Jen wants to see if people score differently on a pop quiz in the stats class she teaches depending on the time of day that the class is taken, so she tests her classes that meet at 9, and at 10, and at 11 am. She also wants to see if the day of the week the class is taken makes a difference in test scores, so she compares the Monday class to the Tuesday class, etc. Her data is shown below, in the form of the mean quiz scores (MQ )of each of the classes and the sum of squared deviations of the quiz scores in each cell from the appropriate cell mean.
9 am 10 am 11 am
|
M
|
M= 17 SS = 225 |
M= 16 SS=232 |
M= 19 SS = 224 |
|
T
|
M= 19 SS= 230 |
M= 20 SS = 227 |
M= 21 SS= 233 |
|
W |
M= 21 SS = 224 |
M= 22 SS = 220 |
M= 23 SS= 221 |
There are 10 scores in each of the cells. Use a significance level of 0.05 for all tests.
- Plot the cell means and calculate the marginal means.
a) Just by looking at the data, does it look like there is a main effect for class day? Why or why not?
b) Just by looking at the data, does it look like there is a main effect for time of day? Why or why not?
c) Just by looking at the data, does it look like there is an interaction between class day and time of day? Why or why not?
- Calculate the appropriate F statistic for the main effect of class day. Should she reject the null hypothesis that day of the week does not make a difference in test scores?
- Calculate the appropriate F statistic for the main effect of time of day. Should she reject the null hypothesis that time of day does not make a difference in test scores?
0 comments