Statistical Sciences 1023A/B Lecture Notes - Lecture 7: Normal Distribution, Null Hypothesis, Sampling Distribution
29 Mar 2017
School
Department
Professor
Document Summary
Hypothesis testing a standard procedure for testing a claim (ie. a hypothesis) about a property of a population. Steps in hypothesis testing: formulate the hypotheses, compute a test statistic, determine the p-value for the test statistic, make a decision based on the p-value. Two-sided hypothesis test not sure which way the coin is biased, so we test both. One-sided hypothesis test only interested in whether aspirin reduces risk. For a given research question there are 2 hypotheses: null hypothesis (h0) Statement of no difference", equal to", no relationship". Observations are the result of pure chance: alternative hypothesis (ha) Observations are the result of a real effect or difference. Sampling variability variation of a statistic from sample to sample due to chance composition of the sample. How hypothesis testing works: assume the null hypothesis is true o. How many standard deviations away is the statistic from the mean, if the h0 is true? o.
