Testimony at PARCC Hearing: Only Test Samples of Students
Last week, the Study Commission on the Use of Student Assessments in New Jersey announced that it would hold three public hearings this week. The first one, scheduled for Tuesday, was canceled due to the blizzard. Wednesday’s hearing in Jersey City went off as planned.
Since this was the closest one to where I work, I rushed out of school to make it to the hearing by 4 PM. Unless you happen to work in Hudson County, it was anything but convenient, and this no doubt had something to do with the light attendance. Despite only having about fifty people in the crowd, the three dozen people testifying took about 3 hears to get through.
The resounding message of the hearing was this: we’re testing too often and it’s anything but productive. Commissioner Hespe complained to NJ Spotlight article, “What is missing from this conversation and what I have asked from testifiers to address is what would they do to this societal problem where half of the students are graduating without the skills and knowledge they need. We didn’t get any of that. But doing nothing is not serving our children well.”
He missed the point, though. Continuing with the current assessment system is not doing “nothing.” It’s doing something. It’s doing something that is detrimental to the education of our children as we focus exclusively on shallow test score gains at the cost of real educational opportunities.
Doing something would be as simple as scaling back or eliminating most of the tests. Tests don’t teach, and the increased frequency, length, and difficulty of our standardized assessments in the last decade has not addressed the problem Hespe mentioned. Instead, it exacerbates it. Freeing schools from the high stakes of testing wouldn’t necessarily fix things, but it would make them less bad.
Anyhow, below is the text of my testimony before the commission. Long story short: Standardized tests are only useful as an audit of the system, so there should be absolutely no high stakes attached to them for students, teachers, or schools. With this in mind, we should use sampling methodologies, like those from the NAEP, to audit the system by only testing a small fraction of our students. The federal landscape for testing is changing, as the Senate considers revisions to NCLB and ESEA, and it’s about time we held up a viable alternative to the current assessment system.
If you were unable to testify last night or tonight, you can submit written testimony to the Study Commission through its website.
Good evening members of the Commission, and thank you for giving me the opportunity to speak today about the use of assessments and standardized testing in New Jersey. I’m currently a high school social studies teacher in East Orange, where I’ve taught for eight years. I’m an Association Rep in my local teacher’s union, and I chair our government relations committee. I’m also a graduate student at Rutgers University, pursuing a doctorate in education.
The increased emphasis on testing in the last decade has led to a toxic environment in our schools, in which administrators are often focused on the shortsighted goal of raising test scores. This problem has intensified in the last few years, as the education field has been consumed by its obsession with being “data-driven.” It’s an unfortunate but foreseeable reaction to the high stakes currently attached to standardized tests. The educational process is cheapened when we are forced to focus on short term solutions to raising test scores, and we lose sight of the long term growth of students. We can’t see the forest for the trees.
On Monday morning, I was sitting in my homeroom chatting with some of my students. An announcement came over the loudspeaker about the PSSS assessment that we were scheduled to administer this week. A student asked, “What’s that?” and I explained that it was a new test for our freshmen, so that we could determine if they were on pace to be prepared to take the PSAT the following year.
I don’t know about you, but I think it’s ludicrous that we are giving a test to prepare students for another test that is designed to prepare students for a third test.
But my student, a senior, thought this was a good thing. She also added that we should have had SAT prep courses built into their schedules, so that students could do better on the SATs. To her, the path to success is simple – you need a class specifically to prep for a given test. Without it, how can you hope to do well on the assessment?
The problem is that she had internalized the haphazard, disjointed approach imposed on our school by administrators and policy makers. It’s just an accepted part of the school culture now that everything is driven by tests.
At the NJEA Convention last November, Pasi Sahlberg spoke about our national obsession with testing. He described it using a wonderful metaphor – a person going to the doctor for an annual check up. If you want a clean bill of health, you should live a healthy lifestyle year round. You don’t prepare for the doctor’s visit by acting extra healthy for a week or two before you go. That’s just silly.
Likewise, we should be focused on providing a sound educational experience year round. If our students perform poorly on tests, thats a symptom of a problem with the system as a whole. If we focus our efforts on short term test prep and manage to raise our test scores, we’re deluding ourselves into thinking we’re providing an education when in fact we’re not.
This leads me to my first point about the proper role of standardized testing: it should serve as an audit of the system instead of a measure of achievement at the student, classroom, or school level.
In my classroom, I assess students all the time. I might ask a question, collect an exit slip, use an online response poll, or give a quiz at the end of the week. In each case, I get immediate, useful “data” about my students’ performance. This is an ongoing process, and it creates a rich and complex picture of my students that I can use to inform my instruction. If I want a summative assessment to gauge my students’ mastery of important concepts, I can design appropriate assessments and collect them in a portfolio. If students struggle, I can provide feedback and give them an opportunity to revise their work.
A standardized test can’t do these things. It’s a snapshot of a single moment in time. It’s not reliable, in the sense that an individual student’s score will vary from day to day. The results aren’t timely, and I’ll never get the data in time to adjust my instruction in a meaningful way. For purposes of assessing student achievement and performing formative assessment, teacher created, classroom assessments are superior in every way.
What a standardized test can do is give us a general measure of how an entire population is doing – say all of the students in the state that share a particular demographic characteristic. These results can help us diagnose the health of our educational system as a whole. This is the only purpose for which we should use standardized assessments like the HSPA or the PARCC.
Which brings me to my second point about testing. If our only purpose for administering standardized tests is to collect aggregate data about populations, there’s no need to test every single student every year. Its an inefficient system that squanders precious time.
To return to the previous metaphor for a moment, if your doctor wants to run some tests on you he’ll draw a few small vials of blood. From this sample, he can determine the health of the body as a whole. Similarly, we can draw samples of students to gauge the health of our schools.
We already have an alternative model. The NAEP is administered using sampling methodology so that by testing a small percentage of students across the nation we can make meaningful comparisons across states. Likewise, the PISA samples students from around the world and allows us to draw comparisons internationally.
The NJ Department of Education could easily create a system to sample students from districts across the state, allowing us to efficiently capture a snapshot of the state of our school system. We would only have to test a tiny fraction of our students, and there would be significantly fewer disruptions to the school calendar.
Now I understand that this system would not currently be plausible under the ESEA. Since 2002, the federal government has required states to implement an assessment system that tests at least 95% of the student population in various years. But this is not set in stone.
The Senate Committee on Health, Education, Labor and Pensions has been working on a new version of the ESEA. In both hearings, testimony was given that supported the use of sampling methodologies to collect aggregate data in a more efficient way. The Committee Chair, Senator Alexander, has signaled his openness to rolling back the current federal requirement for annual testing.
In this context, it is imperative to create an alternative vision to the system that we currently have.
The federally mandates tests have not magically eliminated the achievement gap. The new PARCC tests, which require even more time throughout a student’s career, will not magically eliminate the achievement gap, either.
Tests don’t teach, and our current system is not working. Instead of increasing the frequency, length, and difficulty of standardized assessments, we should instead reduce our reliance on these tests.
The Commission should recommend that the state consider the plausibility of a new assessment system that relies on sampling methodology to collect data about target populations without including any high stakes for students, teachers, or schools. Thank you, and good luck in preparing your final report.