Wednesday, June 5, 2013

Virginia ed tech is for ed testing

The high-stakes testing season in Virginia is drawing to a close. I'll probably have more to say later about this year's testing season, but in the meantime I wanted to address this Washington Post article about Virginia being a "model" for the nation when it comes to on-line testing:

All Virginia students will have to log on to a computer to take this year’s Standards of Learning tests, making Virginia one of the only states to wholly abandon the nearly ubiquitous paper-and-pencil bubble sheets. 
With spring testing in reading and math underway in many schools this week, the move to electronic tests means that Virginia, one of the few states that did not adopt national academic standards, has become a model for the dozens of states that did. Those states are scrambling to meet a fast-approaching deadline to implement corresponding online tests. It took more than a decade of school technology investments and upgrades for Virginia to get to this point.

As I have written ad nauseum on this blog, I am generally opposed to the current high-stakes testing regime. That being said, I acknowledge that on-line assessments are an improvement in principle. However, in practice, there are several problems with on-line testing and I shudder when I hear Virginia is to be a "model" for the rest of the country.

First of all, on-line testing means that testing-related computer skills drive instruction in technology. This leads to, for example, kindergarteners in Virginia being asked to practice over and over again computer skills that they are not developmentally ready to master. Activities like this are frustrating and a waste of valuable time.

Second of all, in some cases the computer version of tests is not easier to manage. For example, many elementary-aged students can not type. But nobody thinks about this before asking the fifth graders to type their writing test. Oooops. The test of writing (which has its own shortcomings) becomes a very frustrating test of typing for which there has been little preparation.

Next, there are all kinds of glitches. My county was all set to test a week or so ago--the students were hunkered down in their classrooms, their regular instructional program cancelled, when the computer system went down. They waited about an hour and then cancelled testing county-wide, delaying it until another day. Add one more day to the testing season. Fairfax County, one of Virginia's largest in terms of student population, has also experienced wide-spread technical difficulties. Remember, it's not just testing that is disrupted by technical difficulties, but real teaching and learning are disrupted for testing--for weeks at a time.

Lastly, and most importantly, the testing drives the technology that's available. From the afore-cited Washington Post article:
To help fund technology upgrades, the General Assembly dedicated nearly $60 million to school districts every year. The state contracted with Pearson, an education publishing and assessment company, to develop the online tests.
At a recent school board meeting in my county the members were getting all excited about all of the newfangled technological tools that are being used in industry and that the students in our district can learn on and about, only to be brought back to earth by an administrator: That all sounds great, but especially with severe budget cuts, remember any technology we invest in needs to be used for testing and hence has to fulfill the requirements mandated by testing. That's right, ed tech in Virginia is being driven by the Pearson-administered SOL tests and not by the tools that facilitate the best and most current learning experiences.

Just as I've said about the Common Core, or any set of national standards, the quality of educational technology and the learning that goes along with it, ultimately rests on the quality of the McAccountability system it's filtered through.

Virginia districts don't really have educational technology departments; they have on-line testing departments. Ed tech is being subverted for on-line testing.


  1. I am concerned that the new online test results are influenced by all kinds of variables other than mastery of standards. I also believe that decontextualized, high-stakes tests are the least effective measure of student learning and performance. I appreciate that the state commissioner of education in California has called for a testing moratorium to determine what makes sense and how to create balenced assessments vs. overboard testing. Thanks for sharing your perspective.

    1. @Pam: Thanks for your comment! Yes, they are influenced by other things, like background knowledge. I've heard that the ES Math SOLs, for example, have questions about restaurants and making coffee, so it's not just the math that's being tested. Most elementary-aged students are just along for the ride at restaurants, if they go to them at all, and they certainly don't make coffee.

      I agree with this that you said especially: "decontextualized, high-stakes tests are the least effective measure of student learning and performance." Decontextualized is such an apt word here.

  2. How do kids do online tests in math? Is their real problem solving or just multiple choice?

  3. @Leonie The new math tests have been explained to me before, but I'm not so good at explaining them coherently myself. They're a mixture of multiple choice and other problem solving questions. I asked Gary Rubinstein to look at some sample questions see here and tell me what he thought. He was not impressed.

  4. "I acknowledge that on-line assessments are an improvement in principle"

    I know you qualify this with that "in principle," but I still think you go too far here. Not only will on-line assessments require massive investments in equipment that, unlike books, will be useless in a few years (I still use some books from the 1960s in my classroom), they will also lead to massive amounts of data that will inevitably be abused to promote this or that "evidence-driven" reform that will squeeze more time from real educational activity! Do we have any evidence that the use of data has ever actually improved education?

    Of course I agree with everything else you say.

  5. @EC Yes, I go too far or rather I didn't go far enough in my thinking about on-line testing--I share all of your concerns. I meant simply that they are an improvement over paper and pencil but I mean for the students (with some exceptions) and for other purposes. The big picture is a different story.