Deck 18: Moving Beyond the Basics of Research and Analysis: How Do I Understand More Complicated Research Designs

Question 1

You have run a one-way ANOVA where there are four conditions. You choose to break down the main effect using Bonferroni corrected t tests. How many t tests will you need to run, and what is the corrected alpha level that you should use?

A) Six comparisons and an alpha level of .050
B) Four comparisons and an alpha level of .050
C) Six comparisons and an alpha level of .008
D) Four comparisons and an alpha level of .008

Answer 1

C
Explanation: There are six comparisons (1v2, 1v3, 1v4, 2v3, 2v4, 3v4) and the alpha level is therefore .050/6=.008

Question 2

Sphericity is an assumption in a repeated measures ANOVA. How is sphericity calculated?

A) By looking at the variability within each condition
B) By looking at the difference between the first and the last condition.
C) By looking at the size of the differences between conditions
D) By looking at variability in difference scores between conditions.

Answer 2

D
Explanation: Sphericity looks at variability in difference scores between conditions.

Question 3

You want to analyse a dataset looking at differences in critical thinking ability between psychology students and physics students and the end of each year of study across a three year degree. What type of design has been used?

A) A factorial mixed design
B) A factorial repeated design
C) A factorial independent design
D) A non-factorial mixed design

Answer 3

A
Explanation: It is a factorial design because two IVs were manipulated. It is a mixed design because there is one independent measures IV (degree subject: psychology or physics) and one repeated measures IV (year of study: 1, 2 or 3).

Question 4

You want to analyse a dataset looking at differences in critical thinking ability between psychology students and physics students and the end of each year of study across a three year degree. How would you describe the analysis that you need to conduct on this data set?

A) 1 * 6 mixed design ANOVA
B) 2 * 3 mixed design ANOVA
C) 2 * 3 repeated measures ANOVA
D) 1 * 6 independent measures ANOVA

Question 5

You want to analyse a dataset looking at differences in critical thinking ability between psychology students and physics students and the end of each year of study across a three year degree. Your analysis shows that critical thinking skills improve greatly across the degree in psychology students, but remains exactly the same in physics students. What kind of effect is this?

A) A significant interaction
B) A not significant interaction
C) A significant main effect
D) A not significant main effect

Question 6

In an ANCOVA, how many difference sources of variability are calculated?

A) 1
B) 2
C) 3
D) 4

Question 7

You want to look at the relationship between two continuous variables whilst controlling for covariance. What type of analysis should you use?

A) Pearson's correlation
B) Partial correlation
C) Spearman's correlation
D) Logistic regression

Question 8

You want to run an analysis where you see which variables are significant, starting with the most significant predictor and then extracting further significant predictors in order of decreasing significance. What type of regression modelling would you use?

A) Stepwise regression
B) Logistic regression
C) Hierarchical regression
D) Multiple regression

Question 9

What type of regression model would you use if you wanted to predict a binary outcome?

A) Multiple regression
B) Hierarchical regression
C) Logistic regression
D) Stepwise regression

Question 10

Which of the following is an advantage of using effect sizes rather than p values?

A) They are only necessary when a finding is significant
B) They are only necessary when a finding is not significant
C) Effect sizes are affected by sample size
D) Effect sizes are unaffected by sample size