Yesterday’s Wall Street Journal had an uplift piece on using gee-whiz data analytics to improve Chicago’s public schools. I found it incredibly depressing. Here is how the article opens:
At 7:15 on a chilly May morning, Marshall Metro High School attendance clerk Karin Henry punched numbers into a telephone, her red nails clacking as she dialed.
“Good morning, Miss MeMe,” she said to Barbara “MeMe” Diamond, a 17-year-old junior with a habit of oversleeping. “This is Ms. Henry, your stalker.
The timing of the call was key. Earlier in the year, Ms. Henry and a co-worker were spending nearly two hours a day calling every student who hadn’t checked into school by 9:30 a.m. But weekly data tracked by their office found that only about 9% of those students ever arrived. So they changed tactics, zeroing in on habitual latecomers like MeMe, and delivering wake-up calls starting at 6:30. On that May morning, 19 of the 26 students called showed up.
“I just stay in bed if no one calls me,” MeMe said. “That 6:30 call be bugging me, but it gets me here.”
Sharief Raines, an 18-year-old senior with a toddler at home, took the challenge after missing every school day in December. In January, she showed up 12 of 19 days. Ms. Calhoun even watched the baby one afternoon while Sharief did homework. “I saw Dean Calhoun was trying to help me,” she said. “I didn’t want to let her down.”
Sharief graduated June 11.
The attendance clerk sounds like somebody getting into the office early to get her job done, and I assume that both MeMe Diamond and Sharief Raines have faced enormous obstacles in their lives. I say this without malice, but no school is going to solve the problems of many students like this. This school exists within a sea of dysfunction that it cannot fix.
Globalization has created trans-national labor pools through a mix of literal outsourcing, immigration, and importing labor content via shipped manufactured goods. We move the people, the jobs, or the merchandise; but either way, workers in Illinois must increasingly compete with workers who live in Eurasia or have immigrated here from Latin America and elsewhere. These are no longer poor people “out there somewhere” for whom we should feel pity and give foreign aid, but people with whom, one way or another, our hourly pay is being compared by those who will decide where new jobs go. Today there are probably hundreds of millions of people on one side of the relevant labor pool who have such a different orientation toward school that the worry is that they’re working too hard, and hundreds of millions of low-skill competitors on the other who are prepared to work for wages much lower than those of even very poor Americans.
Within less than one year, MeMe and Sharief will have to compete in that environment. There is no fixed lump of labor. By specializing in what we do best, and then trading with ever-larger numbers of others who can afford to buy our output, we can become wealthier. What will MeMe and Sharief specialize in? Who in an open market will pay enough for their time to create sufficient income to support them (and Sharief’s child) in a humane manner? (It’s easy to read this as scornful, but I really just feel sympathetic, in that if dealt the same hand of cards, I think I would be in pretty much the same place.)
By extension, where are large chunks of the American labor force are headed? How much dysfunction can the productive economy carry on its back as the level of global competition rises ever higher?
The answers to all of these questions are, in my opinion, very troubling.
I don’t have any great solutions, but then again, I don’t think anybody else does either. “The Answer” is probably not there to be found. I doubt there are any silver bullets, just lots and lots of scut work in many areas, each of which can make a small contribution.
“Data-driven schooling,” if done with this perspective in mind, can certainly make an incremental positive contribution. But it’s easy to do it in a way that actually makes things worse.. If focused on short-term carrots-and-sticks that ignore character effects; if divorced from the right incentives for the participants; and if not focused on careful evaluation of the actual success or failure of interventions against validated outputs, it’s likely to be a huge waste of scare time and money.