Study guide final

Nine questions, three hours Print all tables (z, t, F, and chi-square) 1. Calculating confidence intervals using Z and "t" statistics for proportions and populations ( M3) ALWAYS FIRST CHECK IF N IS >= 30. IF IT IS, DO THE Z TEST, OTHERWISE START DOING THE T TEST. - Use critical t when: o We have the sample mean and sample standard deviation (s) o If n < 30 o We do not know the population variance o Read the table: df = n-1 - Use z test when: o We are told the population standard deviation (sigma) o If n >= 30 o Use z test for proportion   For binomial, once you know the mean, you know the standard deviation - Population test: 1. Use the given formula Z value: use the table 2. Similar concept for t-test - Proportion test: 1. Use the given formula - If given error (x- mu), alpha level, and sigma, then you can find the sample size n

o - Refer to m3 as example for determining the CI for both proportions and populations 2. Conduct hypothesis tests for equalities and inequalities for proportions and populations (only one-parameter, there will be NO questions on two populations) - Equality (>, <) - Inequality (=/): two-sided hypothesis test - - - Still do the same method for finding the confidence interval that is around the true mean - To see if you reject the H0 (null hypothesis), find z. if z is greater than the alpha level z value, then reject null hypothesis

- - 3. Calculate p-values for one and two-tail tests - Use excel - If two-tailed, need to double the p-value - If p value is less than the alpha, then reject the null hypothesis - If p value is greater than alpha, then fail to reject the null hypothesis - To find the p value o Once you find the z value, then find the respective area (go to the z table) and use z to find the area. That is the p value. o If it is two sided, then multiple the area by * 2 = p value 4. Complete and interpret an ANOVA table. At least two of three "sums of squares" will be provided. You will NOT have to calculate any values from raw data. - - Formulas: