SAT, take 2 Suppose the AP calculus students complained and insisted that we should use SAT math scores to estimate verbal scores (using the same data from exercise 47). Here is the regression analysis of Math SAT vs. Verbal SAT. Dependent variable is Verbal s = 72.77 Variable Coefficient Intercept 171.333 Math SAT 0.6943 a) What is the correlation? b) Write the equation of the line of regression predicting verbal scores from math scores. c) In general, what would a positive residual mean in this context? d) A person tells you her math score was 500. Predict her verbal score. e) Using that predicted verbal score and the equation you created in Exercise 47, predict her math score. f) Why doesnt the result in part e) come out to 500?

Week 2: Lecture 3 ● Spread: How much the data values vary around each other. ○ Range: How far apart are the two extremes. Also, sensitive to outliers. (e.g. range= max-min) ● Interquartile Range: the value that is just above the lower 25% of the data values, namely Q 1 ○ Range of middle half of the data ■ IQR = Q - Q *IQR is like median, not affected by outliers or 3 1 skewness of distribution ● Upper Quartile (3rd Quartile): the value that is just above the lower 75% of data values, namely Q 3 ■ E.g. 150 150 152 154 156 I 156 157 157 158 159. Answer: Q =152, Q =157 1 3 ● Standard deviation: measures how far the data values are f