The Limits Of Education

In theory, public education is the road to social advancement. In practice, student demographics and family life have prevasive impact on educational success at all levels, and students tend to stay at the level of achievement relative to their peers where they begin the process.

In the Denver Public Schools, "In reading, 77 percent of whites were proficient in 2008, while 42 percent of blacks were proficient and 36 percent of Latinos." Notably, "in Denver about 43 percent of the students who took the test in English are still learning the language." The proficiency level of whites in the Denver schools is comparable to that of students in the well regarded Boulder Public Schools and the Cherry Creek Schools.

Consider the tale of Colorado CSAP scores, in the news once again after the release of statistics for last year's round:

"[S]tudents already behind will have a difficult time catching up. Many of those students are poor or ethnic minorities.

Only 30 percent of students behind in reading — those rated partially proficient or unsatisfactory — are growing at rates that will bring them to proficiency over three years. . . . In writing, 26 percent are growing at a catch-up rate, and 13 percent behind in math are growing toward proficiency."

I've previously noted at this blog that almost all of the CSAP achievement gap is already present at the first test taken by children, in the 3rd grade.

A study by one of the authors of controversial book The Bell Curve, James J. Heckman at the University of Chicago (cited by New York Times columnist David Brooks and in turn by NewMexiKen), backs away from his previous hereditary IQ is destiny position, but continues to argue as Brooks summarizes that: "By [age] 5, it is possible to predict, with depressing accuracy, who will complete high school and college and who won’t.”

The study indicates that most of the factors that influence socio-economic success at age eighteen are largely in place by age 5. Heckman now concedes that early childhood could have effects in addition to heredity that are important, but stands by the evidence that elementary and secondary education have surprisingly little statistical impact on achievement.

Empirically and on average, public education does not have a leveling effect. We know that attociously bad childhood situations (e.g. exposure to violence or extreme isolation and neglect) hurt achievement. We have some tantalizing evidence from the KIPP charter school program, in particular, and also from the positive effects that increasing education levels have had on national economies, that effort far above and beyond that usually expected staus quo in public education, either more time per year or more years of it, with a good curriculum can make an impact. But, in the data that I have seen over the years, the extremes that matter appear to be quite small. For the middle 90% or so of students in public and charter schools, it is very hard to see statistically that public education does much to help lower socio-economic status kids advance themselves, although there isn't much evidence that elementary and secondary education is holding kids back either.

The cases where there have been major changes at the level of a large school district, such as Denver's big gains this year in middle school CSAP scores, seem likely to have as much to do with changing student bodies in those schools as they do with what in going on the classrooms.

We know that exposure to toxins in utero and in early childhood, for example, lead, is important to childhood development. We have reasonably good evidence that pre-school can have a positive impact, although it isn't clear that those benefits last until adulthood if not vigorously maintained. But, we also know that academic ability has a strong hereditary component.

Moreover, from the perspective of educators in the K-12 education system and beyond, it really doesn't matter if academic ability solidifies at conception, during pregancy and birthing, during a pre-school period, or some combination of these time periods.

In short, while most people who look at the issue closely agree that what we really want to evaluate schools based upon the knowledge added by the schools to their students, rather than the absolute levels of achievement which are sensitive to student body demographics, the evidence seems to show that the vast majority of schools are about the same in their ability to advance the knowledge of individual students. There seem to be real limits to what an educational model close to what we have now can make much of a difference in helping poor and minority students secure high academic achievement.

This is depressing news. It is contrary to the ideals that drove the creation of public education in the first place. But, we need to acknowledge what the evidence shows us in some way as we craft solutions, rather than simply doing the same thing and expecting different results.