TOWNSEND, Del. – On a recent afternoon at Townsend Elementary School here, a little boy squinted at a computer screen and gripped his mouse. He was stuck. Half of the screen contained an article about rainforests. The other half was filled with questions, some multiple-choice, some not.
One question asked the boy to pick two animals that belonged in the rainforest from a list of pictures and written descriptions. Then he was supposed to drag the animals across the screen onto the rainforest background. Next, he had to move two correct descriptions of rainforest characteristics into boxes. He raised his hand.
“I don’t understand,” he whispered to his fourth-grade teacher.
“Read the directions again,” she whispered back.
Delaware is one of a handful of states that has moved all of its testing online. On a recent visit to Townsend, students were filing into the computer lab throughout the day to take tests. But if a multi-state effort to create better tests is successful, the vast majority of American schoolchildren will be taking standardized math and English tests online in three years.
Some education reformers and technology experts are hailing the move, which has the backing of the Obama administration, as a revolution. They are promising more well-rounded tests, less frequent cheating and immediate feedback for both students and teachers, as students’ answers are transmitted quickly over the Internet to states and the results are then sent back to districts.
But other educators and experts point to a host of potential problems. Shrinking school budgets could make it difficult for districts to purchase new equipment, and states that pioneered online tests have dealt with network meltdowns. Some worry that the move to online testing could take time away from learning.
More coverage
As exams move online, students spend more time testing
Online testing debacle in Wyoming provides a warning to other states
The online format allows states to give standardized tests—once just a weeklong ordeal at the end of the year—as often as four times a year. It’s an opportunity that early adopters like Delaware have already embraced.
“This is so thrilling and exciting for those of us who work with schools,” said Joe Willhoft, executive director of the Smarter Balanced Assessment Consortium, one of two groups developing the new tests. “Not only will we have the end-of-the-year test, but we will also have tests that teachers can use throughout the year that can help students.”
Townsend Elementary, which is located in the Appoquinimink School District, gives students additional computer-based tests each year that teachers say are more fine-tuned than the state exams. “It used to be testing week,” said Charles Sheppard, the principal at Townsend. “Now we just test.”
Forty-four states and Washington, D.C., plan to adopt new tests by the 2014-15 school year under a program funded by the Obama administration. The states have divided into two groups that are sharing $330 million from a federal competition to develop different versions of the online tests, which will be tied to a set of common standards established in 2010.
But states that have already experimented with online testing, including Virginia and Wyoming, provide a cautionary tale against shifting to computer-based tests too rapidly.
Wyoming switched from paper-and-pencil to online tests in 2010, but technical problems popped up everywhere. Online testing was such a debacle that voters threw the state superintendent out of office and the state sued NCS Pearson Inc., the company hired to design and administer the test. The state went back to old-fashioned paper exams.
In Virginia, by contrast, the switch to online tests went more smoothly. Over the course of a decade, Virginia expanded online testing incrementally, starting in high school and moving down to earlier grades. The state also invested nearly $650 million in new technology.
But despite its careful rollout, in 2007, nearly 10,000 students were unable to complete online exams—administered by Pearson Educational Management—after a series of technical glitches.
Bryan Bleil, Pearson’s vice president for online and technology implementation, says the company is working with states and districts to help them make the transition to computer-based testing—ensuring they have enough Internet bandwidth, for example, to handle the cyber traffic during testing times.
The company stands to gain as states contract out work on test development. In January, Pearson won a $500,000 contract from the state groups developing the tests to create a “technology readiness tool” for districts, to help them determine whether they have enough computers, for example.
The states in the two groups adopting online tests will launch them in a fraction of the time that Virginia took. And unlike Virginia, many don’t have money to put toward technology upgrades.
Maryland, which has administered science tests online for four years, plans for all of its tests to be taken on computers in three years. But Bill Reinhard, a spokesman for the Maryland State Department of Education, says his state has asked test-makers to keep paper-and-pencil exams as a backup. “We don’t have enough hardware,” he said.
