exam 3 chapter 9 Flashcards
what is the purpose of longitudinal design
allows the researchers to establish temporal precedence in their data. Measuring the same variables in the same people at different points in time to study changes in a trait or ability as a person grows older
what are the three correlational techniques to make a causal claim, and are these variables measured or manipulated?
longitudinal design, multiple regression analyses, and the pattern and parsimony approach. All are measured, NOT manipulated
what is the purpose of multiple regression analyses
help researchers rule out certain third variable explanations
what is the purpose of the pattern and parsimony approach
the results of a variety of correlational studies all support a single causal theory
are the longitudinal design, multiple regression analyses, and the pattern and parsimony approach bivariate or multivariate
multivariate involves more than 2 measured variables
what 3 individual correlations do multivariate designs give?
cross-sectional correlations, autocorrelations, and cross-lag correlations
what is cross sectional correlations
test to see if two variables measured at the same point in time are correlated. This can’t establish temporal precedence because they were measured at the same time e.g., observing aggression in third graders as well as their tv violence preference at the same point in time.
what is autocorrelation
same as test-retest reliability. Determines the correlation of one variable with itself, measured on two different occasions. Can establish temporal precedence
EX: tv violence in third grade compared to tv violence in ‘13th’ grade
what is cross lag correlations
researchers are usually most interested in this. shows whether the earlier measure of one variable is associated with the later measure of the other variable. This addresses the directionality problem and helps establish temporal precedence
EX: aggression in third grade—is this associated with tv violence in 13th grade? And vice versa, is tv violence in 3rd grade associated with aggression in 13th grade?
if the CI contains zero, the correlation is… and if it does not contain zero, the correlation is…
not significant, significant
which of the three criteria for causal relationships do longitudinal designs meet?
covariance: when the CI does not contain 0, significant relationships
Temporal precedence: because each variable is measured at different points in time, they know which came first. Using the strength of the cross-lag correlations, the researchers can see which path is stronger. If one is stronger than the other, the researchers move closer to determining which variable came first, causing the other.
internal validity: When conducted simply, longitudinal studies do not help rule out third variables. Careful researchers may be able to design their studies or conduct the analysis in ways that address some third variables.
what does control for mean
holding a potential third variable at a constant level while investigating the association between two other variables. this suggest its a regression analysis
criterion variable
the variable in the multiple regression analysis that researchers are most interested in understanding or predicting is also called the dependent variable
predictor variable
(the rest of the variables that are not the criterion variable) a variable in multiple regression analysis that is used to explain variance in the criterion variable. Also called the independent variable
what is beta
there is one beta for each predictor variable, and it is similar to r. A positive beta is a positive relationship between the predictor and criterion variable when the other predictors are controlled for. A negative beta is a negative relationship between the predictor and criterion variable when the other predictors are controlled for. A beta of zero represents no relationship when the other predictors are controlled for. the smaller the beta, the weaker the relationship, and the larger the beta, the stronger the relationship.
can you compare betas from different regression tables?
Don’t compare betas from different regression tables because they change depending on the predictor variable.
what does adjusted for mean
researchers conducted multiple regression analyses
what does considering mean
researchers conducted multiple regression analyses
does regression establish causation
no, because it doesnt always account for temporal precedence and additional third variables not measured by researchers may be affecting it as well
parsimony
the degree to which a scientific theory provides the simplest explanation of some phenomenon or for a causal claim: the simplest explanation of a pattern of data-the theory that requires making the fewest exceptions or qualifications
mediator
A variable that helps explain the relationship between two other variables. asking the “why”
does a study have to be correlational to include a mediator
no experimental studies can also test them
what statistical techniques are used to test for mediation
compute relationships between all variables, multiple regression, structural equation modeling
because mediation hypotheses are causal claims, mediation is definitively established only in conjunction with…
temporal precedence
