The following table shows the number of 20- to 24-year-olds in the United States in various years since 1980. Assume that the N-values in the table are middle-of-the-year values. Year, y 1980 1990 1995 2000 2002 2004 2006 20- to 24-year olds, N (millions) 21.32 19.14 18.30 18.96 20.35 20.96 21.11 Source: 2008 Statistical Abstract of the United States. Find the best-fitting quadratic model for these data. (Although both cubic and quartic regression models fit these data more closely than the quadratic model does, neither are suitable models. Look at the graphs of these higher-degreed polynomial models along with the scatterplot to see why.)

Question 15 (1 point) Find the quadratic model's SSE. Only enter the number, in millions2, rounded to two decimal places. Your Answer:

16 (1 point) Find the quadratic model's average error. Only enter the number, in millions, rounded to two decimal places. Your Answer:

17 (1 point) Find the quadratic model's coefficient of determination. Only enter the number, rounded to four decimal places. Your Answer: Question 17 options: Answer Question 18 (1 point) What percentage of the variation in the data is explained by the quadratic model? Only enter the number, as a percent, rounded to two decimal places.

Question 19 (1 point) According to the model, in what year of the 1980's were there 20 million 20- to 24-year-olds in the U.S.? Question 19 options: 1980 1981 1982 1983 1984

Question 20 (1 point) In what year does the model predict that there were the fewest 20- to 24-year-olds in the U.S.? Question 20 options: 1991 1992 1993 1994 1995 Question 21 (1 point) Use the model to predict the number of 20- to 24-year olds in the U.S. in 2008. Only enter the number, in millions, rounded to two decimal places.