Chapter 9 Flashcards
Multivariate design
Involves more than two measured variables
Will get closer to causality than bivariate
Three criteria for caustion
Covariance: Are the two variables related?
Temporal precedence: Does one variable cause the other variable?
Internal validity: Is there a third variable that might explain the relationship of the first two variables?
Longitudinal design
can provide evidence for temporal precedence by measuring the same variables in the same people at several different times; used by developmental psychologists; gets us closer to a causal claim
Repeated observations of the same variables over short and long periods of time
Longitudinal designs yield three types of correlations
cross-sectional correlations, autocorrelations, and cross-lag correlations
Cross-sectional correlations
whether two variables measured at the same point in time are correlated
Autocorrelations
the correlation of each variable with itself across time.
Cross-lag correlations
correlation of the degree to which an earlier measure of one variable is associated with a later measure of the other variable; examines how people change over time. Researchers are most interested in cross-lag correlations because they help establish temporal precedence.
Covariance 9
when the cross-sectional correlations are significant, covariance has been established.
Temporal precedence 9
if one of the cross-lag correlations is stronger than the other, this it moves us closer to figuring out which variables causes the other.
Internal Validity
by measuring only four variables, as we have in the TV and aggression example, longitudinal designs do not rule out third-variable explanations. Researchers might be able to design their studies in particular ways or do subsequent statistical analyses to address some third variables.
Multiple regression (also known as multivariate regression
helps address questions of internal validity by ruling out some third variables
Criterion variable
dependent variable
What you’re most interested in.
Either in the title or the top row of table
(for example, pregnancy risk)
Predictor variables
independent variables.
Listed below criterion variable in the table
for example, amount of sexual content teenagers reported viewing on TV and the age of each teen
Parsimony
simplicity; the degree to which a good scientific theory provides the simplest explanation of some phenomenon