Skip to Content

Support MinnPost

Students -- and schools -- sweating high-stakes tests as critics seek better ways to measure results

They don't call them high-stakes tests for nothing. This week, tens of thousands of students are bent over their desks, No. 2 pencils sharpened and brows furrowed, sweating — really sweating — the Minnesota Comprehensive Assessments.

For a shocking number of them, the point of anxiety isn't whether they can fill in the right bubble to demonstrate proficiency at reading, math and science. It's what's at stake for their school or, worse, a beloved teacher.

The tests, popularly known as the MCAs, are Minnesota's way of complying with the controversial federal No Child Left Behind Act, which mandates that schools show continual progress in test scores. If enough kids meet minimum standards, their school will escape unscathed. If too many do poorly, it could face sanctions.

Also, lots of Minnesota pupils have learned to expect an additional battery of exams delivered in the fall and again right before school gets out. That's because the MCAs in the past haven't measured other needed data, such as how much individual children learn each year, where they excel or what they need help with.


For some schools, tests eat up full two weeks
For some kids, this means two full weeks of each school year are spent taking tests. And that's not counting prep time which, in struggling schools, can go on for months and yet result in nary an uptick in scores.

Educators, pupils and parents alike are to be forgiven, then, for the ripple of disappointment that has greeted the news that the Obama administration's keenly awaited fix for the mess is … another test.

The U.S. Department of Education has promised $350 million to up to two groups of states to develop computerized "growth-model" tests that would measure individual students' year-over-year learning and yield data that could be used to gauge how well teachers, schools and even teacher-training programs are performing. In addition, the tests must show whether student learning meets "Common Core Standards" that 48 states are working to establish.

Under the terms of the competitive grants, "consortia" of 15 or more states have until June 23 to apply. The Education Department hopes to award two "comprehensive assessment systems" grants of $160 million each, as well as a $30 million grant for the creation of so-called end-of-course tests for high-schoolers. Potential applicants from across the country will meet in Minneapolis on Thursday to hear details.

Seven years after NCLB's implementation, however, there are plenty of people who don't care what those details are. Increased use of standardized testing has failed to spark broad improvement in student performance or stimulate progress in struggling schools, they say.

The Bush administration reform called for every student in the nation to demonstrate proficiency in reading and math by 2014. Obama's proposed overhaul of the law would essentially eliminate that deadline but would keep the emphasis on tying teacher pay and school autonomy to test results.

One principal fights back
Steve DeLapp is the principal of Barton Open, one of Minneapolis' most sought-after schools, where students consistently post great tests scores. Irked that his staff had to stop their normal lessons several times a year to prepare kids for testing, he started needling district brass a few years ago for permission to drop some of the tests.

"The state tests are enough," said DeLapp. "I mean, they take up four different days for kids."

All school districts are bound by state law to administer the MCAs, so DeLapp proposed dropping the second test, which is known by different acronyms in different locales. In Bloomington, where it was developed, it's known as the CALT. In Minneapolis, and most other districts, it's called the MAP.

Students take the tests on computers, which are programmed to respond to correct answers by making problems harder. In this way, instead of determining, say, whether they have mastered a list of things third-graders should know, as the MCA does, teachers and parents get specific information about where each pupil is.

The so-called growth-model tests have a number of advantages. Most notably, they deliver thorough data on the highest- and lowest-achieving students, whose needs and particular challenges don't show up on more conventional tests.

DeLapp, though, is unmoved. "The best teachers, they don't need that test to know which kids are struggling as readers and which ones aren't," he said. "If they are looking at a standardized test to tell them what a kid knows, they are in trouble."

Better, in his view, to concentrate on the school's own curriculum. "Our stance as a school has been that the best preparation we can give our kids for whatever standardized test they encounter throughout their lives is a good, authentic, rigorous education," said DeLapp.

He stopped administering the MAP — and the sky didn't fall.

His criticism — that "teaching to the test" is a poor substitute for good instruction — is a common one.

"I always grind my teeth when I hear about anyone prepping for the test," said Jim Angermeyer, director of research and assessment for Bloomington Public Schools and the person who developed the CALT.

He agrees with critics who say it shouldn't be used to evaluate teachers and schools. "On the surface, it seems more fair," he said. "But really, the truth is some kids grow more than others. If you have a class with two or three particularly unruly kids, that's going to affect everyone's score."

Critics not pleased with reform plan
Like other critics, Angermeyer is concerned that the reform plan touted by Obama and Secretary of Education Arne Duncan replaces sticks with carrots without addressing fundamental issues raised by NCLB's failure, such as the appropriateness of using student tests to rate teachers. Just as George W. Bush modeled NCLB on a supposed miracle reform in Houston schools, Obama is basing his on Duncan's successes in overhauling Chicago schools.

Education Secretary Arne Duncan
REUTERS/Larry Downing
Education Secretary Arne Duncan

"The testing part of it isn't the evil entity — it's the political pressure," he added.

If different tests are going to be the order of the day, Minnesota is ahead of the curve. Under state law, the MCAs must be revised every few years. When Duncan announced the testing grants, the Minnesota Department of Education was already starting to revamp them.

This year, for the first time, the tests track individual student growth from one year to another. Another shift under way is toward replacing paper tests with interactive versions. In 2008, Minnesota became the first state to administer an online science exam.

Creating a test takes at least two years, according to Dirk Mattson, director of testing and assessment at the state Department of Education. There's plenty for Minnesota's test developers to do, but they can only go so far forward until they know what the feds will require and what new unified standards states are now devising look like.

