Let's probe the data at the state level, and take a peek at SAT testing and education spending.
Across the country, 46 percent of high school graduates took the SAT in 2009, but that percentage varied enormously by state. Generally speaking, students in the Northeast were more likely to take the test (with highest participation rates in Maine at 90 percent, and New York at 85 percent). Students in the Midwest/Plains area were least likely to take the test (with lowest test-taking rates in Iowa, North Dakota and South Dakota of 3 percent).
On the other hand, students in the Midwest/Plains area had the highest scores on every single section.
Maybe students in the Midwest are really that much smarter than East Coasters. But my guess is that this is a case of selection bias.
In New York taking the SAT is expected of almost all students (85 percent). That means the highest-achieving students taken the test alongside the lowest-achieving students. But perhaps in areas of the Midwest, where the standardized test is less pervasive, only the very brightest students take the test. That means the average score for those areas will look higher.
Or maybe in some states the testing centers are fewer and further between, so only the students who live near testing centers have the option to take the SAT. Or maybe more students are encouraged to take an alternative standardized test like the ACT instead of the SAT in these states. Whatever the reason, the great variation in participation rates among states should make anyone wary of discussions of “national trends” in SAT scores.
As we’ve covered before, New York spends the most money per student, while Utah spends the least.
As always, though, correlation doesn’t necessarily mean causation.
No comments:
Post a Comment