Legislating to the Test

by Rachel Levy
January 26th, 2012

Posted in Education News, Opinion | 3 Comments »

It’s legislating season here in Virginia. One bill by state Senator John Miller (D-Newport News) would remove the Science and Social Studies SOL (Standards of Learning) tests from third grade, not because there are too many tests and not because Senator Miller thinks science and social studies shouldn’t be taught, but so that teachers can spend even more time preparing students for the Reading and Math SOL tests, the Reading Test in particular. Here’s the rationale:

“Miller told the subcommittee that the JLARC study showed that 95 percent of third graders who pass the reading proficiency test will pass the reading SOL in fifth grade, while those who fail have a “50 – 50 chance” of failing the fifth grade test, and ultimately failing in school.”

I read this right after writing a post for the Virginia Education Report explaining how we might advance literacy in Virginia, as our Governor says he wants to do. My second suggestion was:

We need to spend much less time teaching reading as a subject and teaching reading strategies beyond their utility and much more time teaching content or subject matters, such as literature, science, social studies, p.e., art music, foreign languages, technical education, etc. Yes, most kids need to be explicitly taught to decode and yes, to a point reading strategies are useful. Of course, content should be taught as reading and writing intensive. However, literacy is largely representative of someone’s background and content knowledge, and knowledge of vocabulary and does not develop or improve without it. As the University of Virginia’s own Dan Willingham says, teaching content is teaching reading. (It’s also much, much more meaningful and interesting for kids.) My regular readers know that I talk about this ad nauseum. In case you’re new to my writing on education, here are some posts that elaborate further: here, here, and here.

So, first of all, yes, we’ve all made that point on this blog a thousand times.

I can tell you as the parent of two public school third graders that plenty of time is already spent preparing for the Reading SOL. I can tell you as a former Virginia public school social studies teacher that the History SOLs seem to be relatively heavy on minutiae and relatively light on essential knowledge and broader concepts. I’ve always been reassured, though, that unlike many other states, at least Virginia has SOLs for numerous subjects and not just for math and reading. Now, whether any of those tests (all multiple choice except for the writing test) are of good quality is another question. Whether the SOL curriculum is of good quality is yet another one. Neither seems necessarily so given what Chris Dovi reports here about what happens to many Virginia public high school graduates who are successful at mastering the Standards of Learning but not very successful once they get to college.

To be clear, while I am pro-assessment and all for data-informed instruction, I am not currently in favor of many aspects of NCLB or high-stakes standardized testing. Even so, I am somewhat sympathetic to the stance taken, by Andrew Rotherham here in this column about cheating scandals:

“We know from research — as well as experience and common sense — that the best way to help students perform well on standardized tests is not to drill them (and certainly not to cheat) but rather to actually teach them. . . . Real teaching is like a well-rounded breakfast: it sustains you. Drilling for a test is like eating a doughnut: it works for a bit, but you’re hungry again before long. After all, what most assessments are testing is the ability of students to encounter and master material that is unfamiliar in its specifics but similar to what they’ve been taught. So the takeaway for parents is straightforward: with good teaching, the tests take care of themselves. When teachers or schools obsess over tests, parents should be concerned — not about the test, but about the school.”

I say “somewhat” because when Rotherham says that, “with good teaching, tests take care of themselves,” he leaves out (or blithely assumes) good curriculum. This comes across as, it doesn’t matter what’s being taught as long as the teaching is good. In that case, it doesn’t matter if you serve doughnuts every day for breakfast so long as the cooking is good; with good cooking, a good report from the doctor’s office takes care of itself. Otherwise, yes, with good teaching and solid curriculum (and an environment where teachers are free to engage in good practice and to teach knowledge-based curriculum), the tests should theoretically take care of themselves.

