18. The results for the amount of violent television watched and the amount of violent behavior toward their playmates for four young children are presented here, including the means.

Child’s Code Number Weekly Viewing of Violent TV (hours)

Number of Violent or Aggressive Acts Toward Playmates G3368 14 9 R8904 8 6 C9890 6 1 L8722 12 8 M 10 6

(a) Determine the linear prediction rule for predicting violent acts from watching violent television, and

(b) draw the regression line. Use the linear prediction rule to figure the predicted number of violent acts for children watching the following number of hours of violent television:

(c) 7,

(d) 9, and

(e) 11.

(f) Determine the standardized regression coefficient, and (g) explain what you have done in all the previous parts of this question to someone who understands the mean and deviation scores but does not know any more about statistics.

(h) Add the dots for the pairs of scores for the four children to your figure from part

(b) and also add a regression line for the linear prediction rule Yn = -2 + 1.2021X2. (i) Figure the sum of squared errors for your linear prediction rule from part

(a) and for the prediction rule Yn = -2 + 1.2021X2.
(j) Explain why the sum of squared errors for your linear prediction rule from

(a) must be lower than the sum of squared errors for any other linear prediction rule. (k) ADVANCED TOPIC: Find the proportionate reduction in error (using SSError and SSTotal).