AnalysisPandemic  

Beyond the Scantron: Former Dallas Superintendent Mike Miles on Why Testing Is Essential to Know Where Students Are At — and Is a Key Tool in Preparing Them for the 2030 Workforce

By Anne Wicks and William McKenzie | September 29, 2020

(David L. Ryan / The Boston Globe / Getty Images)

This piece is part of “Beyond the Scantron: Tests and Equity in Today’s Schools,” a three-week series produced in collaboration with the George W. Bush Institute to examine key elements of high-quality exams and illuminate why using them well and consistently matters to equitably serve all students. Read all the pieces in this series as they are published here. Read our previous accountability series here.

Mike Miles is the founder and CEO of Third Future Schools, a charter school network in Colorado and Texas. Before starting that organization, the West Point graduate and Army Ranger served as a superintendent in both Dallas and Colorado Springs.

A Broad School graduate, Miles spoke with us about how his schools rely upon demonstrations of learning to assess students each day. The former Foreign Service officer also explains the importance of reliable, comparable and valid state exams so parents and teachers can know whether a child is learning at grade level. Those are part of the requirements, he says, to prepare students for the year 2030 — and to allocate the resources to get them there successfully.

What role does high-quality assessment play in teaching and learning? And what role should it play in accountability?

We first have to define what testing means, and whether we’re talking about formal or informal assessments. I wonder if parents and teachers would understand testing and its value more if they were to think of tests as demonstrations of learning.

We are one of the few systems that require a demonstration of learning every day. I could walk you around our schools and you would see a lesson objective and a demonstration of learning up on the wall. The demonstration of learning, which we are now doing both in-person and online, takes between five and 10 minutes in every core class, every day.

We haven’t had any parents say that is too much testing. But it does give the teacher an idea of how proficient the student is and how much they’ve learned.

Getting back to the question, there are four reasons to test. Number one is to identify student proficiency. Where is their skillset, especially in subjects like reading and math? Most parents, most teachers, and, I think, most of the world wants to know whether a student can read and at what level. We need some sort of assessment that’s not subjective to determine the answer.

A sub-point is how much did this student grow in reading? Where are they at the beginning? Where are they at the middle? Where are they at the end? Assessments help us know that.

You can’t just rely on teacher assessments for those. They won’t be calibrated, and may be too subjective. You need a formal assessment that’s calibrated nationally. Our jobs are so mobile that an assessment for reading or math should be national. At the least, it should be at the state level. A calibrated formal assessment would allow kids, teachers, parents, and everybody to know where a student is.

Number two — you need to test to assess what works. How do you know that your reading program works, or whether your reading interventionist is making a difference? How do you know whether your teacher or their method is making a difference? You won’t know if you don’t have some way to assess kids.

The third reason for an assessment is to assign or prioritize resources. This is predicated on school districts and states identifying what they want kids to know and what competencies they want them to have.

If we are agreed that reading, and reading at a particular level, is something everybody should have, then we should put some money and energy behind that. We should fund the training of teachers or fund resources for districts to prioritize reading. For example, you may hire a reading interventionist over a Social Studies interventionist. We have a reading interventionist.

The last reason, and probably the most controversial, is accountability. I don’t think it should be as controversial as it has become. I don’t avoid the “A” word. We should have accountability, and the lack of accountability has hurt our performance and hurt progress for kids.

If you hold people accountable without supports or if you’re unreasonable, then that breeds fear. It also undermines this notion of accountability. But you can have a fair accountability system with supports.

Given what we experienced with the pandemic at the end of the last school year, what could assessments tell educators, policymakers and parents once school starts this fall?

If I were a parent or an educator, I would want to know where my students are. What level of proficiency? And if it’s a lot of loss, where are they?

We don’t need to blame anybody. We don’t need to blame anything. We just need to know where they are.

I don’t know how you can put together a curriculum or plan to improve if you don’t know where students are. And we need to know that so we can assess if our interventions are working. Once you build a plan, you need to know if it is working.

With regard to accountability, I am sympathetic to kids and teachers for the disruption of the last quarter. I’m not sure it would be fair to hold teachers or students accountable like we normally do, especially since they were at the mercy of the skillset of teachers to move quickly into synchronous live teaching. I think we’re going to have to take a pass on accountability for last year’s results.

Now, testing accountability is only one small part of accountability. It should not be the be-all, end-all; maybe just a third.

In my network, we were able to hold teachers somewhat accountable because we hold them accountable for the quality of instruction all the time. The quality of instruction is probably more important with regard to your ultimate achievement.

It is easy to say that we are going to grow four percentage points. But what are you going to do to make that happen? It’s clearer and more rigorous to say teachers will have high-quality instruction, and that means X, Y and Z.

What do you wish more teachers understood about assessment? And how has that impacted how you have hired and recruited faculty?

I wish they understood that the way to grow student achievement is through these demonstrations of learning, this daily assessment of kids. If you are focused on that you don’t have to focus on the ultimate exam, which is the state test. You can have tools that monitor progress, like a district common assessment, but those still are not as important as what you do and identify in the classroom every day.

The other important part is improving your practice. If you get high-quality teachers, the scores will come. Our schools don’t do a lot of testing outside of the demonstration of learning. We leave the periodic exams to teachers. We do a fast reading assessment for our K-8 students. We do the interim Measure of Academic Progress test that Northwest Evaluation Association provides, but we don’t practice for it. And we take the state Partnership for Assessment of Readiness for College and Careers test in the spring.

But we don’t even focus on the PARCC because we know that the MAP exam is correlated. And we know that if we get almost two years’ growth on it, PARCC will take care of itself.

