Chapter 9: Multivariate Correlations

What is multivariate vs bivariate correlational research?

• Bivariate - an association that involves exactly 2 variables

• Multivariate - associations that involve >2 measured variables

What are the 3 criteria for causal claims?

• Covariance - the variables are related

• Temporal precedence - one variable comes before the other

• Internal validity - no 3rd variables

What are two types of multivariate correlational designs?

• Longitudinal designs

• Multiple regression analyses

What are longitudinal designs?

Studies in which the same variables are measured in the sample people at different points in time

What are cross-lag correlations?

• A correlation between an earlier measure of one variable and a later measure of another variable

  • *Looking at the variables across times

What is an autocorrelation?

• Bivariate correlation between one variable measured at different times

• *They determine the correlation between one variable with itself, measured on two different occasions

What is a cross-sectional correlation? (2)

• Bivariate correlation between 2 variables measured at the same time

• *Different people measured on the same constructs at the same time

What can longitudinal studies provide us?

Establish covariance & temporal precedence

How do longitudinal designs help with temporal precedence?

Measuring the same variables in the same people at several points in time

What is a multiple regression?

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

What does it mean to statistical control for a variable?

Holding a potential third variable at a constant level while investigating the association between two other variables

What is a moderator, and what kind of question does it answer?

• A variable that changes the relationship between two other variables

• Are there certain groups or situations for which the two variables are more strongly related?

What is a mediator, and what kind of question does it answer?

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

• Why are two variables related?

Daily diary designs/ecological momentary assessments (5)

• Collects regular data on a particular variable over a period of time

• Usually qualitative assessments

• E.g. dream journals

• Pros: immediate assessment of a state, eliminate memory bias

• Cons: intensive, high attrition (ppl quit participating quickly)

Cross-sections vs experiments (2)

• No random assignment or manipulation

• All variables are measured

Multiple regression & internal validity (2)

• Cannot establish _________

• We can rule out third variables, but only the ones we measured

• Cannot establish temporal precedence

What is parsimony?

The degree to which a scientific theory provides the simplest explanation of some phenomenon