Myth Busting

Here are a few excerpts from Chapter Two (PDF) of The Outrage of Project Follow Through: 5 Million Failed Kids Later. There are a few more excerpts that deserve a post of their own. Hopefully I'll get to them.

On Parental Support

Increased parental support is frequently trotted out as a cure for our education woes. My position has been "great if you can get it, but in all likelihood, you won't get it for the kids most in need of it." Here is Engelmann (p. 14):

One Piagetian-based approach was that of Ira Gordon from the University of Florida. His idea was to provide stimulation in school and use trainers to visit parents and instruct them in how to work with their children at home on the same Piagetian activities that were presented in school. I thought he was either misguided or the kids in Florida were a lot different from the ones we worked with in Urbana, California, Chicago, and Pennsylvania. Gordon’s strategy might have worked with the kids in Toronto [i.e., strong concerned parents], but not with kids who need more help. With the inner-city population, there is little likelihood of either being able to work extensively with the parents or of the parents working successfully with their kids. This is not to say Gordon's goals were ignoble, simply impractical. If it’s not taught thoroughly in school, the parents of the kids most in need are not going to be successful at augmenting the program.

Engelmann's identified two problems with trying to get increased parental support: 1. you can't get it from the parents who would most benefit from it, 2. even if you can get it, the parents will likely be capable of providing the needed instructional support, 3. even if they can, success still depends on how well the material is taught in school in the first place. And, there we are right back where we started, poor instruction in school clogging up the works. Remember, if the instruction in school were better, schools wouldn't be looking for parental support in the first place.


On Why Kozol is an Idiot

Kozol and his ilk believe that the achievement gap of poor blacks is the result of a social problem that is rooted in history and caused by discrimination against blacks. Their solution is to negate the social problem by putting poor kids into more affluent schools. This solution is based on Thomas Pettigrew's analysis of the Coleman Report data which found that "black students attending mostly white schools had achievement levels much higher than blacks in segregated schools. Also, in these schools, the white students’ performance was no worse than that of whites in segregated schools." This conclusion begot bussing. Bussing failed miserably. Engelmann explains why (p. 7)

The conclusion drawn by a thoughtful person who was not racially prejudiced would probably have been, “So what? Just as there are middleclass whites, there are middle-class blacks. Middle-class kids perform at a high level whether their neighborhood schools are segregated or integrated.”

Once more, policy makers confused correlation with causation and drew the conclusion that if black children in integrated schools performed higher, putting black children into white schools would create an integrated school, and the children in this integrated school would perform as well as the children in schools that were “naturally” integrated.

This reasoning is both romantic and cruel. Jackie and Alan couldn’t perform on fourth-grade instructional material in any school. They had learned only about two years worth of skills in four years. Putting them in a fourth-grade classroom with children who performed two years above them would constitute incredible punishment, not intelligent education that ostensibly addresses “the local needs of the children.”

Yet the meme not only persists today, it remains very popular, as do all non-educational solutions to what is in essence an instructional problem.

On the Failure of John Dewey

(p. 21)

I found the naiveté of proponents of progressive education shocking. If people knew much about John Dewey, they would have known that he and his wife operated two lab schools based on radical progressive education and both were unquestioned failures. The first was at the University of Chicago. It failed so categorically within three years that Dewey was forced to leave Chicago. He went to NYC where he and his wife founded his famous Lincoln School in Manhattan. This effort failed in two years. Yet, a hall full of grown, educated people from NYC found the slogan, “learn by doing,” something of both a battle cry and religious experience, regardless of the amount of student failure it generated.

On why you want it teach it correctly the first time

The reason why you want to teach it right the first time is because if you teach it wrong it becomes much harder to correct. (p. 46):

Our compulsion to obtain current data on the performance of every student stemmed from some facts about learning and relearning. The relationship is simple: The longer a student has misunderstood something, the longer it takes to teach the content correctly. If the teacher uses a good reading program, at-risk beginning readers will learn to discriminate the words a and the with perfect accuracy after about 40 trials. If the teacher uses a poor program and children confuse a and the, their confusion becomes solidly ingrained. When they are fourth graders, it takes about 400 trials to re-teach them so they are perfectly accurate at reading these words. If the remedy is delayed until the students are in high school, it will probably take more than 1,000 trials to induce the correct behavior.

I think it's safe to say that most teachers don't go over anything approaching forty times in the first place. And 1000 times in high school to correct something taught back in first grade? No wonder why high school teachers think it's hopeless to remediate. It is.

On cognitive ability and instructional pacing.

Charles Murray created quite the stir two weeks ago but his underlying premise is accurate--high IQ kids can learn considerably faster than low IQ kids even when the instruction is exemplary (p. 46):

Our initial plans for securing data involved both record keeping and frequent observations of teachers. We first set minimum expectations for every instructional group. Highest groups were expected to progress the fastest through the program—about 1.5 lessons a school day. The expectation for average groups was about 1 lesson a day, for low groups about 3/4-lesson per day. These projections were to be modified as we discovered more about the performance level of each child.

The high performing groups are capable of learning at twice the pace of the low performing group. (Engelmann has said elsewhere that he believes that the high performers can proceed at three to four times the lowest rate.) Today, the lowest group gets extra instructional time to make up for their slower pace.