What we are insistent on are demonstrations of learning and high-quality instruction. Our teachers focus on that religiously.

What do you wish parents understood about testing?

Number one, I wish they understood that our network of charters is not over-testing. I suspect some districts do, but we don’t have practice exams. Teachers incorporate some PARCC examples in their coursework as we get close to the exam. But we don’t have three days of PARCC practice tests or anything like that.

Number two, I wish they would understand that the assessments that we do have are just to help identify where their kids are and assess their growth. None of the tests are punitive for the kids. We don’t use the PARCC or MAP exams as part of the student’s grade.

What role do validity, reliability and comparability play in state exams? And why do they matter?

Our mission, and I think it should be everybody’s mission, is to think about the year 2030 and the workplace needed for then and a mobile society.

If we say a student is proficient in reading that needs to actually be the case. You could have a parent or teacher say a kid is proficient in reading, but that would not hold much validity with me. States can set the bar that determines proficiency, but I would like a national bar.

That’s why the MAP test is important for us. We can see how our kids are doing comparable to kids across the nation. We are such a mobile society that we should be able to say that if a child can read in Colorado, they can read in New York, Florida or Texas.

A national bar would help the nation prioritize its resources, too, whether for schools serving students in poverty, special education needs, or something else.

Your proficiency is not reliable or valid unless you can compare it to some state or national exam. And the exam must be done in a way that will not make the results invalid. If it is not administered consistently and reliably, then I’m not sure you can trust the results.

Let’s talk more about tests helping you to allocate resources equitably. What has that looked like in your experience and why is that important as you look forward to that 2030 vision for all kids?

COVID-19 brought into stark relief the inequities inherent in our education system. We knew that before but now we know it even more. And the inequities are not just about Chromebooks, hotspots, and internet connectivity. Those are important, so I’m not discounting them. But it’s also about inequities with regard to instruction, academic resources, and programming.

One strategy that will help address equity, although it won’t fix everything, is to assign your strongest-performing teachers to your lowest-performing students. Or have a policy where no class of kids challenged by poverty can be taught by a non-proficient teacher.

You can say you are a great teacher all you want, but I’m not going to believe you’re a proficient teacher if your kids are not growing academically. And you have to know which kids are not growing academically, you have to know where they are.

That is why assessment is part and parcel of this allocation of resources, whether it’s human capital resources or more reading interventionists or a new electronic curriculum for kids who need more time at home or remotely.

As you think ahead toward 2030, what assessment innovations intrigue you? What do you think will be important?

A grievance we have all had around state testing is the amount of time it takes to do. PARCC is given over a two-week period. For most kids, you’re talking about several days.

But the MAP test only takes about 90 minutes. It’s not as comprehensive as a PARCC test or the STAAR test in Texas. But it would help if states could use technology so its process becomes much faster.

There also are fast, adaptive assessments that are software or electronic curricula. So we wouldn’t have to take a unit exam, a mid-year exam, or a district level exam. You could use these technologies to gauge where the kids are.

At some point, we might even be able to get rid of our demonstrations of learning because the technology would do the same thing for us. We could put a child online for a little bit, and peg where he or she is that day.

There are disadvantages to taking the assessments out of the teacher’s hands, but we can overcome that.

Now, we have to think about assessments in slightly different ways, depending on what you want kids to know and do. Math, reading, and science will still be really important in 2030. We will want to assess them.

But there are other things that we want to assess, and we have to get away from multiple choice or full response tests for them. This year, we are assessing a couple of things more fully than previously.

One is our Art of Thinking class. We are creating our own project-based or performance assessments for that class. A student will get a problem and then have to solve it. They will get some scenarios, and have to think through the problem, talk about it, and use their logic and critical thinking skills. We are designing assessments now for our third-thru-eighth grade Art of Thinking class.

Measuring proficiency leads us to cut scores. What should go into determining those and adjusting them?

The reason cut scores have become problematic and controversial is because they have become political. If you set a cut point for proficiency in reading, which we finally know we can do well, people then want to chip away at the cut point to give people credit for hard work, even if they don’t reach the cut point.

The cut point should be clear to everybody. For example, if proficient reading on the Texas state exam is not proficient on the National Assessment of Educational Progress, then let’s figure out why and establish a real cut point. If we want to lower the cut point for political reasons, then let’s just be honest about that. Or say that we are going to adjust the score to give you way more points for growth.

That’s hard.

It is. And we should focus on growth for communities that are challenged by poverty or language barriers. The growth in your reading ability or proficiency is more important than your starting point or your proficiency level. Eventually, you have to catch up, but if you’re growing more than a year in a year’s time, then eventually you will catch up.

Politically, that is the easiest to defend. You don’t want to punish a teacher or a student, if they grew two years in one year’s time and they’re still not proficient. There can be a lot of grace for that.

Is there anything that we didn’t ask you that you wanted to put on the record?

We have an opportunity to use the crisis to reset and get people to understand how testing or demonstrations of learning can help and must necessarily be used to close the opportunity gap. In Colorado, assessment and accountability have a bad rep, and we’ve got to turn that around.

Without testing, the kids who already have advantages will continue to have their advantages. The status quo will prevail because you don’t reprioritize and allocate resources; you don’t address the gap because you don’t know what the gap is; and you don’t address equity because you don’t know where kids are.

Anne Wicks is the Ann Kimball Johnson Director of the George W. Bush Institute’s Education Reform Initiative.

William McKenzie is senior editorial advisor at the George W. Bush Institute.

Related

Sign up for The 74’s newsletter

Submit a Letter to the Editor