SOCI 328 Lecture Notes - Lecture 17: Linear Regression, Empirical Relationship, Standard Deviation
Document Summary
There is a relationship between x and y. Have plausible explanatory rationale for causal link. Direct: x causally influences y directly (x-> y) Indirect or chain causal relationships: x1 causally influences x2 which then causally influences y. Spurious relationships: third variable x2 causally influences both x1 and y, where empirical relationship exists between x1 and y but relationship is not causal. Statistical interactions: x1 influences y differently for diferent values of x2, also called conditional or moderating relationship. Multiple causality: x1 and x2 = distinct effects on y. B = slope of regression line: positive value of b = positive slope, b = change in y for unit change in x. If x increases by value of 1, y changes by an amount of b. R^2: proportion of total variability in y that is explained by x. Beta = standardized regression coefficient for x: represents change in y in standard deviations for one standard deviation change in.
