Class Data II

The exercise uses data collected from 180 students enrolled in STAT 100, a beginning statistics course at the University of Illinois. The objective of this exercise is to compare numerical summaries for men and women on various measured characteristics such as height and weight. Click on Data under Chapter 1 in the left panel and then select Open remote ... under the File menu in the right panel. Select class96.xml from the Open File Dialog and click OK. Note that Height, Weight, and DogsCats are assigned the Y role and Gender the Z role by default. Change the role of DogsCats to None. Select y|z from the Graph menu.

  1. Click on the Height histogram to bring it to the front. Look at the two histograms for Height for men and women (by clicking on the conditioning variable Gender). Do they seem to have similar spreads, or is one much more spread out than the other? Discuss.
  2. Find the difference in the mean Height between men and women.
  3. What are the standard deviations of men's and women's heights? Are they close?
  4. Select Quantile Plot from the Plot menu. Find the difference in median Height between men and women. Is the median difference about the same as the mean difference in Question 2?
  5. Click on the Weight histogram to bring it to the front. Look at the two histograms for Weight for men and women. Do they seem to have similar spreads, or is one much more spread out than the other? What are the standard deviations of the men's and women's weights? Are they close? Which is more spread out, the men's weights or the women's weights? Why"
  6. Find the difference in mean Weight between men and women.
  7. Select Quantile Plot from the Plot menu. Find the difference in the median Weight between men and women. Why is the mean difference larger than the median difference?
  8. Change the variable roles of Gender, Height, and Weight to None. Close all the histograms. Display the histogram for DogsCats. Which is larger, the mean or the median number of dogs and cats owned? Why? Hint: Make the histogram display larger so that the outlying values are visible.