Chapter 9 - Multivariate Correlational Research

Question 1

Multivariate Designs

Answer 1

A study designed to test an association involving more than two measured variables.

Question 2

Why can’t a simple bivariate correlational study meet all three criteria for establishing causation?

Answer 2

The three criteria for establishing causation are covariance, temporal precedence, and internal validity.
(1) Covariance: yes, suggests that there is a correlation!
(2) Temporal Precedence: cannot say which variable came first and caused the second variable
(3) Internal Validity: third variables are not usually controlled for!

Question 3

Longitudinal Design

Answer 3

A study in which the same variables are measured in the same people at different points in time.
- Can provide evidence for temporal precedence by measuring the same variables in the same people at different points in time.
- Can be adapted to test causal claims.
- Statistical relationships in longitudinal designs help establish covariance.
- Internal validity: by measuring only the two key variables, longitudinal studies may not help rule out third variables.

Question 4

Cross-sectional Correlations

Answer 4

In a longitudinal design, a correlation between two variables that are measured at the same time.
- Test to see whether two variables measured at the same point in time are correlated.

Question 5

Autocorrelations

Answer 5

In a longitudinal design, the correlation of one variable with itself, measured at two different times.
- Evaluates the correlation of each variable with itself across time.

Question 6

Cross-lag Correlations

Answer 6

In a longitudinal design, a correlation between an earlier measure of one variable and a later measure of another variable. Investigates how one variable correlated with another one (that’s the “cross” part of the name) over time (that’s the “lag” part).
- Shows whether the earlier measure of one variable is associated with the later measure of the other variable.
- Addresses the directionality problem and help establish temporal precedence.

Question 7

Why is a longitudinal design considered a multivariate design?

Answer 7

Within the longitudinal design, more than two variables can be measured within the design.

Question 8

What are the three kinds of correlations obtained from a longitudinal design? What does each correlation represent?

Answer 8

• cross-sectional correlations: correlation between different variables at the same point in time.
• autocorrelation: correlation between the same variable but at different points in time.
• cross-lag correlations: a correlation between an earlier measure of one variable and a later measure of another variable.

Question 9

Describe which patterns of temporal precedence are indicated by different cross-lag correlational results.

Answer 9

Either variable 1 at time 1 is correlated to variable 2 at time 2, or variable 2 at time 1 is correlated to variable 1 at time 2, or there might be correlations between both variables at both times when cross-lagged with each other (i.e., all of the above mentioned).

Question 10

Multiple Regression (or multivariate regression)

Answer 10

A statistical technique that computes the relationship between a predictor variable and a criterion variable, controlling for other predictor variables.

Question 11

“Control for” meaning

Answer 11

Holding a potential third variable at a constant level (statistically or experimentally) while investigating the association between two other variables.

Question 12

Criterion Variable

Answer 12

The variable in a multiple-regression analysis that the researchers are most interested in understanding or predicting.
- Also called a dependent variable.

Question 13

Predictor Variables

Answer 13

A variable in multiple-regression analysis that is used to explain variance in the criterion variable.
- Also called an independent variable.

Question 14

How many criterion variables are there in a multiple-regression analysis? How many predictor variables?

Answer 14

One criterion variable and at least two predictor variables.

Question 15

What is the relationship between the 95% CI for beta and the beta’s statistical significance?

Answer 15

When the 95% CI does not include zero, we can say that the beta is statically significant, p < .05.
When the 95% CI does include zero, we say that the beta is NOT statically significant, p > .05.

Question 16

Give at least two phrases indicating that a study used a multiple-regression analysis.

Answer 16

• The phrase “adjusting for” signals a regression analysis, thereby ruling out those variables as internal validity problems.
• The phrase “even when other factors were considered” or “considering” indicates the researchers used multiple regression.

Question 17

What are the two reasons that multiple-regression analyses cannot completely establish causation?

Answer 17

• Although multivariate designs analyzed with regression statistics can control for third variables, they cannot always establish temporal precedence.
• Researchers cannot control for variables they do not measure. Multiple regression can control for any third variables the researchers do measure in the study; however, some other variables they did not consider could account for the association.

Question 18

What is ‘Beta?’

Answer 18

Betas are similar to correlations (r) in that they denote the direction and strength of a relationship (between the predictor and criterion variable).
Betas change depending on what other predictor variables are being used - being controlled for - in the regression.
The beta that is associated with a predictor variable represents the relationship between that predictor variable and the criterion variable when the other predictor variables in the table are controlled for.

Question 19

Parsimony

Answer 19

The degree to which a theory provides the simplest explanation of some phenomenon. In the context of investigating a claim, the simplest explanation of a pattern of data; the best explanation that requires making the fewest exceptions or qualifications.

Question 20

Why do many researchers find a pattern and parsimony an effective way to support a causal claim?

Answer 20

In psychology, researchers commonly use a variety of methods and many studies to explore the strength and limits of a particular research question. Often using a broader, simple explanation for how variables relate can keep confounds/third variables limited that occur when one gets more specific with a question.

Question 21

What is a responsible way for journalists to cover single studies on a specific topic?

Answer 21

Journalists should report on the entire body of evidence and the theoretical background for a particular claim.

Question 22

Mediator

Answer 22

A variable that helps explain the relationship between two other variables. Also called mediating variable.
- Study can be correlational or experimental; the researcher might use mediation analyses using multivariate tools such as multiple regression

Question 23

What three variables would you have to measure in order to test your mediator hypothesis, and when would you have to measure them?

Answer 23

The three variables should be measured at different points in time, in this order:
- predictor variable
- mediator variable
- criterion variable

Question 24

What are examples of a ‘mediator?’

Answer 24

We know conscientious people are more physically healthy than less conscientious people. But why? The mediator of this relationship might be the fact that conscientious people are more likely to follow medical advice and instructions, and that’s why they’re healthier. Following the doctor’s orders would be the mediator of the relationship between the trait, conscientiousness, and the outcome, better health.

In the image, the overall relationship exists because there are two other relationships: a(b/w deep talk and social ties) and b (b/w social ties and welling being). i.e. social ties mediate the relationship between deep talk and well-being.

Question 25

What is the difference between a moderator and a mediator?

Answer 25

A mediator variable explains the process through which two variables are related, while a moderator variable affects the strength and direction of that relationship.

Question 26

Give an example of a question you would ask to interrogate each of the four validities for a multivariate study.

Answer 26

Construct validity: asking how well each variable was measured.
External validity: asking whether the participants were sampled randomly and from what kind of population.
Statistical validity: ask about the point estimates and the confidence intervals, and ask whether the study has been replicated.
Internal Validity: longitudinal designs help establish temporal precedence, and multiple-regression analysis helps rule out third variables, this providing evidence for internal validity.

Question 27

Mediator / Mediating Variable

Answer 27

A variable that helps explain the relationship between two other variables.