STAT 1000Q Lecture Notes - Lecture 30: Type I And Type Ii Errors, Null Hypothesis, Sample Size Determination
STAT 1000Q verified notes
30/30View all
1
STAT 1000Q Lecture Notes - Lecture 1: Prostate Cancer, Statistical Parameter, Sample Size Determination
2
STAT 1000Q Lecture Notes - Lecture 2: Design Of Experiments, Statistical Unit, Sampling Error
30
STAT 1000Q Lecture Notes - Lecture 30: Type I And Type Ii Errors, Null Hypothesis, Sample Size Determination
Document Summary
If is unknown, use the t-test: blind taste test: cola a vs. cola b . They are asking whether 57/100 is greater than half. What they should have said: based on the sample data, can you really infer that more than half of cola drinkers preferred cola b? well, p-value = 0. 81, insu cient evidence to conclude that. If our n = 1000 and cola b preference drinkers = 570, then the z-statistic would be tripled, and p-value = (4. 2, 1099) = 1. 3 x 10-5. The p-value would now be minuscule and it would be a no-brainer rejection of the h0. This is called increasing the power of the test. Type i error: when you reject a null hypothesis that is true. That will minimize the likelihood of a type i error: stats test menu. Independent random samples selected from two normal populations produced the sample means and standard deviations. Choose stats tests 2-samp ttest stats.
