Deck 17: Factorial Anova

Question 1

If there is a treatment effect, the F ratio will be:

A) 1.
B) smaller than 1.
C) greater than 1.
D) any number that is not negative.

Answer 1

B

Question 2

The formula for determining the df_Total in a between subject ANOVA model is: ______.

A) k_i - 1.
B) N - ab.
C) (df_A) (df_B).
D) N - 1.

Answer 2

D

Question 3

The statistical notation G denotes ______.

A) grand mean
B) great mean
C) global mean
D) gain mean

Answer 3

A

Question 4

Alpha levels determine the level of ______ error we are willing to make.

A) Type I
B) Type II
C) Both Type I and Type II
D) None of these

Question 5

The three assumptions of a factorial ANOVA include: normality of distribution, independence of cases, and ______.

A) sphericity
B) homogeneity of variance
C) expected frequency equals zero
D) Both sphericity and expected frequency equals zero

Question 6

The three types of factorial ANOVAs designs are between-subjects, within-subjects, and mixed.

Question 7

A researcher plans to conduct a research study comparing two treatment conditions with a total of 20 scores in each treatment. Which of the following designs would require the smallest number of participants?

A) Between-subjects design
B) Mixed design
C) Within-subjects design
D) None of these

Question 8

What is Eta Squared when SS_between = 30, SS_Total = 46?

A) .6521
B) 1.533
C) 1380
D) 0.76

Question 9

The F ratio is always positive in the ANOVA model.

Question 10

If F(3, 11) = 6.20, we ______ the null hypothesis at alpha .

A) Fail to reject
B) Reject

Question 11

Hans would like to examine whether he has violated the homogeneity of variance assumption. What test should he run?

A) Levene's test
B) Tukey HSD test
C) Dunnett's test
D) Correlated groups t-test

Question 12

An ANOVA model with 2 levels of an IV was conducted and indicated a non-significant main effect. Specify why post-hoc analyses would not be conducted?

A) Post-hoc analyses are not necessary when the IV only has 2 levels.
B) Post-hoc analyses are not necessary following non-significant main effects.
C) Both post-hoc analyses are not necessary when the IV only has 2 levels and post-hoc analyses are not necessary following non-significant main effects.
D) None of these.

Question 13

For an experiment comparing more than two treatment conditions, why should you use analysis of variance rather than separate t tests?

A) Separate t tests would require substantially more computations.
B) A test based on variances is more sensitive than a test based on means.
C) Conducting several t tests would inflate the risk of a Type I error.
D) There is no difference between the two tests, you can use either one.

Question 14

If F(2, 20) = 3.50, we can reject the null hypothesis at α =_______.

A) .05
B) .01
C) .001
D) .00001

Question 15

Out of the following three common post-hoc analyses, which one is the most conservative?

A) Bonferroni correction
B) Fisher's LSD
C) Multi-stage significance testing
D) None of these

Question 16

Emma conducted a 3 x 2 ANOVA, which revealed a significant interaction. If she is interested in interpreting this interaction, which analyses should she now conduct?

A) One sample t-test
B) Chi-Square test
C) Simple effects
D) Wilcoxon Signed-Rank test

Question 17

A 2 x 3 factorial ANOVA has:

A) 2 factors and 3 conditions.
B) 2 factors and 6 conditions.
C) 3 factors and 2 conditions.
D) 3 factors and 6 conditions.

Question 18

What is p the probability of?

A) p is the probability that the results are due to chance, the probability that the null hypothesis (H0) is true.
B) p is the probability of observing results as extreme (or more) as observed, if the null hypothesis (H0) is true.
C) p is the probability that the results would be replicated if the experiment was conducted a second time.
D) p is the probability that the results are not due to chance, the probability that the null hypothesis (H0) is false.

Question 19

Michael is testing the hypothesis that anger has a negative impact on decision making. He reported these findings: F(2,12) = 6.93. How many people were in Michaels study?

A) 9
B) 14
C) 15
D) 36

Question 20

ANOVA is often robust to violations in the skew of the data.