Chapter 9 Flashcards
Why are simple bivariate correlations not sufficient for establishing causation?
It always indicates covariance but can not always indicate temporal precedence or internal validity - so it can not establish causation.
How can longitudinal correlational designs establish temporal precendence?
Longitudinal designs start with two key variables where the same group of people are measured at multiple points in time. Researchers can tell which variable came first, therefore establishing temporal precedence.
How can multiple regression analyses rule out some, but not all, third variables?
In regression designs, researchers start with a bivariate correlation and then measure other potential third variables that might affect it. With multiple-regression analysis, researchers can see whether the basic relationship is still present, even when they statistically control for one or more third variables. But, they cannot establish causation because they can only control for possible third variables that the researchers happened to measure. Only an experiment can definitively establish causation.
What is pattern and parsimony and why is it valuable?
Pattern and parsimony investigates causality by using a variety of correlational studies that all point in a single causal direction. Parsimony is the degree to which a scientific theory provides the simplest explanation.
How does a variety of research results support a single, parsimonious causal theory?
Researchers can approach causal certainty through pattern and parsimony by specifying a mechanism for the causal relationship and combining the results from a variety of research questions. When a single causal theory explains all of the disparate results, researchers
are closer to supporting a causal claim.
What is a mediating variable and what is it’s function?
A mediating variable is one that answers why one factor causes another and their relationship
What is a multivariate design?
One where researchers measure more than two variables and look for the relationships among them. Examples include longitudinal designs, multiple-regression designs and pattern and parsimony approach.
What type of correlations do longitudinal designs produce?
- Cross-sectional correlations - correlations between two key variables at any one time point
- Autocorrelations - correlations between one variable and itself, over time
- Cross-lag correlations - correlations that show whether the earlier measure of one variable is associated with the later measure in the other variable > addresses directionality
What can researchers determine from cross-lag correlations?
By comparing the relative strengths of two cross-lag correlations, researchers can infer which of the variables probably come first in time.
What does the beta tell in multiple-regression analyses?
If the beta is still significant for the key variable when the researchers control for the third variables, it means the key relationship is not explained by those third variables. lf the beta approaches zero when the researchers control for a third variable, then the key relationship can be attributed to that third variable.
How do researcher’s deal with the mediation hypothesis?
In a mediation hypothesis, researchers specify a variable that comes between the two variables of interest as a possible reason the two variables are associated. After collecting data on all three variables (the original two, plus the mediator), they use statistical techniques to evaluate how well the data supports the mediation hypothesis
Why would one do a longitudinal study instead of an experiment?
Sometimes, variables can not be manipulated and it may be unethical to induce certain behaviours that are bad
What is the difference between mediation and moderation?
while mediation determines the why in a relationship, moderation looks at for whom the relationship applies to