The idea that testing isn’t the problem, though, lets policy makers off the hook. Educational malpractice cannot solely be laid at the feet of bad teachers or bad teaching. Senator Miller’s bill is case in point that practice and curriculum are influenced by policy. This bill essentially dictates bad practice. This well-meaning legislator in Virginia said expressly that he was legislating to the “test,” passing a bill that is meant to mandate that teachers spend more time preparing for a reading test, the stated goal being to get those reading scores up. The stated rationale is not: it’s better educational practice (it’s not); it will make for better education (it won’t); or, explicit preparation for standardized reading tests make students better readers (it doesn’t); rather, it’s getting pass rates up.

If we want teachers to stop teaching to ill-conceived tests then lawmakers are going to have to stop legislating to those tests, lobbyists are going to have to stop lobbying to the tests, and reporters are going to have to stop reporting to the tests. While I think good policy can create the conditions to spur meaningful education reforms, I have serious doubts that we can directly legislate better teaching and more meaningful, knowledge-based learning. If the powers that be are going to try anyway, may they at least legislate sound practice and a broad and rich curriculum, and not more vapid reading test prep.

UPDATE: The bill has been passed 33-7 in the Senate (and, by the way, was supported widely by groups representing Virginia educators). Senator Miller said,“I believe it makes common sense to concentrate on reading and math, and give a good basic foundation in those two core subjects for our students.”

Unfortunately, this is what is accepted as common sense in education today, but it’s far from common sense. People learn to read and there are some reading strategies that can be of great use, but people do not learn reading; it’s not a subject. By assuming and then legislating as if it is, we undermine our students’ acquisition of knowledge and their literacy development. I understand that many Virginians want fewer SOL tests and I don’t blame them, but all this bill will likely do is replace subject matter instruction with more reading instruction, and make it so that Virginia’s kids struggle with literacy more. Without background knowledge, exposure to vocabulary, and instruction in content, literacy does not develop.

Rachel Levy is a writer and a former (and likely future) ESOL and Social Studies teacher who lives in Ashland, Virginia, with her husband and three children. She blogs about education at All Things Education.

When Reading Tests Attack (Content)

by Rachel Levy
August 3rd, 2011

Tags: , , , , ,
Posted in Assessment and Testing, Teaching, accountability | 21 Comments »

During  my first year of full-time teaching ESOL and Social Studies at an inner-city Washington, DC, high school, my principal approached me and told me that my students had come to her saying how much they were enjoying history class. I explained to her my intent to teach content but with a reading and writing intensive emphasis, to build those skills which were quite low among our school’s students. She was enthusiastic and I was thrilled.

A few weeks later, she attended our Social Studies department meeting where she explained to us that since there were no standardized tests in Social Studies, from that point forward, we were required to spend one-fifth of our class time teaching the Stanford-9 Reading Test.  For each of my students, I had to make charts based on testing data showing the skill (for example “context clues”) and how they did on that skill.  Then I was supposed to target my lesson plans to teach and remedy each student’s individual weaknesses. This didn’t seem right, but there was no protesting this: I wanted to help my students, she was my boss, and she was telling me what to do. Furthermore, such instruction and data collection had to be documented in our lesson plan books and during classroom observations.

This is where and how NCLB-applied pressure and high-stakes testing cause poor practices. Some counter, “The testing itself doesn’t cause teaching to the test in an ineffective way. Why don’t teachers simply adopt effective practices, like Core Knowledge?” While testing shouldn’t (in principle) encourage poor practice, unfortunately, my experiences in the classroom and now as a parent shows that national policy incentives mandating high stakes testing change classroom teaching for the worse. Ground-level feedback can help us to see how to fix accountability better than philosophical debates about the nature of testing.

I agree that it’s completely logical, obvious even, as Andrei Radulescu-banu put it on Robert’s recent post, that A=>B (Please read  his comment in its entirety). Certainly, test scores will gradually rise if a well-rounded and knowledge-rich curriculum is implemented. However, many educators are hindered in following this logic by performance pressure and by belief.

