*THIS PROBLEM IS TO BE DONE IN THE GROUPS THAT YOU CHOOSE.*

*THIS PROBLEM IS TO BE DONE ON THE COMPUTER. PLEASE ATTACH YOUR COMPUTER
PRINTOUT TO YOUR GROUP REPORT. YOU NEED TO USE EXCEL.*

Take the data from the class that was obtained on 3/28 and 3/30. If you want the data on disk see Dr. Sherren. Dr. Sherren plans to e-mail you the data and also to post it on the 210 Web Site. This project should be several hours.

Group 1 should use the data from Color 1, Group 2 the data from Color 2, Group 3 should use the data from Color 3, Group 4 from color 4.

1. Using a bar graph plot all of the data. Use % color for the y-axis and the initials for the x- axis.

2. Determine the mean and standard deviation for your data for both sets. Determine the confidence limits at the 90% level for your data. Sketch your confidence limits on the graph.

3. Reject any data that falls outside the CL (90%).

4. Recalculate the mean and standard deviation for both sets of data. Should you change your confidence limits. If so calculate these limits.

5. Calculate the combined mean and standard deviation for the good data used in step 4.

6. Do your confidence limits still hold? reject any outliers. After the rejection calculate the new mean and standard deviation. Would Q-test have let you reject any of the data in step 4.

7. Test the two sets of data to see if they came from one or two populations. Use the t-test on Excel. If Step 6 left too few points use Step 5 Data.

8. Compare your means from step 5 with the following true values.

Red - 20%, Green- 10%, Yellow 20%, blue 10%, Brown 30%, orange 10%. For regular colors.

Pink - 25%, Blue - 25%, Green - 25%, Yellow - 25% Easter

Orange - 50%, Black 50% Halloween

Red - 50%, Green - 50% Christmas

Red - %, White - % Pink Valentine

Problem is due on Friday, April 6. Preferably by noon. Definitely by 4:30 PM.

ATS 2/17/01