Factorial Analysis

14. In this problem, you will use one-way ANOVA to compare means from three unrelated samples. The data in tab 14 represent the scores received by three groups of women on a 20-point knowledge assessment of breast cancer risk following a visit to a breast cancer screening clinic. Women in the “Face to face” group received a brief discussion on breast cancer risk from a physician, those in the “Pamphlet” group received a written pamphlet, and those in the “Nothing” group received no specific information about breast cancer risk. (The women were randomly selected and randomly assigned to one of the three groups. All women received a face to face encounter with a physician following the experiment.)

a. Using the appropriate tool from the Data Analysis Add-Ins, perform an analysis of variance to determine whether there were any differences in knowledge among these groups. (5 points)

b. Perform the appropriate test(s) to identify where the differences between groups, if any, were significant. Remember to adjust the ‘p’ value to account for the multiple comparisons computed. (5 points)

15. his problem, you will examine and practice using Factorial Analysis of Variance, which is a tool for comparing means when the means are grouped by two or more categorical variables (factors). The data in tab 15 represent the in-hospital length of stay (LOS) for 36 patients. These patients were discharged from one of three hospitals and had one of three different diagnoses. These data include four patients for each of three diagnoses at each of three hospitals. The data are grouped by hospital and diagnosis.

a. Using the appropriate tool from the Data Analysis Add-Ins, perform an analysis of variance to determine whether there were any differences in LOS among these patients. In your report, include a table displaying the mean LOS for 1) each hospital, 2) each diagnosis and 3) for each diagnosis at each hospital (a total of 15 means) plus a copy of the ANOVA table from Excel. (5 points)

b. Provide an interpretation of the ANOVA table. That is, report whether LOS varied significantly by diagnosis. Report whether LOS varied significantly by Hospital. Was there significant statistical interaction between diagnosis and hospital in these data? (5 points)