Oneway Between Subjects Analysis of Variance
1. Does coffee help people become sober more quickly after drinking too much? A sample of 40 volunteers is randomly divided into four groups of 10 subjects each. One of these groups serves as a control and receives no alcohol. Subjects in each of the other three groups drink a fixed amount of alcohol in a 1-hour period. During the next half-hour, subjects in the second group drink two cups of decaffeinated coffee, subjects in the third group drink two cups of regular coffee, and subjects in the fourth group drink two cups of water. Finally, all subjects are given a reaction-time test to determine mental alertness with the lower the score on the test, the quicker the reaction time and thus greater alertness. The following data were gathered:
CONTROL
DECAF
REGULAR
WATER
184
221
189
196
197
218
219
201
189
226
206
215
207
230
214
198
179
205
221
229
215
216
192
228
201
195
215
209
193
216
199
192
188
191
231
231
196
205
225
216
Complete the ANOVA using an α = .05 for all statistical tests and answer the following questions:
a) What is the purpose of the random assignment of subjects to groups?
) Do we meet the assumption of homogeneity of variance? What specific information on the printout did you use to come to this conclusion?
c) Is the average reaction time the same for all four groups? What specific information on the printout did you use to come to your conclusion?
d) What proportion of variance in reaction times is associated with differences in the methods used in attempting to sober up the subjects?
e) Use the Tukey procedure to determine which pairs of groups differ in reaction time. Calculate Cohen’s d effect sizes for those pairwise differences found to be significant.
f) Use the Bonfe
oni procedure to determine which pairs of groups differ in reaction time. Do these results agree or disagree with the results of the Tukey test?
g) To whom, if anyone, can we generalize our findings?
h) Write a
ief Results section describing the findings of this experiment based on the analyses you have completed. You do NOT have to present results in tables.