To obtain instructions for doing regression, correlation, and scatterplots
on Excel,  Click Here.

1.     You will need the Footprnt.XLS data set.

2.     Use the chart wizard to develop a scatterplot of Height, the height of a student,
        and Reported Foot Size, the size of the student's right foot (in inches).

3.     Does there appear to be a linear relationship between Reported Foot Size and Height?  If so,
        is it a positive or negative slope? What does your answer tell us about the relationship
        between foot size and height? Is it what you expected?     Explain.

4.     Sort the data by male and female data points (Gender = 1 for male and 0 for female) .
        Then construct a scatterplot of Reported Foot Size and Height for females, then for males.
        Describe any differences you expect in the two relationships.
        SORT is under the Data button on the taskbar.  Be sure that you bring foot size, height,
        and the other variables  along as you sort the data by male and female.  A window will
        ask for this information.

5.     Run regression models for Height = f(Reported Foot Size) for all students, for females,
        and for males. Interpret the coefficients in each case. Do the results appear to verify your
        expectations from Question 4.     Explain.

6.     A crime witness stated that she was sure that the perpetrator was a student in my statistics
        class. The footprint at the scene of the crime is 13.25 inches long.  Predict the height of the
        criminal.

7.     After further reflection the witness is sure that the perpetrator was a male.  Would you predict
        a different height now?

8.     Repeat #5 three times (for the combined data set then for males and finally for females) with
        the following changes.  First use Shoe Size, and, then use Shoe Length as the
        independent variable instead of footsize.  Suppose the print at the scene of the crime was
        actually a shoeprint rather than a footprint. How would you change your prediction?   Explain.

9.    Write down the 9 equations that you have estimated and bring this list along with your regression
       printouts to class.

Idea for this worksheet from Eugene Giles and Paul H. Vallandigham, Height Estimation from Foot and Shoeprint Length, Journal of Forensic Sciences, Vol. 36, No. 4, July, 1991, pp. 1138-1139.

Return to Worksheet Menu
Return to Statistics Home Page
Return to Jarrell Home Page