Let's begin at the beginning.
While at an elementary school doing research for a book about the impact of standards and testing on American education, I spent a lot of time watching a girl I called Whitney. Among other disabilities, Whitney had mild mental retardation. Although she was in fourth grade, she could sound out words only on the level of a first-grader, and her ability to comprehend what she read and heard seemed no more advanced.
It was bad enough that Perlstein wrote an entire book condemning NCLB based on her observation of a single school (n=1). Now she's going to
repeat compound that error by condemning NCLB based on a single student.
Right off the bat we learn, that something has gone seriously wrong with Whitney's education. She's in fourth grade performing at a kindergarten level (sounding out is really a kindergarten skill). She can't read yet. Mild mental retardation notwithstanding, it should have taken the school one year to teach Whitney to read. The school has had four years and failed to teach a year's worth of material. The school has most likely done something wrong if Perlstein has given us an accurate description of Whitney.
How do I know this? I know this because there is at least one instructional program out there, DI, that has been used to teach thousands of children, including many in the "mildly retarded" range, how to read in one school year such that the creator of the program has confidently stated many times that:
[We have] consistently demonstrated that if a reading sequence is properly implemented in kindergarten, virtually all at-risk students with the exception of the profoundly retarded and the very frequently absent will read by the end of the year. No program that purports to be a model of reform should have a standard less demanding than “Read by Grade 1.”
Whitney is not profoundly retarded and there is no indication that she was frequently absent from school. She should have learned how to read by first grade. This is a critical point which undermines most of Perlstein's conclusions in this Op-Ed.
I once saw a teacher spend 15 minutes, as the rest of the class worked independently, trying to explain to Whitney that when you sell something you get money for it, a concept crucial to understanding the story at hand. Teaching homonyms was exhausting, if not futile, because at least one word of every pair (dew, grate) was something Whitney had never heard before and could not grasp once she did. When a special education teacher told Whitney that synonyms have the same meaning, she asked, inexplicably, "Like a science experiment? Like a dinosaur?"
Perlstein's implicit message in this paragraph is that Whitney is somehow defective and this defect is the reason why she isn't learning. Wrong. It is the teaching presentation that is defective. The concepts of homonyms and synonyms are taught in the second grade in DI. In fact the word "dew" is explicitly taught early on in second grade. It is not easy to teach "mildly retarded" kids these concepts, but it can be done in a carefully designed sequence that is competently taught.
In fact, low-performers like Whitney can be mainstreamed if they are are given some extra time before a lesson in which the material is pretaught to them. In this way, these lower-performing students can keep up in a regular classroom while learning the material and experiencing a great deal of academic success which will be highly reinforcing to their motivation.
Here's the kind of performance gains we'd expect to see if the students were competently taught. The data comes from the thousands of students that went through Project Follow through.
Notice how the low IQ kids are keeping pace with the high IQ kids and that by the third grade, the first time they are tested under NCLB, the average student with an IQ below 71 should be approximating grade level performance.
But a large problem remains: Under the versions of the law under discussion, Whitney will still be given the fifth-grade test in fifth grade, the sixth-grade test in sixth grade and so on. She will probably fail these tests -- no surprise to her teachers -- and whatever progress she makes, unless it is so miraculous as to wipe away her deficiencies altogether, will go uncredited. Worse, her time and her teachers' time will be badly misused.
Whitney will likely fail these tests because she has been not properly taught. This is the condition that NCLB is trying to eliminate. As the Follow Through data shows, we should expect to see plenty of Whitneys who are able to pass grade-level tests. This is the counterexample that proves Perlstein wrong.
Perlstein has written off all the Whitneys based on the limited observations of one Whitney at one school which probably has never successfully taught a student like Whitney. Perlstein's understanding is that Whitney has learned as much as she can learn. So, when she agitates for a growth-based measures in NCLB 2.0 you can bet the amount of growth that she believes is appropriate for a child like Whitney is the kind that permits a student to be a non-reader in fourth grade. Whitney's time has been badly misused and if we follow Perlstein's advice you can bet that condition won't change.
It's not just that Whitney's progress can't be properly measured by a test that's way above her head.
The question is why is the test way above Whitney's head in the first place. The answer to which Perlstein has not adequately researched. If the answer is that the teaching that Whitney has received has been inadequate, then the problem is with the teaching and not NCLB's requirements. I've shown above how there exists some good evidence where one could conclude that this may be the case for many, though not necessarily all, kids like Whitney.
It's that by taking to heart the law's mandate of every student in a grade working toward the same target, administrators are making bad instructional decisions that permeate classrooms nationwide. Teachers follow pacing guides that tell them what to teach each day, no matter where their students are. Students take benchmark exams each quarter and unit tests each week that correspond to how much time has passed, not what those particular children need to learn.
Administrators are making bad instructional decisions because they don't know how to reliably teach kids like Whitney. That's the fundamental problem. Pacing guidelines merely tell teacher's what ought to be taught by what time. The problem is that teachers don't know how to teach the Whitneys in a way that the Whitneys keep up with the pace. As I've shown above, many Whitneys should be able to keep up with the pace if they are competently taught.
You can blame No Child Left Behind, the climate it's induced or the questionable choices people make in its name. Whichever way, as long as students are judged only on grade-level tests, no matter their needs, and as long as the education they get the rest of the year hews to that goal, they will lose out.
Students also lose out if they are capable of performing at grade level if taught properly but never receive adequate instruction. If, as Perlstein suggests, we loosen the NCLB standards what incentives to improve will exist then?
The only time I saw Whitney make progress was the hour she spent each day with a specialist who guided her in blending letters to make sounds -- hardly a skill in the fourth-grade curriculum. Is it too much to ask that children such as Whitney be taught what they need to learn in order to make their own adequate yearly progress?
The question remains though why wasn't Whitney taught this first grade skill in first grade? Was it because Whitney was incapable of learning the skill in first grade or was it because she didn't receive adequate instruction? Perlstein completely fails to obtain an answer to this question. How does she even know the amount of progress that a kid like Whitney is capable of given adequate instruction? These are the hard questions that Perlstein should have asked and found answers to before penning the Op-Ed.
Update: Keven Carey, Erin Dillon and Aftie Michele weigh in.