《Lec 8 regression_amp;_matched_t_test》.ppt

Linear Regression Linear correlation ? Regression A perfect correlation: Imply the ability to perfectly predict one score from another Perfect predictions: Very simple, especially with z-scores: the z-score you predict for the Y variable is the same as the z-score for the X variable Perfect prediction: No prediction Prediction with Z Score When r is less than perfect, this rule does not give the best predictions The correlation between these two variables is always far from perfect, there is plenty of room for error. The lower the correlation, the greater room there is for error. Predicting with z scores Modified rule: Regression toward the mean Francis Galton’s study Tendency for unusually tall people to have children shorter than themselves ; the children were usually taller than average. This phenomenon was seen as some sort of natural pressure pushing toward mediocrity, but now scientists recognize that regression toward the mean is just a consequence of the laws of probability when correlation is less than perfect. As r decreases, predictor gets closer to z = 0 (regresses toward mean) Graphing Regression in Terms of Z Scores Raw Score Graph Exam2=1.12*Exam1+41.8 r = .909 Z Score Graph ZScore2=.909*ZScore1 r = .909 The raw-score regression Predictions based on raw scores When r = 1, Ex) height and weight; assume r = 1.0, slope = 5 Perfect r, but you wouldn’t expect a 1 lb change in weight for every 1 inch change in height We can’t predict weight with just the slope because the line does not go through the origin; we need the intercept Quantifying the errors around the regression line Quantifying the errors around the regression line Regression equation give us the straight line that minimizes the error involved in making predictions Residual: difference between actual Y value and predicted Y value: Y – Y’ It is the amount of the original value that is left over after the prediction is subtrac