PSYC 3500 Study Guide - Quiz Guide: Univariate Distribution, Dependent And Independent Variables, Joint Probability Distribution
0 views3 pages
PSY 3500: Correlation & Regression
• Univariate distribution - We focus on one variable in time.
• Bivariate distribution - Statistics related to the relationship between two variables.
• Correlation - A statistic that times whether not to distribution/variables are related;
specifies the nature ( + or -) or degree of the relationship.
• Requires pairs of the scores for each participant.
• Regression - A process in which we can predict values of one variable from values of
another variable; takes a correlation one step further by adding in a line of best fit.
• Range of r - -1 - +1
• If r is a positive number - You have a positive correlation.
• As one variable increases the other variable increases.
• If r is a negative number - You have a negative correlation.
• As one of able decreases the other variable decreases.
• If r equals zero - We have no linear relationship between two variables.
• Scatterplot - How we graph scores of bivariate frequency distributions.
• Interpreting r - The larger the absolute value of r the larger the correlation
• Weak - r=.10
• Medium / moderate - r=.30
• Large / strong - r=.50
• Lower third - r<.20
• Middle third - r=.20 - .30
• Upper third - r>.30
• What does r not tell us? - R does not imply a percent of association
• R=.8 is not two times the effect of an r=.4
• Why are correlations useful? - Prediction