Learn by Doing: Side by Side Boxplots

The objectives of this activity are to:

• Teach you how to use technology to produce side-by-side boxplots and the relevant descriptive statistics,
• Let you practice comparing and contrasting distributions, and
• Help you gain more intuition about variability through the interpretation of your results in context.

The percentage of each entering Freshman class that graduated on time was recorded for each of six colleges at a major university over a period of several years. (Source: This data is distributed with the software package, Data Desk. (1993). Ithaca, NY: Data Description, Inc., and appears in http://lib.stat.cmu.edu/DASL/)

In order to compare the graduation rates among the different colleges, we will create side-by-side boxplots (graduation rate by college), and supplement the graph with numerical measures. Follow the instructions, and then answer the questions based on the output you got.

1. Click https://s3.amazonaws.com/wamapdata/ufiles/1958/graduation.txt to see the raw data of the graduation rates. Notice that there are two columns. The first column (labeled COLLEGE) shows the college (college A represents the first college, etc.) and second column represents the graduation rates for a given year.

2. Go to this website: lock5stat (this website makes histograms and boxplots for multiple sets of data, such as this one)

3. click EDIT DATA

4. press CTRL+A to select the data that is already there, then press DELETE to delete it

5. copy and paste the college data (both columns)

6. if you pasted the column headings, then make sure the box for "data has header row" is checked.

7. click OK

8. click the BOXPLOT tab, and see all the boxplots, side by side.

Now Answer the following questions:

1.Compare and contrast the distributions of the graduation rates at the different colleges. Be sure to address center, spread and outliers.

2.If you had to choose one college among the six colleges based on this data, which college would it be? Explain your reasoning.

3.If you were debating between colleges B and F only, which one would you choose based on this data? Explain your reasoning.