"We are always making judicious decisions about how far into the future we go," said Mattson. "We are always the caboose."

Right now, plans call for Minnesota students to take standardized tests up to three times and submit only the best score starting in 2012.

That might be welcome news for failing schools desperate to escape wholesale restructuring, but it's doubtless unpleasant news for kids, and for teachers trying to cram their regular lesson plans in around testing seasons.

DeLapp plans to maintain Barton's insistence on teaching its curriculum and assuming its students will do just fine on whatever tests he is forced to administer. "That's been the culture of this school," he said. "Keeping the focus on learning — it gets harder and hard to do that in a testing culture."

For his part, Angermeyer hopes parents realize that in a political system, they have some power. "Schools can only push back so much," he said. "Parents are going to have to say, 'Enough!' at some point."

Beth Hawkins writes about schools, criminal justice and other topics.

Get MinnPost's top stories in your inbox

Related Tags:

Comments (7)

"And that's not counting prep time which, in struggling schools, can go on for months and yet result in nary an uptick in scores."

Did you make that up? Or, can you back it up with, like, you know, a fact or statistic or quote?

If I am recalling correctly, a recent report by the McKinsey and Co. found that the world's highest performing school systems, e.g. Finland, are using relatively few or no tests to determine the educational system's effectiveness; contrast this with our emphasis on testing, testing, testing. It makes me wonder what other differences there might be between those high performing school systems and our low performing school system, and why. (Actually, many practices of these high performing school systems are enumerated in the above mentioned McKinsey report.)

The reality is that we do not have an education 'system'; it is dysfunctional; the individual sub-systems are working at odds with each other, similar to having a car with a Rolls-Royce engine and a Corolla transmission with Mack truck wheels; its no wonder we're not getting anywhere.

I share Mr. Angermeyer's concern regarding government dictating education policy. I believe it is important for parents to be able to express their educational preferences, preferably by choosing the school and curriculum their children will attend.

Taking the Brookings Institution's recommendations for a web site with individual schools graded for their willingness to entertain parent/student choice of schools one step further, I suggest also including indexes of the ability of students to choose or design their own curriculum; i.e. democratic education.

Hank, antecdodentally

Hank, We could be like Fox and make up the news for example you just quoted a sentence, it must have some validity then right? Actually just having some fun but myself as a reader and as a parent I have heard from 2 or more different individuals at 2 seperate times that a lot of time can be spent in prep.

Steve DeLapp is right that his school should focus on the curriculum and not the test, but he has spent too much time in a high-functioning school if he doesn’t see the overall value of testing. Perhaps his teachers don’t need the structured and normalized feedback on the effectiveness of their work, but there are many schools that need more of this information to become more effective.

Minneapolis schools have used the CALT, and now the MAP, because, as the story mentions, these tests measure growth, whereas the MCA has not. But Minneapolis has also used the CALT/MAP in the fall to accurately identify exactly which kids are “in the yellow zone” of achieving a threshold MCA score the following spring. Interestingly, the schools completely disregard preparation for the CALT/MAP, but have an almost religious preparation for the MCA tests.

What this means is that all of the hype and stress and teaching-to-the-test is really just focused on a few “borderline” kids in an effort boost the score of the school. The MCA tests are essentially inconsequential to kids who are far from the borderline in either direction.

Public schools need to see testing data as information that can help them improve, rather than seeing it as just another hoop to jump through. Perhaps DeLapp, and Finland, have high-functioning schools because they have a focus on creating curriculum for all students, rather than an obsession with a few tests scores, which leaves most children behind.

Diane Ravitch, former assistant sec. of ed. for research under George Bush, is disillusioned with No Child AND with charter schools, now imagined to be magical improvement-factories as compared to regular schools but in reality not achieving higher results.

Nor, apparently, is Secretary Arne Duncan the miracle-worker he is portrayed to be or the "improvements" in Chicago's schools nearly as impressive as they are said to be. See blogs.edweek/high-school-connections/2009/07/chicago_miracle_or Chicago_mir.html for Catherine Gewertz's "Business Group Gives Chicago Schools a Poor Grade" (with a link to the group's report).

I believe Mr. Duncan may be part of a national effort to privatize the public schools insofar as possible by replacing them with non-public-school charters (check Detroit) and to kill the teachers' unions. Thanks, Mr. Reagan and all your followers including George Bush and now, apparently, Barack Obama.

Minnesota should take not a penny from Duncan's program lest it do more harm than good.

I'm no expert on the subject of testing, but I thought it was interesting that this article featured Barton Open as an example of a school that declines some District testing, and still is a sought-after school with great test scores. Barton is a low-poverty school in a high-poverty district, and therefore I'm not sure it should be held up as an example of why or how much to test.

I'm not defending NCLB as a whole, but by demanding that all schools disaggregate their test scores, we get some useful information regarding progress made by ALL students at a school, not just the majority. For example, Barton's small subset of students in poverty perform markedly lower on these tests than students in poverty at other, higher-poverty schools. The No-Child-Left-Behind laws challenges a school move EVERY child along, no matter who they are or where they come from.

It seems to me we need to do a lot more than look at a middle-class school's test scores, hold it up as a great school and an example of why we don't need to test so much. I want to know what that successful, high-poverty school is doing, and learn from them. They are the schools, teachers and staff that are doing the hard work of the District. Do they think they need less testing? Let's ask them.