this is equal to the average of the variances within each of the nine groups in the experiment.
Save this value to use in Chapter 13.
11.32 Strayer, Drews, and Couch (2006) ran a study in which they compared the driving behavior
of a control group, a group that was at the legal limit for alcohol, and a group that was talk-
ing on a cell phone. I have modified their study slightly, but the results are consistent with
theirs. The three groups are given below (the data are available on the Web site for this book
as Ex11–32.dat).
Control:
808 757 773 937 726 788 806 792 751 765 853 655 626 721 630 722 683 709
718 812 703 791 586 864 737 701 799 844 639 705 822 935 842 827 784 838
795 823 791 819
Alcohol:
631 656 621 706 937 538 947 855 661 887 750 945 663 880 873 830 751 855
815 593 666 730 1021 906 821 956 606 660 802 961 629 603 826 531 828 959 743 745 922 829
Cell Phone:
909 712 805 852 859 781 841 822 740 910 900 912 863 785 863 809 927 847 918 810 788
929 798 863 981 842 1021 827 876 736 640 851 787 703 942 758 843 781 969 872
From these data is there evidence that cell phones lead to longer reaction times than baseline
conditions? How does the cell phone condition differ from the alcohol impaired condition?
11.33 Gouzoulis-Mayfrank et al. (2000) examined task performance of users of the drug Ecstacy
and compared that with a group of Cannabis users and a control group of Nonusers. There
were 28 participants in each group, and the Ecstacy users were almost all users of Cannabis
as well. Performance was evaluated on several different tasks, but we will focus on a test of
abstract thinking. The data given below were created to have the same means and variances
as in the original study, and higher scores represent better performance. The data are avail-
able on the Web site as Ex11–33.dat.
Ecstacy: 25 25 23 32 21 28 34 26 23 22 26 21 29 28
23 24 29 23 30 18 25 25 25 25 32 23 29 32
Control: 29 31 31 25 33 21 18 40 35 32 29 31 25 32
33 34 28 28 25 22 27 34 38 31 30 31 26 30
Cannabis: 28 28 28 29 37 34 27 25 34 28 25 27 30 29 31 32
30 29 32 25 32 31 32 22 31 25 28 34
a. Run an analysis of variance comparing the means of the three groups.
b. Calculate Cohen’s dto examine the pairwise effect sizes—in other words, calculate d
on the comparison of each group with each of the other groups.
c. What is a reasonable set of conclusions from this study?Discussion Questions
11.34Some experimenters have a guilty conscience whenever they transform data. Construct a
reasoned argument why transformations are generally perfectly acceptable.
11.35 In the study by Conti and Musty (1984) on the effects of THC on activity, the means clearly
do not increase linearly with dosage. What effect, if any, should this have on any magnitude-
of-effect measure?
11.36 With four groups you could have the means equally spaced along some continuum, or you
could have three means approximately equal to each other and a fourth one different, or you
could have two means approximately equal but different from two other approximately
equal means, or some other pattern. Using very simple data that you create yourself (hold-
ing within-groups variance constant), how does the Fstatistic vary as a function of the
pattern of means?Exercises 361