Colorado's Amazing New Toy

May 27, 2009

Alright, I'll admit it, I'm a geek. I love playing with numbers, watching patterns develop, seeing if data can help us predict the future in everything from baseball to student achievement. I'm also as much a fan of new technology as the next person, so when you give me a new toy that blends numbers, imagery, and cool tech, well I'm pretty much in heaven.

So all of this makes Colorado's amazing new toy something that I've grown a bit addicted to in the last few hours. What is the toy, you ask? Click the link and have a see for yourself (but don't worry, I'll give some more info as well). If you've decided to click on the link, don't worry about watching either the introduction or tutorial video, instead click "select by name" and start playing right away by clicking on different cities and seeing how the schools there stack up.

If you'd rather just read what this whole thing is about, I'll start here: the toy is called the "Colorado Growth Model Public View" and it's pretty much the most important public data system concerning school quality that's every been released. Here's the long and short of what it does:

  1. First, it does something run-of-the-mill that plenty of other public data sets on school quality do (such as, and others): it tells you how well a given school's students do on standardized tests. Take Trailblazer Elementary School in Colorado Springs, for example. Of the 328 students enrolled in the school, 82 percent of the students are proficient or better on the math portion of Colorado's Student Assessment Program. Pretty good, right? Compare that to Bruce Randolph High School in Denver, where of the 680 students, only a paltry 5 percent were at or above proficient in math. Not as good a school, right? If all you had was some of these other websites or the data available in any other state, you'd quickly draw the conclusion that Trailblazer must be a good school and Bruce Randolph must be a bad one. But not so fast...
  2. The second thing Colorado's Growth Model does is show something that no other public website shows: how much a given school's students have actually learned over time. Here, suddenly Trailblazer Elementary School doesn't look so good: it's only in the 24th percentile of student growth, meaning that three out of every four schools in the state are improving their students' learning ability more than Trailblazer! Turns out the parents who are excited to be sending their kids there are getting a bum deal: their kids may start out at a high achievement level but they'll regress towards an average level by the time they finish elementary school--or worse. And what about Randolph High School? Turns out it is in the 70th percentile for improving student achievement, meaning that it may be getting a bunch of students who start out below grade level in the 9th grade but the school seems to be doing a pretty darn good job at bringing those students up to speed.

So what kind of school did you go to? Unless you grew up in Colorado, it may be some time before you find out for sure. But ask yourself two questions:

  1. Did most of the students at my school have a good chance at passing the standardized tests?
  2. Did most of the students learn a lot every year, regardless of whether they would pass the tests or not?

You might think the two questions are asking the same thing, but they're not. The first question often gets to background characteristics of the kinds of students attending a school and not the school itself--in other words, even a pretty crumby school would have a hard time taking Bill Gates's 9 year old daughter and educating her so poorly that she would fail the 3rd grade math proficiency test. But the second question is arguably more important: how much are the students learning each year? A school in a tough neighborhood may be teaching a lot to its students without much success on the first question (i.e. Bruce Randolph School), and a school in a great neighborhood might be failing its students by failing to push them to do much more than be proficient.

Oh, and in case you were wondering, Bruce Randolph School was found to be such a miracle case that Secretary of Education Arne Duncan visited it earlier in April to tout the power and potential of school reform. Quite simply, the numbers don't lie.

Aaron Tang is the co-director of Our Education, a non-profit organization working to build a national youth movement for quality education. He also teaches 8th grade history in Saint Louis, MO.





Aaron, thanks for the kind words regarding our Colorado Growth Model interface. A great many people have worked very hard on this project, and it is rewarding to hear positive comments from education practitioners such as yourself.