A researcher was interested in the following questions:

1) whether five different kinds of problems have the same difficulty (i.e., whether students'

performance on the different kinds of problems is the same)

2) whether students' performance on the problems is affected by the speed of the test

3) whether the difference in the difficulty among the five problems is dependent on whether

the test is speeded or not.

The dependent variable is student score on a problem while each problem is worth 16 points.

Two independent variables are speed (factor A, 2 levels, speeded and non-speeded) and problem

(factor B, 5 levels). The observed scores in each treatment combination are presented in the

following table:

B

b1 b2 b3 b4 b5

a1 2 3 6 8 11

a1 3 3 9 8 12

a1 2 6 8 9 14

a1 4 5 7 10 13

a1 6 8 10 12 15

A a2 4 5 9 12 15

a2 5 7 7 10 12

a2 7 4 8 9 14

a2 4 7 10 12 13

a2 6 6 7 11 16

Suppose these data are obtained from a two-way within-subjects design and the scores in

each treatment combination are obtained in the order of Subject 1, 2, 3, 4, and 5.

1. How many subjects are in the study?

2. What are the appropriate null and alternative hypotheses for the three appropriate

research questions? (i.e. main effects and interaction effect)

3. Enter the data into SPSS in an appropriate format. Perform appropriate analysis and fill in

the following ANOVA table.

Source SS df MS F p

A

B

AxB

(Error) S/AB

Total