= uly. H0: p >= 0. 01% , h1: p < 0. 01% H0: uq <= uuoft, h1: uq > uuoft. Calculate how likely it is to observe data as extreme as the data we have collected under that assumption. We reject the null hypothesis if it is sufficiently unlikely.