In Exercise A.47, we used the percent of the population graduating high school to predict the percent to graduate college, using data in US States. It is likely that the mean household income in the state might also be a reasonable predictor. Figure A.16 shows a scatterplot with regression line for these two variables.
(a) Describe the scatterplot in Figure A.16. Is there a linear trend? Is it positive or negative? Are there any really obvious outliers?
(b) Use Figures A.15 and A.16 to decide which variable, percent to graduate high school or household income, is more strongly correlated with percent to graduate college.

Figure A.15 and A.16

(c) For the state with the largest positive residual, estimate from the graph the household income, the percent graduating college, and the predicted percent to graduate college.
(d) For the state with the largest negative residual, estimate from the graph the household income, the percent graduating college, and the predicted percent to graduate college.

Exercise A.47

The dataset US States includes information on the percent of the population to graduate from college (of those age 25 and older) for each US state. Figure A.7 shows side-by-side boxplots for percent of college graduates by region of the country (Midwest, Northeast, South, and West.)

Figure A.7

Transcribed Image Text:

45 40 - 35 - 30 25 20 80 86 90 78 82 84 88 92 94 High School College 45 40 - 35 30 - 25 - 20 30,000 40,000 50,000 60,000 70,000 Household Income College