relative vs. absolute improvement in policy

Last week, Elizabeth wrote about Finland’s educational system. Like many high performing school systems, Finland relies on a relatively elite teacher corp. However, Elizabeth and other commenters were skeptical that the same approach would work in the US. I.e., you can’t get improvement by firing the worst teachers.  The commenters responded that the issue was that the “new” worst teachers would still be matched with the lowest SES students. This response is not persuasive because it conflates two issues: relative improvement and absolute improvement.

While we would love all students to get exactly equal treatment in school, the most realistic goal for an institution in the short term to seek improvements with existing resources. I.e., the typical school in the South Side of Chicago needs *better* teachers, not the exact same teachers as the elite school in Winnetka. There are two reasons for this.

First, even modest improvements in outcomes matters. The typical low SES school instructor probably won’t have the same effect as the elite math teacher in Winnetka, but improving a graduation rate from 55% to 60% would result in literally hundreds of low SES kids having the high school credential or admission to college. It matters.

Second, in a system of local control, it is not entirely clear why we should expect random assignment of teachers to schools. The way the teaching profession works is that schools compete for teachers by offering higher salaries, nicer facilities, and higher SES students that are easier to teach. One can imagine a Federal system for assigning teachers to schools, but that isn’t coming any time soon. For now, we work with what we have.

Bottom line: Yes, firing the worst teachers will almost certainly increase the educational outcomes of a school. Don’t give up real, achievable gains in an attempt to stamp out all inequality.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz


Written by fabiorojas

June 25, 2014 at 12:01 am

2 Responses

Subscribe to comments with RSS.

  1. Fabio,

    There are many twists and turns between “eliminating tenure”, “firing the worst teachers at the school”, and “increasing the educational outcomes of a school.” I’m not convinced that A will lead to B, nor that B will lead to much of C – even in an absolute sense (let alone in terms of reducing gaps between rich and poor schools). Here’s a useful summary and analysis with lots of links from Lawrence Mishel at EPI, Teachers, “Tenure,” Due Process, and Truly Helping Disadvantaged Children:

    “The due-process protections afforded by tenure, at the very least, ensure that teachers who do stay in high-poverty schools can speak out against these inequities and be advocates for a more just system for their students.

    Scott Lemieux writes that Vergara not only takes aim at the wrong target, it actually harms the low-income and otherwise disadvantaged students it purports to save. “Treu simply assumes, not only without evidence but in the face of logic and reasons, that there is a group of highly skilled teachers waiting to fill the least desirable teaching jobs in the California school system,” he writes, “despite the fact that these aren’t particularly remunerative and, thanks to Treu, now must also be insecure.””

    Is it not possible that the size of this effect – the disincentive from having your job not only be an under-resourced school but also not even having job security at that job turning away potential good teachers – will wash out any gains from the new ease in firing the absolute worst? Do we have any good evidence pointing in one direction or another? What’s the right empirical model?


    Dan Hirschman

    June 25, 2014 at 2:54 am

  2. I agree with Dan. I’d just like to add that the proposal to improve education by firing the “worst” teachers is nothing like what Finland has done. Finland (which, let’s be honest, is a small country with very little poverty) developed its education system by eliminating standardized test scores completely and trusting that a high level of teacher professionalism would improve education. It did. In the US, what is being proposed is to leave the existing teacher training model in place, and then to fire the “worst” teachers based on things like value-added measures (which are calculated using student test scores).


    Zach Griffen

    June 25, 2014 at 1:21 pm

Comments are closed.

%d bloggers like this: