Inferential Statistics

Question 1

Additivity

Answer 1

Assumption that If you have several predictor variables, their combined effect is best described by adding their effects together

Question 2

Linearity

Answer 2

Assumption that the outcome variable is linearly related to any predictors (i.e., their relationship can be summed up by straight line)

Question 3

When does Linearity matter?

Answer 3

When the predictor variable is continuous

Question 4

What can you do when data indicates a non-linear relationship?

Answer 4

Or you can use a non-parametric test to analyze your data

Question 5

Normality

Answer 5

Assumption that the sampling distribution of the parameter you are trying to estimate (not your data itself) is normally distributed

Question 6

Central Limit Theorem

Answer 6

As the sample size increases, the sampling distribution of those parameters becomes increasingly normal

Question 7

Homogeneity

Answer 7

Homogeneity means that as you go through the levels of the predictor variable, the variance of the outcome variable should not change

Question 8

How can you spot violations of homogeneity?

Answer 8

Using numbers: Levene‘s test. A significant Levene‘s test indicates that homogeneity assumption is violated

Question 8

How can you avoid that a homogeneity violation biases your results?

Answer 8

Use the Welch t-test instead of the student t-test. The Welch test corrects for any violation of homogeneity

Question 9

Independence

Answer 9

Assumption that the errors in your model are not related to each other (e.g., measurement error of Participant 1 should be unrelated to measurement error of Participant 2)

Question 10

Sphericity

Answer 10

Assumption only applies to within-subject designs with 3 or more conditions

Sphericity assumption means that the variances of the differences between conditions should be equal

Question 11

How can you spot violations of sphericity?

Answer 11

Using Mauchly‘s test (included in SPSS repeated measure procedure). A significant Mauchly‘s test indicates that sphericity assumption is violated

Question 12

How can you avoid that sphericity violations bias your results?

Answer 12

Greenhouse-Geisser (more conservative)
Huynh-Feldt (less conservative)

Question 13

When to use and independent samples t-test

Answer 13

between subjects design

Question 14

When to use paired samples t-test

Answer 14

within subjects design

Question 15

Directional vs. Non-directional hypothesis

Answer 15

A model that tests a directional hypothesis is called a one-tailed test, a model that tests a non-directional hyp. is a two-tailed test