Nevertheless, when the legislature presented Johanns with a bill to pay for uniform testing statewide, he vetoed it. Johanns felt that standardized testing might be inevitable, but that the same test in every one of the state's schools was not. Instead, his administration came up with a plan to allow each of Nebraska's 537 districts to choose among five different state-approved tests. Those can be used for up to a third of the total assessment of a student's performance. The other two-thirds is completely up to the local school board and administrators. Some schools used classroom projects or tests created by their own teachers; other districts designed tests drawn from so- called test banks--collections of thousands of questions, from which a small percentage are asked in a given year. Out-of-state consultants go over each of the testing systems to make sure they are sufficiently rigorous.
It's a somewhat cumbersome system, but the state has been happy with it. After the first year, the outside consultants reported that in 80 percent of the cases, the locally designed tests were as good as or better than anything school systems in other states had come up with. "Not only did we take local control as the guiding principle," crows Nebraska's education commissioner, Doug Christensen, "they're doing better assessment than anything in the market." You don't hear many calls around the Nebraska legislature anymore for a single uniform statewide test.
Nevertheless, the state may have to create one. The federal Education Act that became law in January requires that all states test students in reading, writing and math every year between grades three and eight if they want to receive federal money. Johanns and Christensen have been warned that their menu of five different tests, plus the homemade local assessments, may not meet the new standards. If Nebraska gives in, however, it will not go quietly. "Over my dead body will we co-opt our system," Christensen says. "We'll do everything we can to try to integrate the federal law into our system, but there are things we are just not going to give on."
Not many educators around the country are as blunt as Christensen, but quite a few of them feel the way he does. Uniform testing continues to be embraced warmly by lawmakers around the country (whose only serious concerns tend to focus on the cost of the federally mandated tests), but educators and an increasing number of parents believe that the rush to standardize may discourage valuable alternatives being developed at the state and local levels. And there are fears that the testing requirements in this year's federal law will work to the detriment of subjects not covered by the tests and may penalize students whose skills and challenges are more complex than a standardized system will be able to measure.
Nebraska may be the rare state that is openly rejecting the prevailing wisdom on tests, but there are guerrilla-style protests in diverse parts of the country, from Scarsdale, New York, where parents have organized boycotts to keep a majority of students home on testing days, to California, where some teachers being rewarded with bonus checks for producing good test scores are refusing the money, handing it over instead to nearby schools that are struggling. "What lawmakers want are measures to show how well the schools are doing," says W. James Popham, emeritus professor of education at UCLA. "Most people who work in the legislative arena assume, incorrectly, that the standardized off-the-shelf tests will do this, but sadly they don't."
A recent Arizona State University study of results in 18 states reached the same conclusion. "If the intended goal of high-stakes testing policy is to increase student learning," the study reported, "then that policy is not working. While a state's high-stakes test may show increased scores, there is little support in these data that such increases are anything but the result of test preparation and/or the exclusion of students from the testing process."
Not that there are many educators who reject the idea of testing per se. "If you believe in performance, you've got to have a way to measure it," says Maryland state Senator Barbara Hoffman, herself a former teacher. "Standardized testing may not be the best way, but it's the most efficient way to figure out whether students are learning and teachers are teaching." The tests are, if nothing else, an excellent method for finding out whether a school, or an entire system, is falling apart. A school is either keeping up with its neighbors, or it's not. No one disputes that a district with a high percentage of failure needs attention, and a student scoring in the bottom few percentiles is in desperate need of help.
"It is a way of measuring teacher performance," says Krista Kafer, a senior policy analyst at the Heritage Foundation. "Why is it that one school with very similar characteristics is performing better than another? Testing really shines a light on what's going on."
Interpreting the test scores, however, is not as simple as it sounds. "The whole process of testing and the science of testing are as complicated as school finance," says Texas state Representative Scott Hochberg. What does it mean to have a majority of students receive a passing grade? The answer to that question depends, among other things, on where the bar is placed. Texas set a minimum "cut score" of 70 on its reading and math tests, but that doesn't necessarily mean passing students got 70 percent of the questions right. The number of questions a student has to answer correctly varies from year to year, depending on the difficulty of the questions. States have not shown themselves averse to lowering the passing score if they are unhappy with the number of students who are passing. "We need to make sure that states are reporting in a way that's clear and usable for parents," Kafer says. "One of the things that's telling is that every state has a different way of defining a 'failing school.'"
The Lone Star State was a pioneer in standardized testing. Measuring the performance of schools as though they were businesses was a pet project of Ross Perot in the years prior to his presidential campaigns in the 1990s. George W. Bush applied many of Perot's ideas and other accountability schemes during his six years as governor, then turned them into a signature piece of domestic legislation as president. In fact, the new federal requirements bear a striking resemblance to the Texas system, which means that the Texas experience offers several useful lessons for other states to think about as they revamp their systems to meet the new requirements.
