Question

# a. a researcher wishes to test the differences among five poplulaiton means, using .05 criterion of...

a. a researcher wishes to test the differences among five poplulaiton means, using .05 criterion of significance. Why is it a bad idea to perform 10 t tests for the 10 possible pairs of means?

b. in one-way analysis of variance, what can cause the group means to vary? how about The degrees of freedom for the numerator is equal to?

(a)
It is bad idea to perform 10 t tests for the 10 possible pairs of means because the repetition of multiple tests repeatedly add multiple chances of error, which result in a larger level than the pre -set level, The comparisons are repeated 10 times in an experiment and an unacceptable increased total error rate of 1 - (1 - )10 is expected for the total comparison procedure in the experiment.

(b) In one way Analysis of Variance, the difference in the relative size of variance among group means (between group variance) with the average variance within group )within group variance) cause the group means to vary.

(c) The degrees for the numerator = k - 1,

where k is the number of treatments.