**chi-squared goodness of fit test**- The observed frequencies and claimed proportions are given, find the expected frequencies.
- Determine the standardized residual for one of the categories.
- Give the number of degrees of freedom.
- Find the critical value.
- Determine if the test statistic lies in the critical region.
- Give the decision.
- Give a conclusion.
- Know how rearranging the order of the categories would affect the test statistic.

**test for independence**- Write the null hypothesis
- Find the expected frequency for one of the cells in the table
- Determine the degrees of freedom
- Find the critical value
- Give the decision.
- Give a conclusion.
- Know how rearranging the order of the rows or columns would affect the test statistic.

**one way analysis of variance**- Write the null and alternative hypothesis.
- Complete the ANOVA table
- Give a conclusion
- Find the pooled variance
- Find the variance of the response variable
- Find the standard deviation of the response variable
- Look at a probability plot and determine if the residuals appears to be normally distributed

**correlation and regression**- Complete the table of coefficients
- Give a conclusion about the y-intercept or slope
- Complete the ANOVA table
- Find the sample size
- Find the value of the coefficient of determination, r
^{2} - Find the residual standard deviation, s
_{e} - Find the center of the confidence interval for the slope
- Write the standard error of the slope (this is given, you just need to know where to look)
- The critical values are given, use them find the margin of error for
the estimate of the slope, b
_{1} - Find a confidence interval for the slope of the regression line.

**multiple regression**- Complete the table of coefficients
- Rank the predictor variables in order from best predictor to worst predictor
- Write the null and alternative hypothesis for one of the rows in the coefficient table.
- Give a conclusion based on the table of coefficient p-values.
- Decide which one variable you would keep or eliminate from the model if you had to.
- Complete the ANOVA table.
- Determine the sample size
- Write the null and alternative hypothesis for the ANOVA table.
- Find the value of R
^{2}and adjusted-R^{2}from the ANOVA table. The formula for R^{2}and instructions for finding the adjusted-R^{2}are given. - Rank the models from best to worst based on either the R
^{2}or adjusted-R^{2}values.

**two-way analysis of variance**- complete the two-way ANOVA table
- Know the three null and alternative hypotheses being tested with the table
- Give a conclusion based on results from the table.

- You will definitely want a calculator.
- You will not need Minitab. The computers will be off during the exam.
- You will be supplied a chi-squared table to use during the exam.
- Each problem is worth a lot of points, but most of them have been designed so that you can continue even if you can't completely get the first part right.
- You may create a sheet of notes to use for this test. Details about what can be on the notesheet follow.

You may use a sheet of notes on this exam. Here are the guidelines for the notesheet.

- You may use the front and back of up to an 8.5" x 11" piece of paper. You may use a smaller sized paper if you desire.
- The notesheet must be handwritten and an original (not duplicate) document.
- You may not photocopy, scan, fax, or otherwise duplicate someone else's notesheet. it is okay if you want to get together, study, and create a notesheet, but each person's notesheet must be an original, handwritten document.
- If I can not tell that your notesheet is an original document, I will not allow it. Consider using pencil or colored pens for portions of the notesheet.
- The notesheet must have your name on it. I will collect the notesheets with the exams. You will get them back with the graded test.
- The notesheet may not have examples on it.
- The notesheet may have notes, explanations, ANOVA tables with an explanation of how to find specific entries, but it may not contain examples with specific numeric values. For example, "the total df is one less than the sample size" is okay, but "if n = 20, then total df = 20 - 1 = 19" is not. You may have "SS / df = MS," but not "SS(Regression) = 351, df(Regression) = 3, MS(Regession) = 351 / 3 = 117"

# | 1 | 2 | 3 | 4 | 5 | 6 | Total |
---|---|---|---|---|---|---|---|

Pts | 13 | 10 | 13 | 22 | 25 | 17 | 100 |