For Data Please Refer:

str(case0502)

attach(case0502)

Spock Trial Study

The stem-and-leaf plots in Display 5.4 and box plots in Display 5.5 are useful for

suggesting some answers to the question of interest and for indicating the appropriateness

of the tools based on the standard one-way classification model. An analysis

of variance F -test confirms the strong evidence of some differences between means.

An application of the extra-sums-of-squares F -test for comparing equality of the

six other judges shows no evidence of a difference in mean percentages of women on

their venires. Assuming that the six other means are equal, a further F -test shows

overwhelming evidence that the Spock judge mean is different from the mean of

the other six. Since this is a test for equality of two means, a t -test could be used.

In fact, the F -test is equivalent to a two-sided t -test when I D 2. The actual

p-value reported in the summary of statistical findings comes from a different test,

not based on the assumption of equal means among the other six judges

Questions1 :

Explain the role of degrees of freedom of the F -distribution associated with the F -statistic.

How are degrees of freedom related to how far the F -statistic is likely to be from 1?

B....: DISPLAY 5.4 Percents of women in 30-juror venires for Boston area U.S. District Court trials, grouped according to the judge presiding Spock Trial Judge Other Boston Area U.S. District Court Judges B C D E F &UN-O 64.87 33,36.50.52,77.86 68 77.97 65 31 70,89 10.34,75.75 43,97 15.79 07.35,64.67.95,98 08.36 20,27,55 05,19,25.38.38 48 19.62 05,89 56 02 Legend: 489 represents a venire with 48.9% women.DISPLAY 5.5 Percentages of women on venires of the seven Boston area judges 50 Percentage of Women 40 30 20 - 10- Judge: Spock's B E 9