COM3710 - Section 1 - Homework 2




riosjosh

Comm 3710 Spring 2011

Assignment 2 (10 points, Due 09-20)
This homework is designed to increase your skill in visually presenting data. As a result, your work will be graded based on accurate completion of the problems and also on the visual appeal of your answers. All of the homework in this set must neatly typed and some of the problems require you to produces charts and graphs using Microsoft Excel or another spreadsheet program. Please insert your charts directly into your homework manuscript. Your homework manuscript must include a cover page and have page numbers). All of the data is available in the “homework two dataset Fall 2011” posted on webct.
For the first three problems, we will work with the variable age “Age” in the “homework two dataset fall 2011” posted on webct. The data represents the ages of award winning professors at the time they won the award at the University of Utah. Start binning the data at 30 and use 10 point bin ranges (30-39, 40-49, and so on).
  1. Bin the data using the criteria stated above and construct a frequency table for the distribution of age scores. Your frequency table should have a title, the standard four columns corresponding to age bins, frequency, relative frequency, and cumulative frequency. Your table should also have several rows. The first row indicates the title of each column, the rest of the rows indicate the age bins, and a row for the total of each column.

|||||||| University of Utah Awards by Age
||
Age
Frequency
Relative Frequency
Cumulative Frequency
30-39
12
35.29%
35.29%
40-49
13
38.24%
73.53%
50-59
5
14.71%
88.24%
60-69
3
8.82%
97.06%
70-79
1
2.94%
100.00%
Total
34
100.00%



A. Do the professors appear to be younger, older, or neither?
They appear to be much younger at the time of their award. ~74% of them received the award before the age of 50.
B. Which bin had the highest frequency?
40-49

C. Which bin had the lowest frequency?
70-79

D. What percentage of respondents are 49 or younger?
73.53%
  1. Create a bar chart using Excel that displays the frequency for each age group. Labeling is extremely important. Be sure to include the five labels noted on p. 110 of the Bennet et al. (2001) reading.


A. Explain whether or not a dot plot would be an appropriate display of this data.
It would not be because you would not be able to see as clearly, compared to the bar chart, the difference between the age bins of 30-39 and 40-49.
B. Explain whether or not a Pareto chart would be an appropriate display of this data.
A Pareto chart would not be appropriate for this data. This is because the more important data to show is the age groups in order rather than the frequencies in order. There is too much significance in the ages or the genres.

  1. Use your word processor to create a stem-and-leaf plot of respondent ID and AGE. Use the categories created in problem one for the stem and use the ID numbers for the leaves. Be sure to include all of the appropriate labels.

|||| University of Utah Awards by Age with IDs
||
Ages
ID #s
30-39
1, 3, 5, 9, 10, 13, 16, 18, 24, 25, 30, 32
40-49
2, 4, 8, 11, 15, 19, 20, 22, 23, 26, 28, 31, 34
50-59
6, 7, 12, 17, 21
60-69
27, 29, 33
70-79
14


  1. Under what circumstance(s) would you choose a stem-and-leaf plot to display this data instead of a histogram?
If the data itself was significant, and we wanted to see the individual values then a steam and leaf plot is better than a histogram. For example in this data set if we wanted to see the ID numbers then a steam and leaf plot is more appropriate.
  1. What is the difference between a histogram and a bar chart?
In a bar chart the bars height shows the number of frequencies versus in a histogram the bars width show the number of frequencies. So in a bar chart the bar’s height has significance and meaning where as in a histogram a bar’s width holds these. In a bar chart the width has to be the same for all bars, and in a histogram the height has to be the same.

  1. Construct a pie chart of the “Genre” and “Percent” data in the “homework two dataset fall 2011” posted on webct. The data represents a reading survey given to 1,000 communication majors at the University of Utah.
    1. What labels did you include?
Chart title, data percentages, and category
  1. Why?
The chart title tells the viewer what the significance of the data is. Categories show what the meaning of each piece of the pie is. The percentages give relevance to each category.
  1. How many students said they were reading Nonfiction?
6% of the students read Nonfiction
  1. For this problem, use Excel to construct a time series diagram of the average class size at the University of Utah from 1990 to 1999. The X axis should be the data for “Year” and the Y axis should be the “Class Size” data in the “homework two dataset fall 2011” posted on webct. Be sure to include all of the appropriate labels.



  1. Is there a clear trend in the data? If so what is your interpretation of the trend?
Over the entirety of the data there is no real clear trend. We might say that this chart is slightly bimodal; however, the second hump isn’t quite significant enough. We can say there is a clear increase in class size between ’90 and ’92 and again between ’96 and ’97. With one significant decrease between ’93 and ’94.