This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Chapter 9: Interpretive Aspects of Correlation and Regression PSY 201 2/20/08 Chapter 9: Interpretive Aspects of Correlation and Regression • Defined correlation as the measure of the degree of the relationship between two variables . • Assess this relationship by determining how each pair of scores (on X and Y) deviate from their means. ( 29 ( 29 Y X S nS Y Y X X r å = ( 29 n z z r Y X å = • Therefore, the size of a correlation really depends on whether each pair of scores have similar deviations from the mean. • Example in z score units: Subject z x z y 1 .23 .23 2 1.54 1.54 3 1.25 1.25 • To the extent that this relationship holds for all pairs of scores, r will = 1.00 R as Slope of a Line • Y = bX + a • b = amount of vertical change/amount of horizontal change • Have seen that z’ Y =r z X, where r = b • When expressed with raw scors, b ≠ r • b = r(S Y /S X ) • Unstandardized Regression Coefficient (B) R and Standard Error of Estimate • When correlation is high, standard error of the estimate is low • Therefore, can express r as the ratio of S yx to S y 2 1 r S S Y YX = 2 2 1 Y YX S S r = • If S...
View
Full
Document
This note was uploaded on 04/17/2008 for the course PSY 201 taught by Professor Arthur during the Spring '08 term at Purdue.
 Spring '08
 Arthur

Click to edit the document details