CHAPTER 5: Correlation and Quasi - Experimental Design Flashcards
causing or producing the effect.
Causation
It is used to establish relationship among preexisting behavior. It can show the relationships between sets of antecedent conditions and behavior.
Correlational Design
It is a design that is used to determine the correlation, or degree of relationship, between two traits, behavior, or events.
Correlation Study
is any observable behavior, characteristics, or event that can vary or have different values.
Variable
It is the relationships between pairs of scores from each subject.
Simple Correlation
is the most commonly used procedures for calculating simple correlations.
Pearson Correlation Coefficient (r)
shows the scores on one variable plotted against scores on a second variable.
Scatterplot
it is the line that is drawn on the scatterplots and it measures scores. The direction of the line corresponds to the direction of relationship (positive or negative).
Regression Line (lines of best fit)
Correlation does not imply causation
even though a relationship exists between two measures, we cannot say that one causes the other, even when such statement is reasonable.
means a behavior could affect each other.
Bidirectional Causation
third agent may actually be causing to behaviors appear to be related.
Third Variable Problem
estimates the amount of variability scores on one variable that can be explained by the other variable.
Coefficient of Determination
determines the degree of relation between two variables.
Multiple Correlation
uses regression equation to predict scores on one variable from scores on the second correlated variable.
Linear Regression Analysis
uses regression equation on one variable from scores on sets of other variables.
- represents “R”
Multiple Regression
determines the subsets of correlated variables within a larger set of variable.
Factor Analysis
uses beta weights from multiple regression analysis to generate possible direction of cause and effect from correlated variables.
Path Analysis
measures the same pair if variables at two different points in time, looks at patterns of correlation across time for possible direction of cause and effect.
Cross-lagged Panel
- after the fact in the effect
- is a study in which the researcher systematically examines the effects of subject characteristics but without actually manipulating them.
- capitalizes on changes in the antecedent conditions that occurred before the study.
Ex Post Facto
are collected for each item or variable for one or more time periods
Data
involves some comparison of data between or among periods.
Analysis
investigates changes across time by comparing groups of subjects already at different stage at a “single point” in time.
Cross-sectional Study
investigates changes occur time by measuring behavior of same group of subjects at different point in time. (long term studies)
Longitudinal
it is a threat to internal validity are not completely controlled because the comparison group is taken from another population and is unlikely to be equivalent to the treatment group.
Non-equivalent control group design
it is often used to assess the effect of naturally occurring events when a true experiment is not possible; sometimes such a design is used in a laboratory studies, but it is a number of problems that reduce its internal validity.
Pretest-Posttest design
a group that took both the pretest and posttest but was not exposed to the “treatment” this design has been called “the old workhorse of traditional experimentation”.
Non equivalent control group
