Name_______________________________
Lab #1: Intro to StatCrunch-Graphs and Descriptive Statistics
STAT 202 Miller
Thursday August 30, 2012
Please review the file under “A Brief Guide to StatCrunch and Lab Submissions” (posted on Blackboard) while completing this lab. You will use StatCrunch to answer the following questions.
Your lab should be submitted as a Word document. Please copy and paste any graphs that you create into your lab document. Any responses must be easy to read and labeled with the appropriate problem number. Please save your solutions to your g:/ drive (or your own computer) and submit the lab via Blackboard.
Background
: A common psychological experiment tests numerical memory-a subjects ability to repeat back a string of numbers. For this particular experiment two trials were conducted with each of 52 subjects; number strings were delivered both verbally and visually. The data was collected from 4 universities and online between July 2010 and December 2010. Along with detailed information pertaining to the experiment (such as the place, date and time of the experiment, and type of delivery), the experimenters collected information about the subject (age and gender) and recorded the longest number of digits that the subject was able to repeat back to them (Limit #).
1. Begin by loading the data into StatCrunch. Use the data file that I have provided (numericalmemory.txt) and follow the directions provided in “Intro to StatCrunch and Labs.”
Make sure that you select “comma” as your delimiter.
Once the file is loaded save your data to your G drive so that you will be able to access it again in the future.
2. Classify the variables “Affiliation”, “Gender”, “Age”, “Trial Num”, and “Limit#” as categorical or quantitative.
3. To get an idea of where the subjects were tested make a pie chart for “Affiliation.” (Graphics-Pie Chart-with data, Select the “Affiliation” column and click next to look at other options and give the graph a title before clicking “Create Graph!”.)
4. To take a look at the distribution of the genders of the subjects tested make a frequency table (Stat-Tables-Frequency) and bar graph (Graphics-Bar Plot-with data) for the ”Gender” data. Note that each subject is listed twice so the frequencies listed are twice as large as the true values but the relative frequencies are accurate.
5. Make a stem plot of the “Age” data (Graphics-Stem and Leaf). Describe the overall pattern of the distribution of “Age.”
6. Now make two histograms (with different bin (interval) widths) of the “Limit#” data (Graphics-Histogram). You can designate a starting point and bin width or StatCrunch will automatically select these for you. After selecting the “Limit#” column click “next” to set your options.
7. How does the bin width affect the shape (skew, symmetry, etc.) of the distribution?
8. Make a box plot for the “Limit#” data (Graphics-Boxplot). You have the option to orient the plot vertically or horizontally: Click on “Next” when in the Boxplot menu and select (or do not choose to select) “Draw boxes horizontally.”
9. Find the mean, standard deviation, variance, median, and quartiles for the Limit# data (Stat-summary stats-column stats). Select the “Limit#” column and then click “next.” Click on the summary stats name to deselect it. Copy and paste your results.
10. Based on all of your graphs and calculations for the “Limit#” data, what can you conclude about the overall pattern/shape of the distribution of “Limit#”? Is it symmetric, skewed or neither? (I suggest comparing the two measures of center as well as using your calculations of spread and how they relate to the center to support your final conclusion here.)
11. Now I want you to look at the distributions for “Limit#” separately for the Males and for the Females (using the default binwidth). To do this choose to Group by “Gender” (there is a pull-down menu on the first screen) and check the box next to “Separate Graph for Each Group.” Compare the two distributions.