Question: The news media have a weakness for lists. Best places to live, best colleges, healthiest foods, worst-dressed women a list of best or worst is sure to find a place in the news. When the state-by-state SAT scores come out each year, it's therefore no surprise that we find news articles ranking the states from best (Iowa in 2007) to worst (District of Columbia in 2007) according to the average SAT score achieved by their high school seniors
The College Board, which sponsors the SAT exams, doesn't like this practice at all. "Comparing or ranking states on the basis of SAT scores alone is invalid and strongly discouraged by the College Board," says the heading on their table of state average SAT scores. To see why, let's look at the data.
Figure shows the distribution of average scores on the SAT Mathematics exam for the 50 states and the District of Columbia. Iowa leads at 613, and the District of Columbia trails at 462 on the SAT scale of 200 to 800. The distribution has an unusual shape: it has one clear peak and perhaps a second, small one. This may be a clue that the data mix two distinct groups. But we need to explore the data further to be sure that this is the case
In this chapter we will learn that to understand one variable, such as SAT scores, we must look at how it is related to other variables. By the end of this chapter you will be able to use what you have learned to understand why Figure has such an unusual shape and to appreciate why the College Board discourages ranking states on SAT scores alone.
