Example Question
Given an ANOVA printout, state which groups have an appreciable difference, and which groups do not have a statistically significant difference.
A question like Example 1 on Lecture 18
1. Tell me any 3 different statistical tests, the technical conditions that must be satisfied, when they would be used (with qualitative or quantitative data or both), and a sample null and alternative hypothesis.
Answer TRUE or FALSE to the following statements: 1) The coefficent of determination (r squared) indicated the percentage of variability in the explanatory variable that is explained by the least-squares line with the response variable. (FALSE)  2) If the Pearson correlation coefficient has an r<0 than the association is positive. (FALSE) 3) Pertaining to a least squares regression line, if you remove a value and it changes the regression equation, than this value is an outlier. (FALSE)
Any question that lays out the complete process of using a two-way ANOVA.
What is needed to conclude a cause and effect relationship?
For the following variables in a regression analysis, which variable more naturally plays the role of X (explanatory variable) and which plays the role of Y (response variable) a) College grade point average (GPA) and high school GPA. b) Number of children and mother's education level. c) Annual income and number of years of education. d) Annual income and assessed value of home.
What is the most appropriate test of significance to use when dealing with one quantitative and one qualitative (multiple responses, not binary) variable?  Justify your answer.
300 males were interviewed on the streets of San Francisco about exercise. 300 females were also surveyed about how much exercise they get each week.  The resuls were broken down by gender and race.  The researchers are interested in seeing if there is a relationship between how long a person works out and gender and race. Female: African American (120) exercise 6 hours a week, Anglo-American (100) exercise 8 hous a week, Mexican-American (80) exercise 5 hours a week. Male: African American (100) exercise 8 hours a week, Anglo-American (120) exercise 4 hous a week, Mexican-American (80) exercise 6 hours a week. a) Enter the above information into a table. b) Define the explanatory variables and the response variable.  c) Which inferential method would you use to analyze the above data? d) If possible, conduct the test of significance, including all steps in the procedure. e) What conclusions can you draw from this test
A chi square test similar to the judge example with Dr. Spock.
You have a sample of racers and you have them compete in three different racing conditions (short intermittent sprints, moderate pace, and a continous pace).  You record the time and preference (qualitative).  What type of statistical analysis would you conduct to see if one condiditon was different from the other?
What would an r^2 value of .89 tell us. What if the number was .23? Where would you see an r^2 value, and why is it used?
What is the definition of residual?
When is a two-way ANOVA used?
A simple regression problem where the technical conditions are violated and we have to transform the data in minitab. Similar to the housing prices example but maybe more complicated, like having to predict the response from the explanatory using the regression line equation.
Interpret and use least-squares regression line.
Perhaps have a question with Minitab output for an incorrect test or procedure.  Ask students to identify why the test performed is incorrect, and ask them to carry out the test properly.
Explain what an R-squared value is.
I absolutely loved the multiple choice independent/dependent practice problem, it gets your mind working.  Given a minitab or SAS output of a regressesion analysis, identify all the important variables and construct a CI.
Something similar to question #3 in homework #8.
something similar to question #2 on HW #8 where we need to remove an observation to see how it affects the regression line.
Explain what each of the following mean: "r", "R-sq" and "R-sq adj"
Since we have only three hours for the final exam, I would recommend very clear questions that we don't spend a lot of time thinking what test can we apply. I really like working on regression when we have at least two quantitative variables.
You will probably ask us to create a "four in one" graph for regression and then check technical conditions.
What is the difference between a sample distribution and a sampling distribution?
A study is trying to figure if there is a relationship between the amount of tv watched per day (in hours) and weight (in pounds). The data can be found in minitab under tvweight.mtw.  Carry out a test of significance to determine whether the relationship between amount of tv and weight is statistically significant. Make sure to include both numerical and graphical summaries in your analysis. Also make sure to check all technical conditions before proceeding.</P>
How do different factors affect the size of the p-value?
Explain the reasoning for why we prefer to use t distributions for quantitative response variable significance tests and confidence intervals over z distributions.
Actually, if you had a question similar to the last PP questions, those were pretty difficult.
Which kind of test do you use for one qualitative and one quantitative variable to test the level of signifigance? what are the technical conditions?
Explain the fundamental differences b/n chi-square, regression and ANOVA procedures.
Based on this least square regression equation is the slope the same based on the different age predictions (show your work)? 14age: 15.02 + 2.750(age) =sodas drank(hat); 15age:15.02 + 2.751(age) = sodas drank (hat).  True or false the slope is the same show the difference?
something about regression, I love that stuff!
What are the technicial conditions for an Anova and a Chi Square?
possible question:  please explain what the R squared value for this regression equation is, why it is important and what it represents.
A study asked "do you prefer coke or pepsi?" This quarter said 75 coke, 54 pepsi.  Last quarter 100 said coke, out of 162. Construct a 99% confidence interval for the difference in proportions between each quarter and interpret the results.
Utilizing the Diet data file in Mini-tab give the 1. the regression equation for the regression line and the predicted weight loss at month number 8?
One can interpret r = .3 as follows: (chapter 9, problem 35)  a) A 30% reduction in error occurs in using X to predict Y.  b) A 9% reduction in error occurs in using X to predict Y comparing to using Y bar to predict Y.  c) 9% if the time Y hat = y  d) Y changes .3 units standard deviations when Y changes one standard deviation. 2) Give an example of a situation when two-way ANOVA would be appropriate, instead of the one way ANOVA. 
Given one quantitative variable and one qualitative variable which test of significance would one run?
name the different test procedures and what type of variables belong with each.
What method of testing would be required for the following statistical information to figure out if it is statistically significant?
Conduct a Chi-Square test for the following two-way table, x, and interpret the results.