¶ … Statistics

Consider this scenario: You are on the Board of Directors of the National Kidney Foundation. Two different research labs are working on treatments. Your responsibility is to review and evaluate both of the results from the research studies and make recommendations for future funding. You can only fund one of the two. Patients and their families are anxiously waiting for the Board of Directors to make a decision which of the two can be funded.

Your first task it to do the following calculations for both of the results that have been submitted to you.

Mean

Median

Range

Standard Deviation (are there outliers you should consider?)

Number of weeks that patients felt better:

Group I results = 3, 5, 6, 6, 8, 9, 9, 9, 10, 11,

Group II results = 3, 4, 7, 9, 9, 9, 15, 15, 16, 17,

Part II.

After you have done this comparison, write a letter to the rest of the members of the board stating which group you are recommending to receive additional funding to support their research. This should be done in Memo Format and you need to discuss your findings and compare them. You can include a table to show the comparison if you want to. Do not be concerned with the small number used in each group in the study.

TO: National Kidney Foundation, Board of Directors

FROM: Jon Doe, Staff Statistician

DATE: June 23, 2013

SUBJECT: Future Research Funding Recommendations

Upon examination of the data submitted by the two research groups using a graphical view (see below), it became obvious that Group I could have an outlier. Using the maximum value Grubbs test, it was determined that this value was indeed an outlier and was excluded from the data prior to further statistical analysis.

Once the outlier was removed from the data it became clear that mean of the Group II data (11) was higher than the mean of the Group I data (see table below), suggesting better results with the Group II treatment; however, the magnitude of the standard deviations indicated the possibility that the two means were not significantly different. Using a one-tailed students t test, the means for the two data sets did reach statistical significance at the 95% confidence level (p = 0.038). Based on these results, I'm recommending that continued funding be approved for research Group II only.

Statistic

Group I

Group II

Mean

7.6

11

Median

8.5

9

Mode

9

9

Range

8

14

Std. Dev.