PSY 301 Chapter Notes - Chapter 9: Regression Testing, Internal Validity, Confounding
Document Summary
Scientists often want to turn correlational research into causation. Experimentation with one variable being manipulated, along with other advanced techniques. Multivariate designs: involving more than two measured variables. Multiple-regression designs, and pattern and parsimony approach. Temporal precedence: whether one came before the other. Internal validity: was there a third affecting variable. Longitudinal design: measuring same variables in the same people at several points in time. Test to see whether two variables, measured at same point in time, are correlated. Determine the correlation of one variable with itself, measured at two different occasions. Cross-lag correlations: show whether the earlier measure of one variable is associated with the later measure of the other variable. Three possible patterns from cross-lag study: first variable affects second, second variable affects first, both affect both. Longitudinal studies and the three criteria for causation. Can help researchers make inferences about temporal precedence. Do not help rule out a third variable.