Ch. 11: Quantifying and Interpreting Relationships Among Variables
Correlation Coefficient A single number indicating the strength of association between 2 variables. To what extent does the association resemble a straight line? Pearson r as coefficient of choice Value ranges from 0 (no relationship) to 1 (perfect linear relationship) Sign indicates direction of relationship
Correlation and Causation Causation implies correlation, but correlation does not imply causation. Temporal precedence problem Third variable problem
Types of Variables Continuous variables Discrete variables Dichotomous variables
Example of a Scatter Plot This person scores a 10 on Exam 1 and a 40 on Exam 2
Scatter Plots of Different Values of the Correlation Coefficient
Pearson r Assesses the linear relationship between two continuous variables Product Moment Correlation Conceptual Formula:
Alternative Formula for Pearson r
Spearman Rho Assesses the linear relationship between two variables that are in the form of ranked data Formula: “D” is the difference between the pairs of ranked scores “6” is a constant “N” is the number of score pairs
Point-Biserial Correlation One variable is continuous and the other is dichotomous Dummy coding is used to quantify the one dichotomous variable Formula is same as Pearson r
Phi Coefficient Both variables are dichotomous Could convert scores to z-scores and use Pearson r formula
Alternative Formula for Phi “2 X 2 Table of Frequencies (or Counts)” A D C B