PSYC3890: Chapter 9
PSYC3890: Chapter 9 PSYC3980
Popular in Research Methods in Psychology
Brittany Ariana Borzillo
verified elite notetaker
Popular in Psychology
verified elite notetaker
This 4 page Class Notes was uploaded by Brittany Ariana Borzillo on Monday October 3, 2016. The Class Notes belongs to PSYC3980 at University of Georgia taught by in Fall 2016. Since its upload, it has received 2 views. For similar materials see Research Methods in Psychology in Psychology at University of Georgia.
Reviews for PSYC3890: Chapter 9
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/03/16
Chapter 9: Multivariate Studies Making an association claim regarding more than 2 measured variables Help to control for some causal requirements o Still unable to claim causation Deals most with covariance, temporal precedence, and internal validity Longitudinal Often used in developmental psychology Same variables with same individuals over the course of time Help establish temporal precedence o measuring the same variables at several points of time o measuring two different variables at two different points of time Covariance o When two variables are significantly correlated there is covariance Temporal Precedence o Helps researchers make inferences about temporal precedence Internal validity o Fail to help rule out third variables Cross-Sectional o 2 variables measured at the same time point o test whether two variables that are measured at the same point of time are related Autocorrelation o Variable and itself measured after a change in time Cross-lag Correlation o Earlier measure of one variable with later measure of the other variable o Two cross-lag correlations address the directionality problem and helps further establish temporal precedence o Allows investigation of how people change over time o Patterns Variable 1 at time 1 is strongly correlated with variable 2 at time 2 while variable 2 at time 1 is not strongly correlated with variable 1 at time 2 Variable 2 at time 1 is strongly correlated with variable 1 at time 2 while variable 1 at time 1 is not strongly correlated with variable 2 at time 2 Both conditions above may be significant Potential Results o Can gain insight into temporal precedence Compare strength of cross-lagged correlations Significance of both is a reciprocal relationship o Still lack control Useful when you can’t randomly assign participants Multiple Regression Similar to correlation o Exception is that there are more than 2 variables Helps control for 3 variable problems Helps determine how well relations hold when other variables are controlled for Regression o Criterion dependent variable o Predictor independent variable o relationship between each independent variable and dependent variable while controlling for the other independent variables interpreted the same was as correlation ® refers to “unique” variance can compare the relative effects of each independent variable may be represented as “b” unstandardized p-values indicate the significance of relationships o regression does not imply causation can’t establish temporal precedence can’t control for variables that aren’t measured o statistically controlling for third variables when looking at the effect of one variable, all other variable predictors are held constant regression does not imply causation Pattern and Parsimony accept results that provide the simplest explanation for the pattern of the data still need multiple diverse studies to decrease 3 variable explanations Mediation explains the relationship between two variables test of mediation o assess c, then a, then b must be related o regression with both independent variable and mediator see how c is affected how moderators differ o influences the strength of a relationship between two other variables 4 Validities internal validity o statistical inferences about causal effects are valid for the population being studied o want to be unbiased and consistent construct validity o still need to assess how each variable is measured external validity o sampling still matters because you won’t be able to assess an entire population o generizability statistical validity o effect size o statistical significance o subgroups o outliers o curvilinear ***I’m not claiming rights to this picture; I merely feel it’ll be helpful for everybody’s studies***
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'