A Test of Standardized Tests

SHOCKING NEWS!!!!!! The federal government has released an interesting report. That would be a Department of Education study, which reveals (or at least seems to ...
by | June 13, 2007
 

Test SHOCKING NEWS!!!!!! The federal government has released an interesting report.

That would be a Department of Education study, which reveals (or at least seems to reveal) which states are dumbing down their standardized tests under No Child Left Behind (NCLB).

The report presents the percentages of students meeting proficiency standards on their NCLB tests (which states design) compared to the percentage who are proficient on "the Nation's Report Card" -- the National Assessment of Educational Progress (NAEP). The subtext: If students in a state are scoring much higher on the NCLB tests than they are on NAEP, then that probably means the NCLB tests are too easy.

The federal report looks at scores for fourth grade reading, fourth grade math, eighth grade reading and eighth grade math. To make the data more digestible, I've averaged all these scores and then found the difference between the NCLB average and the NAEP average.

You can see the full results below. The five states where the gaps between the two tests are largest are Tennessee, North Carolina, West Virginia, Georgia and Mississippi. In Mississippi, for example, 88% of fourth graders scored proficient on the state reading test, but only 18% did on the NAEP reading test.

Without more information, these states shouldn't be judged too harshly. There are some inadequacies in the data (for example, only 40 states were included in the federal study and of those 40, some had only partial results) and it's also worth noting that all these scores date from 2005.

What's most interesting to me is the question this report raises as Congress considers renewal of NCLB: Should the federal government design the nation's standardized tests?

If you're looking for a law that reflects a coherent philosophy on federalism, NCLB isn't it. The feds decided how often schools should test their students, what demographic groups they should be concerned about, by what year schools had to show improvement and what consequences they would face if their students didn't score well enough.

But designing these all-important tests? That was left to the states. As a result, some feel that states have an incentive to make their tests as easy as possible, to avoid all the consequences the law stipulates for failure.

One solution: Require every state to use NAEP or something similar. That's what, in response to the Department of Education report, the Washington Post opined in favor of this weekend, declaring, "America can no longer afford the quaint tradition of each state defining success differently." (Secretary of Education Margaret Spellings disagreed in an op-ed.)

However, the report could also provide fodder for a very different conclusion: That NCLB has led to "teaching to the test." In most states, the gaps between NCLB scores and NAEP scores are like kids in algebra class: yawning.

One explanation is that the NCLB tests aren't inherently easy, but rather that schools have gotten really good at teaching students how to pass them, without actually making their students much better educated. From that, some will conclude that standardized testing produces young people who can pass standardized tests, but who aren't prepared for the real world, a lesson that would imply that the federal government shouldn't reauthorize NCLB in anything resembling its present form.

In other words, this study can be used to argue, quite plausibly, that the federal government should take on a bigger role in education or a smaller one. And folks wonder why Congress never gets anything done.

State              Difference between NCLB proficiency and NAEP proficiency

Alabama No data Alaska 45.25 Arizona 37.5 Arkansas 21.25 California 22.66667 Colorado 49 Connecticut 37 Delaware 38 Florida 29.5 Georgia 52.5 Hawaii 14.25 Idaho 48.75 Illinois 33 Indiana 38.5 Iowa 41.75 Kansas 40.5 Kentucky 25 Louisiana 39.5 Maine No data Maryland 37.75 Massachusetts -2.33333 Michigan 33.5 Minnesota No data Mississippi 52.25 Missouri -0.5 Montana 45 Nebraska No data Nevada 26.5 New Hampshire No data New Jersey 36 New Mexico 23.25 New York 32.5 North Carolina 54.25 North Dakota 36.5 Ohio 34.75 Oklahoma 48.5 Oregon 31 Pennsylvania 29.5 Rhode Island No data South Carolina 2.75 South Dakota No data Tennessee 62 Texas 45.25 Utah No data Vermont No data Virginia No data Washington 31 West Virginia 53.75 Wisconsin 43.5 Wyoming 5

Josh Goodman
Josh Goodman  |  Former Staff Writer
mailbox@governing.com

Join the Discussion

After you comment, click Post. You can enter an anonymous Display Name or connect to a social profile.

More from View