As an 11th grade teacher, I play “Devil’s Advocate” a lot. On any given day, I find myself arguing from 10 different sides of the same issue. When a witty student is hip to my game and calls me on my crap show, I usually just say a standard line.
“Ok, sure, ‘devil’s advocate.’ You got me.”
Inevitably, some kid, who probably had a religious grandma, will respond with the equally popular phrase.
“The devil doesn’t need an advocate.”
That student might be right. The devil might not need an advocate, but I think standardized tests do.
I understand that this idea may frustrate some people. It is not a popular title. It is much more sexy to write an anti-test post that decries the evil of our schools producing bubble-filling robots who imagine the world can all be solved by choosing the best of four choices. I could try to write a graphic novel, where students use #2 pencils to shoot test bubbles out of the hands of stomping politicians.
If You’re the kind of person who would rather read that, try to read this to the end. It might not be as bad as you think. I’m not being paid off by big publishers or Arne Duncan, but I am trying to provide an honest interpretation from the perspective of one classroom teacher. We tend to divide our debates into rather extreme camps. In this case, you have the “opt-out” folks on one side, and the archaic, drill and kill, NCLB, proficiency or bust people on the other. As is usually the case, the correct approach lies somewhere in the middle.
As there is plenty of literature attesting to the evils of testing, I am hoping to offer a counterpoint.
Teachers use test data every day. I use it to generally gain information on the literacy skills of the students in front of me. At my best moments, I can individualize instruction based on repeated deficits that we notice in testing data. As any teacher knows, some of the best information we get on our students is from the 10 months of work we see in our class. But we need longitudinal data as well. It helps us see where students are growing, what interventions are working, and where students are declining. In my class, I can measure growth over months, but I have no way other than tests to speak larger learning patterns. Some of our measures need to be from sources outside our classroom, and some of our measures need to be standardized.
I also use these test scores to celebrate when students improve from previous years, performing in percentile levels far higher than in the past. We can also celebrate growth in certain areas from year to year. My job has never been about proficiency, but the growth that comes from working to get better. I’ve watched students cry with joy when their hard work in class is rewarded with tangible growth. And anecdotally, it almost always does. I also use that same data to notice troubling trends early. There are many times when test scores, often paired with classroom work, has led to really intense conversations about what is happening in the classroom or elsewhere that keeps students from learning.
Most teachers that I know use every indicator they have to build meaningful learning for their students. We use classroom performance, behavior, test scores, conversations and observations, and grades from previous years. Testing is certainly not the only thing to notice, but it is extremely important.
They also show us, sometimes glaringly, where we need to change. We still live in a society where academic outcomes are painfully tied to race and neighborhood. In fact, we are not much better now than we were in the 60s, when the Civil Rights Act was passed. In Pittsburgh today, 32% of black students in grades 3-11 are achieving proficiency on reading tests. 64% of white students meet that mark. If you leave the city and travel just 5 minutes north, you find districts where well over 80% of the students are reading at this level. Philly is just as alarming, as 55% of white students are reading proficiently and only 27% of black students. Equity does not primarily mean funding, or teachers, or anything else that we sometimes throw at these numbers. For equity to mean anything, it has to primarily be about outcomes. Outcomes still depend too heavily on zip code.
I’m hesitant though, to even write the paragraph above. I want to be sensitive to the way many respond to numbers representing this opportunity gap or these comparisons. We quickly jump to reactionary conclusions about schools, neighborhoods, or worse, people.
The problem was never tests, but the way we respond to these tests. To the extent that data can be used to grow, it can be also used to classify in terrible ways. It can lead educators to make really bad decisions. For some reason, we talk very little about schools and teachers using data in really productive ways that help to close gaps in opportunity, and we focus on the bad decisions that adults make about testing.
To be fair, some of there are some terrible decisions being made. We buy into the lies that we need to teach to a test, or that a score tells the whole story, or that we should avoid schools that don’t achieve “A” ratings or high school performance scores. We start to think that because students will take a multiple choice test, we have to present information in multiple choice ways. We make a relatively low-stakes test really high-stakes by the way we handle the information. We scare kids into test anxiety and nerves.
It doesn’t have to be that way.
Sometimes I wish the world were more multiple choice. It seems more simple. I can more easily prepare for a world that falls neatly into categories of good and bad or right and wrong. But it’s not, the world is complex. One of those complexities is illustrated here. In the past, when my students have "opted out," I'm missing an important piece to help them. If standardized tests go away, I lose a valuable source of information that informs the growth of students.