Chapter 17—Factorial Analysis of Variance
MULTIPLE CHOICE QUESTIONS
17.1+ A factorial analysis of variance has
17.2 A 2 4 factorial has
17.3 The difference between a one-way analysis of variance and a factorial analysis of
variance is
17.4 If the analysis of variance is significant, we are pretty sure that
Test Bank
Use the following research scenario to answer the next four questions:
A researcher was interested in the effects of 1) alcohol consumption and 2) content of a
videotape, on how likely one is to support rape myths. The researcher randomly assigned
60 college-aged males to one of the following three groups: no alcohol consumed, a
moderated amount of alcohol consumed, and a large amount of alcohol consumed.
Additionally, half of the participants were shown the educational video on rape myths.
The other half of the participants watched a documentary on owls (a control condition).
At the end of the study all the participants filled out a survey on rape myth acceptance.
Higher scores on the survey indicated higher acceptance of rape myths.
17.5 The results indicated that the participants who watched the educational video
scored significantly lower on the rape myth scale compared to the group that
watched the owl video. What does this suggest?
17.6 How many cells does this experiment have?
17.7+ What type of statistical analysis would be most appropriate for this experiment?
17.8 What would you suggest if the researcher found that alcohol consumption
increased rape myth acceptance, but only when the participants had watched the
owl video?
Chapter 17
Use the following ANOVA summary table to answer the following four questions:
Source
df
SS
MS
F
Gender
1
240.25
240.250
29.94
Group
4
1514.94
378.730
47.19
Gender
Group
4
190.30
47.575
5.93
Residual
90
722.30
8.026
Total
99
2667.79
17.9+ The summary table suggests which of the following conclusions?
17.10+ Why does Group have 4 degrees of freedom?
17.11+ What does the significant F for Group most likely mean?
17.12+ How many cells are there in this design?
17.13 Which of the following is not an advantage of factorial designs over one-way
designs?
17.14 In a factorial design a cell is
Test Bank
358
17.15 The notation
..X
stands for
17.16 In the text the Eysenck study of recall as a function of Age and Instructions
allowed us to see that
17.17 The main effect of a variable is
17.18 To look at an interaction effect we must
17.19 A simple effect is defined as
17.20 To calculate the sum of squares for a treatment effect in the analysis of variance,
we would work with
17.21+ Which of the following is NOT true in a factorial analysis of variance?
Chapter 17
359
17.22 The degrees of freedom for an interaction in a two-way factorial are equal to
17.23+ To calculate the F for the interaction in an analysis of variance we
17.24 In a factorial analysis of variance you cannot have
17.25 If you have a significant interaction, you should
17.26+ Which of the following graphs is most likely to portray an interaction?
(I) (II)
(III)
Test Bank
17.27+ In the three graphs above, which one is most likely to have a main effect for
quarters (the variable that increases along the X axis)?
17.28 In graph I above, the most apparent simple effect is for the line represented by
17.29 A simple effect is calculated by
17.30+ To calculate the F for a simple effect you
17.31 If you have a significant interaction,
17.32+ Unequal sample sizes in a factorial analysis of variance
17.33 Unequal sample sizes in a factorial analysis of variance are
Chapter 17
361
17.34+ In the factorial design analyses discussed in Chapter 17, the different cells
17.35+ The following is a printout from SPSS.
Tests of Between-Subjects Effects
Dependent Variable
210.854
a
5
42.171
3.984
.005
12707.521
1
12707.521
1200.373
.000
122.792
2
61.396
5.800
.006
67.688
1
67.688
6.394
.015
20.375
2
10.188
.962
.390
444.625
42
10.586
13363.000
48
655.479
47
Source
Corrected Model
Intercept
GROUP
EDUCATIO
GROUP * EDUCATIO
Error
Total
Corrected Total
Type III
Sum of
Squares
df
Mean
Square
F
Sig.
R Squared = .322 (Adjusted R Squared = .241)
a.
From this table, which of the following conclusions would be wrong?
17.36 To calculate the magnitude of effect estimates for a factorial design, the methods
17.37+ In a study which investigated the effects of amount of coffee consumption and
mood (good or bad) on driving speed, the magnitude of effects estimates were as
follows: Coffee:
12.
2=
C
; Mood:
20.
2=
M
; Coffee x Mood:
19.
2=
CM
.
Together, how much of the variability in driving speed is accounted for by
Coffee, Mood, and their interaction?
Test Bank
362
17.38 If 2 is calculated to yield the magnitude of effect estimates instead of 2 for a
particular experiment, the 2 estimates would probably be
17.39 Suppose you are interested in convincing high school students to avoid taking
drugs, and you have three different videos you could show them. You want to
know whether there is a difference in effectiveness of the videos, and whether the
effectiveness differs for males and females. You set up a design of different
students to watch the films as a 2 X 3 ANOVA with a rating of tendency to avoid
drugs as the dependent variable.
17.40 Which of the following has a main effect for Gender and a significant interaction?
(Give the best answer.)