Texas has generally been free of scandal; it has not been plagued by the sorts of cheating disclosure that have been widely publicized in Maryland, Kentucky and several other states. But Texas has also shown that in pushing students to achieve a fixed passing score, there are plenty of ways to game the tests. Some Texas fourth graders, for instance, are told that when they come upon a challenging passage on the reading test, they shouldn't even try to complete it. Instead, they should find and circle the most important nouns in the test question, and then simply go back to see whether they can find those nouns in the passage. This produces a fair number of correct answers, but it doesn't do much to help students improve their reading skills.
Some teachers report spending as much as two months of their school year on standardized tests--not just administering them, or assigning exercises on material similar to the expected test questions, but offering detailed instruction in basic test-taking techniques. They want to make sure, for instance, that students know their scores will be better if they take a guess on each question, rather than leaving some of the bubbles blank. Sometimes it seems as though the schools are training the next generation of game show contestants.
Of course, not all classroom time, even in the most test-obsessed schools, is being given over to test-taking techniques. Students are receiving legitimate instruction in the subject matter they'll be tested on. What's not so clear is how much help they are getting in subjects that aren't included on the tests. Teachers of non-academic subjects, from art and music to agriculture and vocational training, protest that they are losing out in funding because they haven't been given a place on the standardized menu. During the most recent legislative session in Texas, science teachers lobbied to have their subject added to the standardized tests in the elementary schools. The association won approval of a fourth grade test in science, but at the expense of the eighth-grade science test that used to be given in middle school.
Meanwhile, in Austin and other state capitals, test scores have become the dominant measure of how well schools are doing. Texas has abolished tenure for principals, judging their performance, and that of district superintendents, almost exclusively on the basis of whether their schools are producing acceptable test scores that are trending upward. Graduation rates are no longer considered a major factor.
Nor are the quirks that often result from the focus on testing. For example, one of the bright spots in Texas is performance in 10th grade. It has been going up nicely. On further examination, it turns out this is largely because ninth-graders who fail even one course can be asked to repeat the entire grade. Many of the students who are being held back don't stick around for that intensive year of test preparation--they drop out. By the time the testing takes place in the higher grades, they are not there to depress the scores. In Houston, for example, there are about 20,000 kids who enter the ninth grade each fall, but only about 8,200 who reach graduation. The scores look better than they used to, but it's hard to see how the students themselves benefit. "It's the opposite of accountability," says Linda McNeil, co-director of the Rice University Center for Education. "These kids are failing, and nobody is being held accountable."
There is an irony to this in that one of the main arguments put forward in favor of standardized tests is that they level the educational playing field. The impetus--even the original title--of the new federal law is derived from the promise to "leave no child behind." And to a certain extent this promise is being kept: When a school performs poorly--thus counting against its district's totals-- better teachers and more money are often sent there.
Within a given school, though, the worst-performing students may be getting less help than ever. Many of the recent state laws impose dire consequences on a local school system if a certain percentage of students in a given school fail to earn a passing grade. In Virginia, for example, if fewer than 70 percent of the students in a school pass the state's Standards of Learning tests, the school risks losing its accreditation. In schools that have a pass rate of, say, 65 percent, this has led to extraordinary efforts to push that last 5 percent over the top. That handful of students gets booked into extra classes in the test subject areas they're having trouble with.
But that type of extra attention isn't paid to everyone. The kids who appear to have little or no chance of passing aren't being offered more help. "What you have is a situation where 30 percent of your kids are expendable," says Shelley Bryant, a teacher who sits on the school board in Staunton, Virginia. "If 30 percent of the kids don't pass, it's no big deal." In some cases, she says, children are being categorized as "special needs" students so they won't count against the 70 percent requirement, even when there's no previous sign that something is wrong with them.
All of these problems and complexities have contributed to the nascent backlash against standardized testing, and they are certain to make the tests themselves into a huge, emotionally freighted issue in the years to come.
Nebraska may turn out to be one of the important initial battlegrounds. Johanns and Christensen continue to insist that their patchwork system is actually a much better evaluator of student performance than any uniform structure. Johanns recognizes that Nebraska's approach is much more unwieldy than a single-test system, and he admits its success has been made easier by his state's small population and enviable class sizes.
But Johanns also is convinced that Nebraska is doing a better job than other states of communicating student performance to all interested parties--students, parents, teachers, districts and the state--and developing better means of making improvements to make sure they learn more. "It's proved to me over and over again that we should not overreact and join the national bandwagon," the governor says, "if what we're doing here is getting good results."
