Correlation and Regression

bivariate

relationship between predictor IV and criterion DV

correlation can show us the

strength, direction, shape and significance of relationships

Pearson’s correlation coefficient

-1 to +1

shows strength and direction of linear bivariate relationships

regression

prediction: plot line the correlation would have and use to predict scores on DV

needs slope and y axis intercept

slope

gradient of line

regression coefficient: how many units of Y increased for every increase in X

Y axis intercept

predicted value of Y when X is 0

Regression equation

Y = a+ (b) (X)

Y= predicted score of y

a = y intercept

b = slope

X = score on X

correlation and regression assume

normality of X and Y

linearity

residuals

difference between actual score and predicted score

line of best fit/least squares regression line

overall minimum distances from all data points

standard error of estimate

mean of residual scores = average distance from data points to regression line

SS

can separate variation in Y into 2 components

variability from error, variability from regression