In a study of problem solving, subjects were randomly assigned to two different treatments, A 1 and A 2 , and a control group, C. At the conclusion of the experiment, the subjects were given a set of...

In a study of problem solving, subjects were randomly assigned to two different treatments, A₁
and A₂
, and a control group, C. At the conclusion of the experiment, the subjects were given a set of problems to solve. The problem-solving scores for the three groups were as follows:

Using dummy coding, do a multiple regression analysis in which the problem-solving scores are regressed on the coded vectors. I suggest that you do the calculations by hand as well as by a computer program. Calculate the following:

(a) R²

(b) Regression sum of squares.

(c) Residual sum of squares.

(d) The regression equation.

(e) The overall F ratio.

(f) t ratios for the test of the difference of each treatment mean from the control mean.

(g) What table would you use to check whether the t's obtained in

(f) are statistically significant?

(h) Interpret the results.