CSC 239: Personal Computing for Science

Final Project (Due Friday November 18, 2005)

Assignment

Working with your group, go to the DASL homepage and choose a data set:
  1. Click on "All Methods"
  2. Choose from either the "Correlation" or "Regression" link
  3. Choose at least two variables that might have an interesting relationship
Conduct the following analysis for each chosen variable of your data set: (15 points)
  1. Graph the histogram
  2. Indicate whether the histogram is symmetric or skewed
  3. Report the appropriate center and spread
Conduct the following analysis for a single variable of your data set: (15 points)
  1. Use the normal approximation to calculate the bottom 5% and the top 10% of the data.
  2. Compute three confidence intervals (90%, 95% and 99%) for the population mean
  3. Test the significance of the claim that the population average is less than the median value of the data (link to hypothesis-testing calculator). Assume alpha equals 0.05. Report your results.
Conduct the following bivariate analysis of your data set: (25 points)
  1. Draw a scatter plot of your data.
  2. Calculate the regression model for your data.
  3. Determine if there are any outliers in your data. If there are, remove them from the data and reapeat steps 1 and 2 above.

All groups will present their projects during the final exam time (2:45 - 5:00) on Friday November 18, 2005.

Guidelines:

Each group will create
  1. a spreadsheet showing the data analysis
  2. a short report summarizing the results of your analysis.
  3. a web page presenting the results of your analysis.

The web page should contain a non-technical summary of your project with a description of the results of your analysis. It should be intelligible to a person who does not know regression analysis. Suppose you are talking to your boss who does not know statistics or to a friend who is not familiar with statistical terminology. Finally, provide some plausible explanation that involves the story behind the data set and is consistent with your results.

The web page will have links to both the data analysis spreadsheet and the short report. Also, if there are external web pages relevant to your project topic the please add links.

There are no rules on the way the page should look. You can use your esthetical skills to add colors and pictures to the page. Be careful not to overdo with colors, though!

Technical Tips:

General grading scheme:

Group grade: Individual grade:

Deliverables

One student will post the web page on his or her shrike account. One group member will send their final projects as three attachments (the web page, the report and the spreadsheet) in an email to cmiller@cti.depaul.edu. Also indicate the URL of the Web page. Each student will submit an individual report using the online submission Web site.