Suppose that you want to compare the mean size of health care claims submitted by five groups of policyholders. Ten claims are randomly selected from amongst the existing claims for each of the five groups. The data is shown in the following table:

Group 1 ($) Group 2 ($) Group 3 ($) Group 4 ($) Group 5 ($)

763 1,335 596 3,742 1,632

4,365 1,262 1,448 1,833 5,078

2,144 217 1,183 375 3,010

1,998 4,100 3,200 2,010 671

5,412 2,948 630 743 2,145

957 3,210 942 867 4,063

1,286 867 1,285 1,233 1,232

311 3,744 128 1,072 1,456

863 1,635 844 3,105 2,735

1,499 643 1,683 1,767 767

Assume that the data violates the assumptions of linearity, robustness, and normality. Do the data contained in the five samples provide sufficient evidence to indicate a difference in the mean levels of insurance claims among the five healthcare plans? Use the 8-step of hypothesis testing sheet.

I am assuming that I would first have to do a statistical analysis to get the mean and then I am thinking using an ANOVA test as well.