Kayleen Irizarry, assistant superintendent for elementary and secondary education in D.C.’s state education office, said glitches are “always a concern.” So next year, some schools in the city may pilot low-stakes exams on computers in preparation for the district-wide launch.
Yet the test developers hope that eventually, technology in schools will improve enough to allow for more challenging and stimulating tests. In these new exams, a student might be asked to use a mouse to move the sides of a shape on screen into an isosceles triangle, highlight the main idea of a passage, or write an essay about two articles supplemented by their own online research.
In Delaware, however, the rainforest question, where students simply click and drag their answers across the computer screen, is “as adventurous as we’ve gotten,” said Michael Stetter, the state director of accountability resources.
Even if the move to more sophisticated tests takes a while, advocates for the new online exams point to other benefits. If a roomful of students switches a wrong response to the right one on the same question—suggesting someone might be coaching them—the computer can easily flag the pattern as possible cheating.
“The big blowups we’ve had with cheating, it’s just not going to happen,” said Doug Levin, executive director of the State Educational Technology Directors Association.
The shift to computer-based testing also corresponds with a push to make students digitally literate. And instantaneous scoring by computers will allow teachers, students and parents to see test results right away, rather than having to wait weeks or months after the school year has ended.
Don Davis, principal of Brick Mill Elementary, in Delaware’s Appoquinimink district, has mixed feelings about the tests, including whether they might widen the achievement gap for low-income students who don’t have computers at home. But, he said, “It’s better than what we used to have.”
A version of this story appeared in The Washington Post on June 25, 2012.
“…we will also have tests that teachers can use throughout the year that can help students.”
So these tests are going to “help students”?? You haven’t spent much time around struggling elementary school kids, huh? These tests doing more good than harm is just not going to happen if (for starters) we don’t get rid of the high stakes surrounding them which distort everything about them and bring lots of collateral damage.
They aren’t even going to be very useful for “informed instruction” because teachers and students aren’t allowed to see the questions and discuss the students’ actual individual responses. (“Oh, but we tell you which standards are associated with the questions the student got wrong.” But then I have to come up with my own examples to see the students’ thinking in action, don’t I? And that’s really what I needed to do from the beginning, as the education professional in the room…) The tests are frankly only “useful” (as in basically “used for”) to assign grades and ranking to students, teachers, and schools. (…oh, and to provide the ed marketeers lots of opportunities to make plenty of profit from the captive audience which is our nation’s schoolchildren.)
Look closely at the boy in the top picture. Does that look like the face of someone who is highly motivated about this whole exercise? Do you think students learn more when they love school or hate it? (Here’s a hint: I’ve been a tutor for a couple of years; and the biggest obstacle to learning that I have come across is the individual student’s damaged attitude wherein they spend all their time with a subject thinking, “I stink at this!” and “I hate this” rather than about the material itself.)
And, as far as literacy goes, you can “reading strategy” and “targeted instruction” the students until you are blue in the face; but unless you have buy-in from those students, they will not become proficient readers. You do not become a good reader unless you actually spend time reading–lots of it–regardless of what kind of instruction you get. (And I hope I’m wrong, but I believe the Common Bore is going to kill whatever fragile motivation young students bring to the party.)
Everything takes time to become adopted and technically perfected. Technology isn’t going anywhere.
Also, teachers are probably not giving it their best shot, in protest. Educators are in a state of fear that technology is going to replace them. Unless they grasp it, it will.
I work with a computer based testing software company and have found that using computer based exams help meet the diverse needs of exam administrators in the academic, certification and licensing realms. Computer based tests provide fair and easy access to everyone. Moreover, enabling the use of computers would improve readability, thereby simplifying the grading process and deter cheating.
So . . . how are students with visual and/or physical disabilities going to do these online tests, particularly ones where they have to look at pictures, and click and drag information into boxes as described here? How do we know we’re testing content and not ability to use technology? What adaptations are being designed to make these tests accessible for all? Or is that going to be an afterthought? In our rush to put everything online, let’s not forget universal design for learning, and the need to make sure of what is being assessed, and how the results of those assessments will be used.