Designing metrics to define a quality education — or to pick a school — is a controversial and surprisingly imprecise science. Image: BensonKua/Flickr
Last week I wrote a somewhat half-baked post describing simple numbers that parents can use to pick an elementary school (the first and second were solid!). This week, I called around to get experts’ take on the topic. Here’s what they said.
First, it turns out that education statistics and measurement is politically fueled beehive of badness and so finding anyone willing to go on record recommending any specific measure was tricky – these metrics are simply too controversial. But off the record I chatted with folks at the Journal of Education Statistics and a numbers guru at the National Center for Education Statistics and have bundled these two anonymous sources into one I’ll affectionately and originally call Deepthroat (conspiracy theorists, start your engines…).
This is Deepthroat’s main point: “Imperfect measures are better than no measures.” Anonymously and with that in mind, Deepthroat was happy to recommend the following three ways to pick a school by the numbers.
First, Deepthroat suggests looking at variability of abilities within a classroom. “If you’re in fourth grade and to the left of you is a student doing first grade math and to the right of you is a student doing eighth-grade math, the fact of the matter is you may not see as much instruction tailored to your learning level,” he says. So instead, perhaps a concerned parent should look for a class in which students have fairly uniform abilities.
Of course, this is the hugely controversial idea of ability grouping or to use the more politically loaded term, “tracking.”
Deepthroat says, “Tracking done well works and tracking done poorly doesn’t.” For example, he says the pullout movement – grouping kids by ability for remedial or advanced math or language arts classes – is largely a failure. He calls this “tracking done poorly” and attributes lack of pullout success to widely misidentifying student abilities and the likelihood that teachers in these pullout classes will be of lower quality than teachers of students’ mainstream classes. Too frequently, ability grouping puts together the wrong kids with the wrong teacher (and that’s without the moral and ethical difficulties that come with imagining its implementation).
In lieu of ability grouping, teachers are getting better at “differentiating” the instruction in any single classroom to meet their students’ diverse learning needs – but it would be much easier and, according to Deepthroat, likely more effective to teach a group of students at the same instructional level.
Second, Deepthroat warns us to be wary of test results. Remember those students who within one classroom are working at the 1st, 4th, and 8th grade levels? This leads to what Deepthroat calls “high variability within a class but low variance across grades.” This homogenizes the way student scores creep up through the grades – test score percentiles shouldn’t jump 20 points from second to third grade classes, and if they do you should question their validity.
“As a parent you should be suspect of high performing schools,” says Deepthroat. “We all know these tests aren’t as solid as they should be and we’ve seen that students, teachers and schools cheat. Don’t make a school over promise. Instead of a 10 or 20 percent jump between grades, the marker of a good school is consistent, routine, year-after-year gains.”
Finally and most importantly, Deepthroat says that good teachers are the key to a good education, pointing to a recent Harvard/Columbia economics study of “value added” by good teachers (based on tracking one million children from a large urban school district from fourth grade to adult). The paper writes that, “Replacing a teacher whose true value added is in the bottom 5% with one of average quality would generate cumulative earnings gains of $52,000 per student or more than $1.4 million for the average classroom.” Yes, this is the language of economics and not sociology or necessarily education and, yes, we want more for our children than good test scores toward the end goal of a high salary, but the point remains: a good teacher makes a big difference.
Unfortunately, according Deepthroat, you can currently only see the value of these teachers in hindsight – there’s no way to predict who these high-value-added teachers are likely to be.
But that’s not to say there couldn’t be a measure in the future.
He suggests the following: “because teachers’ subject competence is so important, why not give teachers the same test you give kids?” Amid the many variables of teacher quality, teachers who demonstrably know the material are better than those who don’t.
Until an objective measure of teacher quality staggers from the Petri dish, Deepthroat suggests that gathering subjective data about the quality of potential teachers might be the most important thing you do for your child’s education. Or you can take Deepthroat’s parting advice when he says, “parents should look for the school that they think best fits their child and that logistically works best for their family.”
[By the way, if usable science from authoritative sources is your thing, please consider taking a peek at my new book, Brain Trust.]
