STAT 101 Lecture Notes - Lecture 19: Null Hypothesis, Test Statistic, Sample Size Determination
Document Summary
Stat 101 - introduction to business statistics - lecture 19: hypothesis testing (continued) Set up the appropriate null and alternative hypothesis. Compare the test statistic to the cut off value or compare the pvalue to alpha. The cut-off value comes from looking up the appropriate quantile in the t-tables, but we often round this value to 2 for simplicity when the test is two-sided and = If the test statistic exceeds the cut off value or the pvalue is less than alpha, then reject the null hypothesis. If we use a cut-off value of 2 for the test, then given the null is true there is only a 5% chance of making a type i error. We call this 5% a significance level and write it as = 0. 05. Likewise, if the cutoff value was 1. 645 then = 0. 10. The test statistic for testing a single mean against a hypothesized value is called the t- test statistic.
