Book review: "Education Myths"

[dmaze]

Over the past week I read through Education Myths: What Special-Interest Groups Want You to Believ About our Schools -- And Why It Isn't So by the (rightish) Manhattan Institute's Jay P. Greene. The phrase "special interest" is only really used on the cover, and inasmuch as organized groups are implicated at all in education's issues only the teacher's unions are specifically brought up. The book cites sociological research to make a point, largely that adding funding hasn't historically improved schools but current research on voucher programs suggests they help. This wasn't really a light fluffy read but it was still pretty well-written, and if the subject matter sounds interesting to you it's probably worth a read.

I have two big criticisms of the book. The first is kind of glaring where it comes up: Greene in a couple of places early on draws conclusions in the explicit absence of evidence. For instance, when discussing teacher pay (chapter 6), do teachers work very long days that result in them being underpaid per-hour? There's apparently only one even vaguely reputable study of this coming up with 43.9 hours per week, but it's not a great study and its conclusion might be within the margin of error, so having discredited all available evidence it must be true that teachers work no more than 40-hour weeks.

The other thing that nagged me through the entire book was more subtle, and didn't hit me until chapter 3 or so. The book cites quite a bit of research (and refutes quite a bit of it as well) but when "Jay Greene and Greg Forster of the Manhattan Institute have developed a systematic method for measuring..." it's not immediately obvious that this is the author. There's nothing particularly wrong with citing yourself, but I expect to see it as "I" or "the author". And in the end there winds up being an awful lot of it.

Connected with this is a tendency to dismiss existing measures of things. This jumped out at me in talking about racial diversity: Greene discusses a couple of standard ways of measuring integration but then says, effectively, "they're all wrong; Greene and Forster have developed an alternative..." Here their alternative makes sense, at least ("integrated schools" == "racial composition of school/classroom is the same as the greater metropolitan area"). But they use it to reinterpret a couple of other papers to support their conclusions rather than the original authors'. Similar reinterpretation happens around class; Greene claims it's standard practice to normalize for various things, and one of the things he thinks should be normalized for is a correlation between affluence and academic performance.

In all, this seemed like a worthwhile book to me. It visibly has a position (which you can guess from the cover and table of contents). I don't have the background to really tell if it's covering almost all of the relevant research or a specific subset, or if the author's research is totally legit. Still, it made an argument supported by data, and there's always something to be said for that.

Comments

[ukelele.livejournal.com]

The teacher per-hour pay thing is a raging controversy. There are a lot of things that aren't clear about how you should measure: should teaching be considered a 12-month job or 9? (It prevents you from taking equally remunerative, professional jobs over the summer (so you need to be able to live on it as a 12-month salary) and you need to put in time over the summer to do the job well, but you're really only at work for 9 months and that's all some people do.) Should only contracted hours or overtime count? (Many teachers, especially new-ish teachers, put in time beyond their contracted hours, but some argue that many professional jobs require this and, eg, lawyers don't go whining about it. (I'm dismissive, in that salaried jobs requiring substantial overtime are, by and large, more prestigious and remunerative than teaching by a fair sight.))

So, yeah, it's very unclear which hours you count. I think you'd also find that the distribution of hours worked is enormous, which means any single number you come up with will be just wrong. (For example, were you to look at only first-year teachers -- who, coincidentally, have the lowest salaries -- I expect you'd find 60-80 hours/week is common. As time goes on, I suspect you'd find a lower average but a much wider range.)

[nonnihil.livejournal.com]

The phrase "special interest" is only really used on the cover

Publishers, not authors, usually have the final say over a book's title; the title is considered part of the marketing of the book rather than part of the book itself in modern contracts. A lot of titles reflect this, focusing on issues that will motivate the people the publisher thinks are likely readers rather than on the actual content of the book.

In general, authors have no final say over anything that appears on the cover except their names; everything else from art to title to blurb is just marketing fluff.

[ukelele.livejournal.com]

(He has, by the way, as of fairly recently, a blog: http://jaypgreene.com/ . It's not in my feed but it gets cited a lot by things that are, so I keep wondering if it should be.)

[dmaze]

Google found me that fairly directly when I was wondering if the Internets had heard of him. Do you know if he's actually well-regarded?

[ukelele.livejournal.com]

Not sure; I've only been following the edublogosphere for a little while. Certainly, as I said, he gets linked to and listened to, but largely by right-wing commentators who would be disposed to listen to him anyway. I don't have wide enough experience to apply the bias filters properly.