URSI 403--Doing Urban Research -- Lecture Notes 8

VII.Parametric Tools

A. Correlation & Multiple Regression

1. standardized covariance (covariance expressed in terms of standard deviation units in x and y)
2. generally, interval measures required (may use ordinal if underlying interval relation can be assumed)
3. correlation may be positive, negative, or zero
4. square of correlation coefficient may be interpretted as PRE (proportional reduction in error) measure
5. measure is fairly robust (tolerates violation of assumptions)
6. highly affected by extreme values (outliers)
7. correlations may differ, even if slope does not
8. does not indicate causation: spurious correlation is possible, even likely
9. multiple regression extends correlation to multiple variables with additive effect on dependent variable (y=a=bx1+bx2...)

B. Factor Analysis

1. orderly simplification of interrelated variables
2. uses correlation to group variables into common (highly inter-correlated) factors

MSU