Reviewing the Three CausalCriteria

1. Why can’t a simple bivariate correlational study meet allthree criteria for establishing causation?

Establishing Temporal Precedence with LongitudinalDesigns

1. Why is a longitudinal design considered a multivariatedesign?

2. What are the three kinds of correlations obtained from alongitudinal design? What does each correlation represent?

3. Describe which patterns of temporal precedence are indicatedby different cross-lag correlational results.

Ruling Out Third Variables with Multiple-RegressionAnalyses

1. Describe what it means to say that some variable “wascontrolled for” in a multivariate study.

2. How many criterion variables are there in amultiple-regression analysis? How many predictor variables?

3. What does a significant beta mean? What does a nonsignificantbeta mean?

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

5. What are two reasons that multiple-regression analyses cannotcompletely establish causation?

Getting at Causality with Pattern andParsimony

1. Why do many researchers find pattern and parsimony aneffective way to support a causal claim?

2. What is a responsible way for journalists to cover singlestudies on a specific topic?

Mediation

1. Explain why each of the five steps examining a mediationhypothesis is important to establishing evidence for amediator.

2. Think of a possible mediator for the relationship betweenexposure to sex on TV and chance of pregnancy. Sketch a diagram ofthe mediator you propose, following Figure 9.11. (It is ok to justdescribe your sketch)

Multivariate Designs and theFour Validities

1. Give an example of a question you would ask to interrogateeach of the four

validities for a multivariate study.