Chapter 17
17.41 In a factorial ANOVA, the interaction is best defined as:
17.42 A pediatrician is studying weight gain in infants. He divides them into 2 groups:
breast fed and bottle fed. Further, he divides them into those whose mothers feed
them on a timed schedule, and those whose mothers feed them when they cry (on
demand). Weight gain is the dependent measure. What type of analysis should
you run?
17.43 _______ are the effect of one variable at one level of the other variable.
17.44 Pliner and Chaiken (1990) wanted to investigate whether the amount of food
eaten depended on the gender of the participant and the gender of the confederate.
It was observed that women eat less than men overall and that women eat less in
the company of men than they do when in the company of other women. The
finding that women eat less than men across all conditions is a(n)
17.45 The finding that women eat less in the company of men then when they are in the
company of other women is a(n)
17.46 A factorial design has at least
Test Bank
17.47 In a factorial design involving the sex of the participant and the sex of the
experimenter’s confederate
17.48 In this graph we can see that there is
Estimated Marginal Means of MEASURE_1
TIME
4321
Estimated Marginal Means
400
380
360
340
320
300
280
260
240
220
LOCATION
1.00
2.00
17.49 When comparing differences in an experiment with two or more independent
variables we should use a(n)
17.50 When you compare the effect of one variable at one level of another variable you
are examining
Chapter 17
365
17.51 This graph represents a/an
Estimated Marginal Means of MEASURE_1
TIME
4321
Estimated Marginal Means
400
380
360
340
320
300
280
260
240
220
LOCATION
1.00
2.00
17.52 The overall effect of an independent variable is called a(n)
17.53 Dr. Gates looked at the effects of frustration on the use of profanity by males and
females. Males and females were asked to write a lab report on computers in a
lab, but half the computers were set up to crash during the session while half of
the computers were not set up to crash. Three observers recorded the use of
profanity by the participants during the task. What is the design of this study?
17.54 In the Spilich et al. study of the effects of smoking that was discussed in the text,
active smokers were found to do better than nonsmokers on a driving task but did
worse than nonsmokers on a cognitive task. However, over all three tasks (the
third was pattern recognition and the groups were not different on that) active
smokers did not differ from nonsmokers on performance. The results suggest
Test Bank
366
17.55 What type of design is the above study?
17.56 When we compute an effect size measure such as
ˆ
d
for a factorial ANOVA we
have to decide
17.57 When we say that a measure is “not of theoretical interest” we mean that
17.58 Using the example in the text of a participant receiving therapy while sitting in a
bath of ice water, what would be the best denominator for calculating
ˆ
d
?
17.59 To calculate the F for a simple effect you
TRUE/FALSE QUESTIONS
variable.
and the other has four.
main effect.
Chapter 17
367
simple effect.
will be calculated.
with three independent variables.
OPEN-ENDED QUESTIONS
17.69 What value appears in each identified cell in the following table?
Fall
Winter
Spring
Classroom A
10
15
16
Classroom B
12
12
14
Classroom C
8
10
12
Classroom D
10
12
10
a) X11
b) X23
c) X41
17.70 Determine whether each of the following effects is statistically significant at =
.05.
Source
df
F
Gender
1
2.56
Treatment
3
2.75
Gender X Treatment
3
3.26
Error
120
Total
127
17.71 In the previous example,
a) how many treatment groups were being compared?
b) what was N?
c) how many cells are there in the design?
Test Bank
368
17.72 A doctor examining the effectiveness of smoking cessation programs wanted to
examine the independent and joint effect of a support group and the patch. The
following data are the average number of packs smoked 2 weeks after the
interventions. Each group consisted of 10 people. Answer the following
questions based on this table of means.
Using Patch
Not using patch
In support group
0
2.5
Not in support group
1.5
3.0
a) What is the grand mean?
b) Calculate the mean packs smoked at each level of support group.
c) Calculate the mean packs smokes at each level of the patch.
17.73 Plot the data from the previous question three times (each main effect and the
interaction effect); interpret the graphs.
A human resources director for a large company wanted to compare salaries based
on minority status and the type of job. Answer the following questions based on
the following SPSS output.
17.74 a) How many levels are there for each factor?
b) Which effects are statistically significant?
17.75 Calculate and explain 2 for the significant effects from the previous data.
17.76 Calculate and interpret F for each effect based on the following data.
Source
SS
df
Treatment group
1700.50
4
Problem severity
1144.57
2
Treatment X Severity
3008.504
8
Error
30412.86
186
Total
36266.434
200
Chapter 17
369
17.77 Calculate and explain
2
for treatment group from the previous problem.
17.78 A researcher noted that there was a significant interaction effect of amount of
time studying and hours of sleep the night before an exam on exam scores. He
calculated simple effects to try to interpret the data. Here are the results. Graph
them and explain the nature of the interaction.
Study 2 or more hours
Study less than 2 hours
t
6 or more hours of sleep
83
75
2.65*
Less than 6 hours of sleep
76
74
1.35
df = 30; *p < .05
Test Bank
370
Answers to Open-ended Questions
Chapter 17
371