Over at Eduwonk, Andy Rotherham poses the following thought exercise: What would you do with $5 billion to improve American education? Great idea.
My favored reform, not surprisingly, is a national curriculum. That would cost about a buck, since it already exists and merely needs to be implemented. What to do with the other $4,999,999,999? Two ideas:
- Scrap existing state tests in favor of a random testing arrangement. If schools only know that they will be tested twice a year, but don’t know which day, grade, or even the subject to be tested, the only way to guarantee good results would be to actually educate kids. Keep existing state reading and math tests, if you like, but use them for diagnostics, not to determine AYP. Until the laws of human nature are repealed, it’s naive to think the current prep-and-test regimen will do anything other than narrow the curriculum, and stress the heck out of teachers and kids. If you insist on testing (and there’s no reason not to; as public servants schools and teachers need to be held accountable) then you have to have a testing strategy that encourages the results you seek. Random testing would also give you a much clearer picture of what’s actually happening in schools. But prepare yourself, it’s worse than you think.
- This one idea will make me unpopular in certain circles, but teaching in a struggling inner city school, and observing in lots of others has solidified my belief that nothing matters more to student achievement than a positive, productive school environment. In a good environment, virtually any curriculum or pedagogy will work. You could put Nobel prize winners in front of every classroom in a dysfunctional school to no good end. Use the money to hire teachers for one-on-one home tutoring for our most disruptive students. The vast majority of kids come to school, even in our most challenged schools ready to learn, but their education is sacrificed minute by minute by constant disruption and discipline problems. I don’t know of any data on this, but I’d bet that the achievement gap is really a time-on-task gap. It is hard to overstate just how profound this problem is. Vast amounts of learning time are sacrificed to discipline problems, and the need to organize classroom management around behavior issues changes the entire classroom dynamic. It turns the teacher into an entertainer, not an instructor. If a child chronically demonstrates that he or she is cannot participate in a classroom setting, that’s a terrible shame. But by allowing that child to completely dominate and alter the school and classroom environment to the detriment of others, we lose not just that child but damage 24 others. Educate that child at home on the school’s nickel, and you help establish the positive, productive, achievement-oriented environment that is a prerequisite of success. This by the way, is probably the real secret of KIPP’s success. Every kid is down with the program. If not, they’re not a KIPP student anymore. The best schools — public, private and charters, show they’re serious about learning. Struggling schools will not improve until we show the students who are ready to learn and fully invested in their education that they’re the most important people in the building.
Feel free to cross post your best ideas here and over at eduwonk.
![[del.icio.us]](http://images.del.icio.us/static/img/delicious.small.gif){kind=small}
Regarding the random testing, I’m just thinking out loud, but what if you had two or three one month windows like October for the 1st quarter standards, January for 1st semester. Counselors and everyone else would be prepared for the logistics, and perhaps they would be notofied a day in advance. This means that testing would have to be done on computers not bubble sheets, so results would come quickly. And you would have to be prepared for bad news. But actually, I see standardized testing as primarily a guage of three things, the quality of your elementary school education and home life, reading comprehension, and morale. Students could figure out the answers of plenty of questions if a) they had practice in critical thinking, and b) motivation to think through the options. Classes with good morale and a learning culture would not have their scores drop. Classes just going through the motion and that are out of control would drop. It would be sorta like an unannounced classroom evalutations. Most students who like their teachers will do everything they can to impress the evaluator to help their teacher. Those classes would pull it together for a random test.
And since the tests wouldn’t be high stakes, we wouldn’t exclude anyone’s scores. Then you would get a real picture of what percentage of the kids who we are reaching.
I’m also intrigued by your idea of tutoring at home for the most disruptive students. In my experience, there are a lot of good hearted teachers who get defeated by the peer pressure that disrupts class, even small special ed classes where there shouldn’t be the time on task problem. But one on one, those teachers could be very effective traveling teachers. It could be very cost effective. The key is whether the idea would be helpful in breaking the ideological logjam that keeps us from investing in alternative slots that are absolutely necessary to create orderly environments in high poverty neighborhood schools. I’m going to try the idea out on a couple of Board memebers.
Lastly, I just want to add my support to the achievement gap really being the time on task gap. Firstly, that’s a profound explanation. Secondly, why not measure time on task. Train selected students and program their cell phones to buzz at set times. The students then fill out a form recounting what they see happening around them. Again, be prepared for some bad news. And what if the time on task report takes place as the random testing is being conducted? We could really get some good info. Then teachers, administers, and others would meet with students to analyse the results.
Thanks, John. I always worry about sounding harsh or hard-hearted in suggested home-tutoring for disruptive kids, but you’ve nailed it. I’ve never seen a kid who didn’t respond to one-one-one tutoring (at least in elementary school), and I agree that it would be very cost-effective.
For time on-task, I’d be content with researchers sitting in classrooms (or in front of videotapes) timing teacher and student time on task with a chess clock. I’d wager that students in low-achieiving schools lose about one-fourth to one-third of their instructional time to disruption or executing routines designed to prevent disruptions.
I’d also wager that low-achieving students spend an inordinate amount of time simply not doing anything due to lack of explicit instructions from teachers, but that’s another subject.