Toward Better Standardized Testing (Accountability)

by John R. Walkup, Ph.D.


Like all summative tests, standardized tests are tools public agencies can use to monitor classroom effectiveness.

The purpose of standardized tests has been hijacked by agencies and organizations who found the massive database too tempting to avoid availing for their own purposes. This has led to numerous abuses by well-meaning organizations who think that they can solve the problems in education if only they can get their hands on the data. (And yes, the U.S. Department of Education is one of the worst abusers and its power-mad actions have stirred much of the present anti-test passions.)

As a result, test-bashing has become a popular on social media. Many blog posts have derided standardized tests for "killing creativity" as teachers "teach to the test." We hear stories of kids crying, teachers quitting, and (worst of all) corporations making money.

Some arguments have even seeped into hyperbole worthy of WWI anti-Kaiser posters. Many claim that the rich are designing the tests to force kids into private schools. Recently, a question on Twitter was raised whether the tests are causing kids to commit suicide.

Time to take a deep breath.

Accountability hierarchy

Standardized tests are not perfect, of course. Nor do they come without problems. But we owe it to school children to bend over backwards in ensuring they experience a positive learning environment.

Some schools are less adept at providing a quality education than others, necessitating the need to continually monitor educational effectiveness and, when necessary, taking corrective action. But what constitutes corrective action? The answer to that question has major implications on the pressures we place on teachers and students. I will address this issue in another blog post, where  I will explain that we're not doing it right.

Any school accountability system must take into account the layers of school governance that form the modern school system. For the public school system to operate effectively, at the very least:*

1. Each state department of education should monitor the performance and growth of individual school districts, then take corrective action.
2. Each school district should monitor the performance and growth of its individual schools, then take corrective action.
3. Each school should monitor the performance and growth of its teachers, then take corrective action.
4. Each teacher should monitor the performance of his or her students, then take corrective action.**

Ideally, this hierarchy works well if everyone does their job correctly. Unfortunately, we can't trust everyone to do their job correctly. As such, we need to modify the model.

Trust versus accountability

"Trust" is now one of the most pervasive buzz words in education. Teachers are repeatedly told to trust their students; school boards are urged to trust their teachers.

Trust-based governance naturally favors the untrustworthy. Such honor systems lead to abuses of power that, if taken far enough, can generate inefficiency and corruption.

Make no mistake about it, trust is the enemy of accountability. Therefore, trust has no place in this argument.

Modified accountability model

Because relying on trust-driven accountability is like having no accountability at all, we should consider modifying the accountability system to allow each entity to dig one layer deeper into the governance structure. This produces a modified form of the original accountability hierarchy:

1. Each state department of education should monitor the performance and growth of individual school districts and their respective schools, then take corrective action.

This goal fits with the state department of education's professional development model, which is really designed to provide support at the individual school level. Note that a statewide teacher evaluation system does not fit this hierarchy. However, a statewide school ranking system does.

2. Each school district should monitor the performance and growth of its individual schools and teachers, then take corrective action.

Teachers are district employees, not school employees. Therefore, allowing school districts to monitor teacher performance and take corrective action aligns naturally with the roles and responsibilities spelled out in most teacher contracts.

3. Each school should monitor the performance and growth of its teachers and students, then take corrective action.

This goal fits with the school's capabilities in providing intervention for individual students. Aiding schools in this regard are academic coaches and counselors.

4. Each teacher should monitor the performance of his or her students, then take corrective action.

Statewide teacher evaluation

As stated earlier, the accountability hierarchy does not support a statewide teacher evaluation system.

This is not to say that such an evaluation system is undesirable. That isn't up to me to decide.

For those states wanting a statewide teacher evaluation system, I have conceptualized a model based on multiple linear regression that overcomes many of the failings of models, such as VAM (which I think is horrible).

I will describe my teacher evaluation model in a future blog post. But if you want to see the conceptual underpinnings of the model, check out my blog post on school ranking systems.

Next up

In the next post in this series, I will counter many of the arguments offered by both sides of the standardized testing debate.

* Whether the U.S. Department of Education must monitor the performance and growth of state departments of education is a matter of states-versus-federal rights. Frankly, I would prefer they stick to what they do well (whatever that is).

*Teachers usually cannot measure growth, as they monitor student performance only within the one school year.