Homework Assignment #6

All homework assignments must typed or legibly handwritten, and must show your name, the assignment number, course number, and the date clearly.  Show all your calculations and other work.

Due: Monday, April 2, at the beginning of class

Part A

Do chapter 8 exercises #2, 6, and 10, using WebStat for the calculations and the plotting.  Make sure that you give an description of the strength, direction, and other features (e.g., nonlinearity, outliers, heteroschedasticity, etc.) of each relationship you examine, even where the exercises in the text don't specifically ask you to.  You can enter the data for each exercise directly into WebStat (if you need a refresher on how to enter data, access the help page [via the "Help" menu] and consult the Data section on "Using the data table."

Run a simple linear regression analysis by going to Stat ---> Simple linear regression.  In the box that appears, click on the independent variable for the x variable and the dependent variable for the y variable and hit "Next."  In the next box, put a check for "plot the fitted line" and hit "Calculate."  The numerical regression results will appear in the bottom half of the WebStat window and a separate WebStat graphics window will open up with the scatterplot.

Part B

Do chapter 8 exercise #7 (parts c and d only) in SDA.  Instead of using the very limited data the text provides for these two variables, use the EDUC and MEMNUM variables in the GSS data set available at SDA.  For part c of #7, compute the errors of prediction for those persons based on your SDA analysis.  Ordinarily, what would you want to do before running such a regression analysis to determine whether linear regression is an appropriate technique to use?