Jencks, page 139

I'll just quote this:

There is a popular theory among both educators and laymen that middle-class children work harder in school than poor children.  We cannot find any good evidence that this is so.  When we compare economically advantaged students to disadvantaged students with the same test scores, for example, we find that they get the same averages high school grades.  We assume that diligence has more effect on school grades than on standardized test scores, and we therefore conclude that economically disadvantaged students probably work as hard as economically advantaged students with comparable aptitudes.  More generally, we conclude that high school teachers reward a set of traits which, with the exception of academic aptitude, are not very class-related.  This clearly contradicts most people's preconceptions.  We suspect, though we certainly cannot prove, that these preconceptions are based on a misunderstanding of the dominant values of working-class families.  While a few lower-class and working-class children behave in ways that schools find unacceptable and try to punish, the great majority evidently do not.  The deviant minority seems, however, to shape middle-class stereotypes of working-class values and behavior.

My main thought from this is that the 60's and 70's seem to have been a better situation than I thought.  Why haven't we made more progress?

Jencks, pages 136-137

I'll just quote two things.  First, on page 136:

A second reason schools have become certification agencies is that this serves the interests of a society that wants people sorted and graded but does not know precisely what standards it wants to use.  If high school diplomas or other certificates of competence were given solely for passing examinations, there would have to be political agreement on what the examinations should cover.  This would be hard to get.  Delegating the problem to the schools is a way of sweeping it under the rug.

In many respects, the mantra of "education is the solution " is a way of passing the buck on social and economic problems.

Second, on page 137:

If credentials measured skills that were only learned in school, equalizing the distribution of schooling would equalize the value of the credentials schools conferred.  But educational credentials also measure traits that people acquire outside of school.  Equalizing the amount of kind of schooling that people receive cannot alter these traits.  This means that an 8-year gap between the best- and worst-educated fifth today may be just as important as a 10-year gap used to be.

Indeed, pushing more people through a system either requires Herculean efforts from the students, the educators, and the people around them, and vast resources for everyone involved, or certain polite fictions that were once called "Gentleman's C's."  Graduation rates are hardly everything.

Inequality: School of Rorschach

On page 90 it is noted that the best predictor of a 12th grader's scores on a great many standardized tests is their scores on similar tests at the start of 9th grade, i.e. where they go to high school is less important than his they start out.  This is not surprising.  Either it is a damning indictment of educators for failing to lift people up or it is the best argument for getting off the backs of educators because outcomes are largely beyond our control.

Inequality: How things have changed

It is hard to come up with a coherent reaction to this book.  The emphasis on detail rather than narrative and theory means that from a blogging perspective there is not a lot to "react" to, just absorb.  But on page 81 the authors point out that while a large fraction of the variance in test scores is heritable (note that "heritable" just means that you get it from your parents, whether via genes or upbringing), a system that based privileges solely on test scores would only let about a third of upper-middle-class kids stay in the class into which they were born.   I am not going to judge the level of empirical support for their claim in 1972; I will just note that while test prep is hardly perfect it can be effective, and the most effective test prep is not a few weekends of Kaplan but rather 12 years of good schooling.  To the extent that upper-middle-class educational arms races have heated up in the past four decades (and I freely admit that my anecdotal observations on that point are no substitute for real data) this could undermine some of that prediction by Jencks et al.  If you tell the upper classes that their kids can no longer get into desirable universities based solely on last name (and legacy admits are fewer in percentage, even if by no means gone) then they will not simply shrug.  They will try to compete and play to their advantages.  This is the same response that I had to Guinier's book.