SOCIOL 3FF3 Lecture Notes - Lecture 12: Regression Analysis, Interval Ratio, Scatter Plot
7 May 2019
School
Department
Course
Professor
Document Summary
Regression allows us to make predictions about a certain outcome, based on sample data. We try to minimize error by estimating ordinal least squares (ols) regression analysis: optimized prediction that reduce the sum of squared differences btwn observed and predicted. April 9 values: x predicted variable, can only be used w interval ratio variables, bivariate: one ind variable, multivariate: more than one ind variable. Linear relationship a relationship btwn 2 interval ratio variable in which the observations displayed in a scatter diagram can be approx. w a straight line. Short for least squared error: reduce difference btwn the data point (observed values and line of best fit) X whatever value of x we are interested in. Y predicted scores of the dependent variable. Calculating y intercept (a: rearrange, = ybar b (xbar, use the means of x and y bc u know these coordinates actually exist.
