Friday, May 6, 2011

-Problems with education

It is clear to many people that primary and secondary education in the US is not as good as in other developed countries (Finland, Korea, for example). You have probably heard or seen in the news that people in education are trying to implement different changes aimed to improved the system.

Michelle Rhee, in Waiting for Superman argues that the biggest problem is that a lot of teachers are bad and there's no accountability and punishing framework for those teachers who underperform. Wisconsin is going against unions, in order to remove collecting bargaining rights. Others argue that salaries and on-the-job welfare of teachers are under par in the US when compared to countries that in which the schools perform better and that that leads to a high turnover rate and low recruiting standards for new teachers.

If you divide the problem into its four working parts you get: the education administrators, the teachers, students and everything else. Rhee and Wisconsin's governor blame the poor performance on having bad teachers; Eggers and Clements Calegari blame the administrators for not having good work conditions. Maybe the two feedback into one another: bad working conditions lead to good teachers leaving (or never entering the field) which opens the door to bad teachers taking those jobs. The main issue I have with these two camps is the lack of data supporting their statements. You can quote the info presented in Waiting for Superman, but to me, that wasn't hard data. That was an opinion formed by a group of people based on a limited (an biased; will the administration ever blame itself for the low performance? will the teachers?). On the other hand, while it is clear that teachers' salaries are low and benefits might not be great, they have other advantages (e.g. summers off, "tenure", etc). Nobody, as far as I know, had looked at the other two parts and the effects that they have on a kid's education. But that's until now...

Michael Marder, a physicist from the University of Texas at Austin, has done some work in this area too (click here for an interview). I think it's good that someone that's close to the problem but not from within is looking at this stuff, mainly because his approach is based on data and, as a good scientist, he's opened to peer-review and/or rebuttals of his claims:

"There's a flood of data. The schools have been gathering an unprecedented amount of information about what students are doing every single year, particularly in reading and mathematics. [...] Everything I'm showing you can be obtained from the web with downloads. Now, the data sets are typically too large to be put into Excel, and so, my background in nonlinear dynamics helped me because I'm used to processing large data sets from experiments and that meant I had software tools available to me that although free, are not available... easily available to everybody. [...] and I encourage everybody to check the claims that I'm making because it's extremely important that as we go forward in this country we base the decisions on what the data tell us"

So, what does he find?

That poverty levels matter a lot. You can see from the video interviews the graphs that show how the higher the poverty concentration (defined as the percent of students eligible for free and reduced lunch) the lower the test scores. It's also clear that there is a correlation between being a minority student and poverty level.

He also looks at the performance of charter schools and finds, that at best (in most cases), they do about as good (or actually, as bad) as comparable public schools.

Are these results striking?

To me, yes and no. Anyone who thought (or thinks) that poverty levels do not matter in the performance of kids is either completely misinformed or plain dumb. There are plenty of documented examples from different countries in which kids from lower income families perform worse than their middle class counterparts. I am actually surprised that it requires a physicist to put it into a graph for it to gain traction.

The charter school results are somewhat surprising though. You can see that in most cases, charter schools are actually worse than public schools. If you watch Waiting for Superman you'd think that charter schools are the solution to the educational problem, but this goes to show that just because someone in the position of "power" says something is the problem it doesn't mean it is. It may well be that for D.C. schools, charter schools and/or better teacher accountability is the solution, but to go to the federal area and push for policies to that end without having data is ridiculous. On the other hand, you can see that most charter schools have a high poverty level. This argues against the idea that bad teachers alone are the problem, and supports the idea that the learning conditions both inside and outside of the classroom are more important parameters.

Is there anything really surprising?

YES! The fact that a lot of people (including many politicians in office) completely dismiss poverty as a factor. It's much easier for them to blame the teachers, and for teachers to blame the administration, than to admit that social conditions in their cities and states are bad and are affecting education.

How do you fix the problem? Marder says:

"The schools are going down, the schools that serve poor children are not doing well, and I believe, that the fundamental problem just like the cracks in the airplanes, is that poverty introduces flaws into the system of schooling that cause the schools to crash. We have to solve that problem. Now, there are theories for how to do this right now, but I think they are as mistaken as the theories that the British were using as the Comets went down, and I think that the data show it. So, what I urge everybody is, look at the data. If you think that accountability by itself; that improving teacher quality... by itself; and that neglecting the influences of poverty and saying that other reforms will help, then I think the data have a lot of things to tell you that say that you're not right."

I find this extremely courageous. Since he has no stake in it, he can really look at it from other angles and speak out. I guess having tenure at UT Austin, and your job not depending on political appointments, can put you in a position to stand for what you think is true and right. Of course, using hard data and objective analysis to reach your conclusions cannot hurt either.

Now, will the people believe what a physicist have to say? I guess we'll have to wait to know the answer...