There were vague and all-encompassing standards (think horoscopes), however there was no social studies curriculum in DCPS at the time (there still isn’t).  By collaborating with my colleagues and relying on my own education and knowledge of social studies topics I came up with unit and lesson plans pretty easily. However, I really struggled to come up with lesson plans for teaching the Stanford-9 Reading Test.  I thought at the time that it was because I didn’t have much background in reading instruction, that I was missing something. Eventually I figured out that it wasn’t that I was missing something, it was because “teaching” the Stanford-9 Reading Test made absolutely no sense (and Tim Shanahan explains here that such an approach doesn’t work). So I taught history and geography as much as I could and I taught what I imagined “teaching the Stanford-9 Reading Test” was only when I had to, and made sure I had passable documentation in my lesson plan book.

In my second year, I got braver. In faculty meetings when we talked in small groups about how to get our test scores up, I voiced my opinion that the way to get test scores up was via an implicit route—to teach content and have students read and write as much as possible. I stated my skepticism that one could teach the Stanford-9 Reading Test or that students could learn the Stanford-9 test, but except in private asides, I had no supporters. And this was before Michelle Rhee came to town, mind you. Before NCLB, teaching content to struggling students was unappreciated; now, it’s practically an act of subversion.

Now, don’t get me wrong, I have little tolerance for the secondary teachers who say their job isn’t to help struggling readers. Yes, kids should be reading well by the time they get to middle and high school. But for whatever reason (and it’s probably a good idea to try and figure out that reason), many aren’t. Accept it and adjust your practice accordingly. If you’re not prepared to teach and help struggling readers or even non-readers, you have no business teaching in most American public school classrooms. But this does not mean giving a student reading on a first grade level a tenth grade textbook and telling her to read it. Rather it means finding content-relevant materials appropriate to students’ age and maturity and books on their level, scaffolding, and building up.

I also don’t blame my principal. Downtown was breathing down her neck, judging her on reading and math scores.  Rather than fight a losing battle that could cost her her position (while she was ambitious for herself, she also cared deeply about the students in her charge and about the school she had built from scratch), she embraced it all, including putting on Stanford-9 pep rallies. I am not making that up; these pep rallies happen.

In addition to what I describe above and the bankrupty of the content of the tests themselves, described by Diana Senechal here, another fundamental problem is that many of the advocates of reforms centered on testing-based accountability actually believe that kids who can’t read (decode) at all well should not be learning content, that they have to learn reading first and then they can learn content, that teaching a content-based curriculum is useless if kids can’t read. “Let’s focus on teaching reading and get the reading scores up and then we can worry about content.” And let’s be honest, even many teachers and educators who are opposed to testing-based accountability believe this. I encounter this all the time, as a teacher in both inner city and high-performing suburban districts, as a parent in my children’s high-performing district, and in my interactions with readers as an education blogger and writer.

I encountered this attitude, that language proficiency is a prerequisite to learning content frequently as an ESOL teacher. People insisted that English Language Learners should master English first before learning content. However, a very effective way to teach the English language is through the “sheltered content” model, where the content is a vehicle to teach the language. Of course, both English Language Learners and struggling readers need intensive and explicit language and reading instruction, but not beyond its utility and not without pairing it with content.

In the vast majority of cases, the “belief” that students have to learn to read before they can learn content is not a result of dysfunction, laziness, or poor intentions. Quite the contrary–this belief is based on the proven correlation between strong literacy skills and academic success and on the understandable urgency to get kids to master such skills. Unfortunately, it is also based on a fundamental misunderstanding of how people learn and of how such literacy skills develop. Some are celebrating NCLB’s increasing poor and minority students’ ability to read the tests they’re taking, but just because they can read the tests doesn’t mean they actually have a grasp of the content or that they know more. Being able to read the road signs is well and good, but it won’t get you very far if you don’t know how to drive or where to go.

Rachel Levy is a writer and a former (and likely future) ESOL and Social Studies teacher who lives in Ashland, Virginia, with her husband and three children. She blogs about education at All Things Education.