Revised 1/16/01
Permanent Residences (PRs)
Frequency Distributions and Histograms on the Computer
Worksheet 3

1. You will need a data set to proceed.  Once you click on the data set name which follows, a
     box appears.  It asks whether to open or save the file, I suggest you save it (probably on
     your 3.5 floppy).  You must give it a name also.  Remember the name and location where
     you save it, so you can open it in Excel and proceed with the worksheet.

    Click on Permanent Residence Data.  After you save the data, open it in Excel and proceed.

    PRD is the approximate distance (in miles) between campus and a student's permanent
    residence.

2. Use bins (upper class limits) for classes of size 50 miles in any column except A to
    construct a frequency distribution on another worksheet in the same workbook.  Click
    Tools, then DATA ANALYSIS, then Histogram. (If DATA ANALYSIS does not appear,
    click Add-Ins... and check Analysis Tool-Pak.  Click OK.  Then DATA ANALYSIS should
    appear on the TOOL menu.)  Indicate the position of thedata set you wish to describe in
    the box beside Input Range.  For instance, if the data is in cells A2 to A38, then type
    A2:A38 in the box.  Indicate the range of cells that contain the bins in the opening beside
    Bin Range. If there is a label in A2 rather than a value, be sure to check the box beside
    Labels.  (Note:  you must include a label for the bins as well if you include a label for the
    data.)  Under Output Options check NEW WORKSHEET PLY.  You do not need to enter
    a name at this point.  Be sure to check CHART OUTPUT in the histogram window to
    obtain the corresponding histogram of the data.

3. Describe the pattern of the data in the results from Question 2.

4. Use bins of size 100 miles to construct another frequency distribution.  Include the
    histogram that results from thisfrequency distribution. Is the pattern of the data different
    now that in Questions 3?  Explain.

5. Suppose the University decides to have a midweek one-day vacation.  The Housing
    Director would like to be efficient and save money by closing one or more residence
    halls and consolidating the students who remain on campus.  (Let's assume that it's
    relatively simple and costless to close dorms and consolidate residents in different
    locations.)  He assumes that students who live within 100 miles will go home and return
    in time for classes the next day.  Suppose the PRD data from your class is "representative"
    (replicates the features, such as distribution, center, and spread of the population) of
    the entire WCU student body. Make a recommendation to the housing director. Explain
    your recommendation and base it on your earlier work for this worksheet or create a
    new histogram to backup your recommendation.  Make and state any other information
    that would help you make the recommendation and any necessary assumptions you
    must make.  Would you argue with the Director's assumption about distances for
    single-day return trips?

6. Again assume the PRD data is representative of the population of PRD values for all
    WCU students.  Suppose also that a local business wants to communicate with as many
    students as possible during the week before classes start.  The business has a budget that
    will afford radio and television commercials for an audience within 200 miles.  What is the
    maximum percent of the students the business can expect to reach?

7. Use Excel to determine the mean, median, and mode of the PRD data. For
    instance, in separate cells type=average(A2:a38), =median(A2:a38), and
    =mode(A2:a38). These  examples assume the data set is in the range of cells from
    A2 to A38. Do the relative magnitudes of these three values seem reasonable for
    the pattern you described in Questions 3 and 4?

8. Try different bin sizes to see if you can change the pattern of the data. You will not
    always be able to accomplish this. Is it better that it is difficult to change the pattern
    or would we want the pattern to be very sensitive to bin sizes? If you were able to
    change the pattern, describe the new pattern.

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