Bivariate Correlational Research Flashcards
Correlational Research
Research that focuses on the degree to which measured variables are related (supports association claims)
Bivariate Correlation
The degree to which exactly 2 measured variable are related (correlations allow you to predict outcomes) (correlations do NOT imply causation)
What can we use to examine correlations?
We can use a scatterplot to examine correlations
Pearson Correlation Coefficient (r)
A static that describes the relationship b/w two variables. Ranges from -1.0 to 1.0. The sign (+ or -) only indicates DIRECTION. The absolute value or r indicates strength. Can only use if you have a LINEAR RELATIONSHIP.
What would be a positive correlation?
As variable A goes up, variable B also goes up
What would be a negative correlation?
As variable A goes up, variable B goes down (or vise versa)
Strength of a correlation
Strong: Can predict variable A from varaibale B correlation.
Weak: Cannot easily predict variable A from variable B. Correlation coefficient (r) is closer to 0
Effect Size
Magnitude/Strength of a relationship b/w 2 or more variables. The closer that r is to 1.0 (or -1.0) the stronger the correlation 9 and the bigger the effect size. The stronger correlation means that one variable can be predicted based on what you know about the other variable
Describing strength of a correlation
No perfect benchmark. In psych, the average size of a correlation is (r = +/- .20)
Precision + Significance of a correlation
In psychology, we often use 95% CI
Factors that distort correlations
Restricted range, curvilinear relationships, outliers, unreliable measures
Restricted Range
Causes researchers to underestimate the strength of the correlation. Researcher wants to estimate the correlation b/w 2 variables in a population, but subjects are selected on X and data for Y are only available for a selected sample
Curvilinear Relationships
Linear Relationship: The nature of the relationship b/w variable A + B remains consistent. Linear Relationships = Straight line relationships
Curvilinear Relationship: Association
The rel. b/w 2 variables is not a straight line. Curvilinear Relationshisp = Curved Line
(EX: The association b/w 2 variables might be positive up to a point and then become negative)
Outliers
Can be problematic b/c they may exert disproportionate influence. Can exaggerate or reduce the strength of relationship. Can be more influential with a smaller sample size
