STA 2122 Lecture Notes - Lecture 9: Scatter Plot, Dependent And Independent Variables
Document Summary
Pearson"s r: correlation coefficient, appropriate for scale (interval/ratio) level variables such as age, education and income, r measures the extent to which variations in one variable explain variations in another variable, types of variables, scale x scale. 1 to 1: value, 0 = no association. 1 perfect: 1 perfect, direction, + indicates positive association, indicates negative association. Strength of correlation: correlations vary with respect to their strength, the strength of correlation between x and y increases as the points in a scatter plot more closely form and imaginary diagonal line. Regression analysis: considers causal direction of the association in addition to its strength, with a real set of data, most people fall to one side or the other of any regression line we might draw through the data. Spss is able to determine the line that would produce the smallest deviations overall (from the mean: spss can calculate the least square regression line (line of best fit)
