OUTLINE FOR CHAPTER
3
Bivariate Correlation
 The Key Concept Behind Correlation: "Relationship"
 The need for data on two variables
 What's the basic question being addressed in discussions about
correlation?
 Three possible answers to the basic correlational question:
 Highhigh, lowlow
 Highlow, lowhigh
 Little systematic tendency one way or the other
 Scatter Diagrams
 What they look like
 How to interpret one
 The notions of "strong," "moderate," and "low" relationships
 The Correlation Coefficient
 How it's symbolized
 It's numerical range
 How to label different points on and sections of the continuum
of possible values
 The Correlation Matrix
 Determining the number of bivariate correlations among k variables
 The purpose and general appearance of a scatter diagram
 Saving space when showing one correlation matrix (or two correlation
matrixes)
 Different Kinds of Correlational Procedures
 Quantitative vs. qualitative variables
 Dichotomous, nominal, ordinal, and "rawscore" variables
 True and artificial dichotomies
 Specific correlational procedures:
 For two rawscore variables: Pearson's productmoment correlation
 For two sets of ranks: Spearman's rho and Kendall's tau
 For one rawscore variable & one dichotomous variable:
biserial r & pointbiserial r
 For two dichotomous variables: phi and tetrachoric r
 For two nominal variables: Cramer's V
 Warnings About Correlation
 Correlation and causeandeffect
 The coefficient of determination and explained variability
 Outliers and their influence on correlation coefficients
 Linear and curvilinear relationships
 Correlation and independence
 The subjectivity nature of adjectives used to denote relationship
strength
 A Summary of the Abbreviations & Symbols Associated
with this Chapter
