Friday, February 19, 2010

Summary of Alfie Kohn's Keynote in Calgary, Alberta, Canada

Palliser Teachers'Convention was keynoted by Alfie Kohn at the Telus Convention Center in Calgary. Here is a brief summary of his hour long talk:

Kohn's primary message revolved around why achieving high scores on Provincial Achievement Tests is actually a really bad thing for kids and schools. Kohn describes how today's high-stakes accountability fad gets four aspects of education wrong.

Today's accountability fad gets school reform wrong. Today, more accountabilty is simply code for more control for those who are not in the classroom upon those who are in the classroom. This kind of top down reform is ultimately undemocratic and will be one day viewed upon by educational historians as a truly dark time in the history of schools. This brand of school reform equates to a 'do what we tell you or else' kind of treatment of teachers and students. Rather than working collaboratively together with teachers, policy makers end up doing things to teachers to bribe or threaten them to comply. And in turn, teachers who feel controlled in turn become controlling.This kind of school reform allows those who are the most ignorant of what good learning looks like to use their political power to mandate their ignorance into law.

Today's accountability fad gets improvement wrong. There is a misconception among many that harder equates to better. We mistakenly describe education as being desirably rigorous when we should be talking about making learning more vigorous.

Cliches like 'raise the bar' and 'higher standards' at first glance seem to make sense; however, if we were to stop and speculate - what would the government say if every child actually scored proficiently on this year's Provincial Achievement Test? Might they respond, "Wow, those teachers sure are doing a fine job." This is laughable. It is far more likely that the response would be something like, "Those teachers are too damn easy on those kids." What this tells us should disturb you - the higher standards movement, by defiinition, is designed to ensure that not all children can achieve them, because if everyone could achieve them, that would be proof that the standard was simply not challenging enough.

Today's accountability fad gets teaching and learning wrong. It gets teaching wrong because it encourages the transmission of a bunch o' facts kind of curriculum. Kids become vending machines - teachers simply input the curriculum into the kid's heads and they will spit out high test scores. The more emphasis we place in raising these scores, the more we emphasize the worst kinds of teaching - sit, get, spit and forget.

The other problem that arises from overemphasizing the need to raise these high-stakes test scores is that the more you prepare students for these exams, the more time we teach to the test, the less valid the scores become. The very high-stakes nature of these kinds of tests work to make them invalid and unreliable which leads policy makers to making bad decisions based on bad data.

Today's accountability fad gets evaluation wrong. We can not measure our way to a better education system. Assessment should never be equated to simply a way to better judge students and schools. If we really care about 'closing the achievement gap' or 'leaving no child left behind' then we have to stop using test scores as an indicator of good teaching or learning.

Keep in mind test scores are a good indicator... of the size of homes that surround the school! If given the income levels of a set of parents, it is very plausible that one could accurately predict their children's standardized test scores with chilling accuracy. Research has shown that 50% to 90% of the variation in student learning is not attributable to school factors but to student, family and community characteristics. This means that too many tests are reporting on what children bring to school and not necessarily what they learn at school.


So when we see test scores go up, we should respond in one of two ways:


So what! We know that these tests measure what matters least and that their results are not indicative of the kind of quality teaching and learning that takes place in the classroom, in the school, in a school district, in a province, or even in a country.

Or we should respond:

Uh oh. What kinds of real learning did you have to sacrifice in order to get those children to score so high on that standardized test. How much less time are they spending on field trips, art, music, recess...


As teachers and parents, we need to be on two tracks of mind. In the short-term, we need to 'play the testing game' as little as possible to ensure that our students succeed in gaining an education, while in the long term we need to get political and engage our School Board and MLA's in a discussion about how focusing on standardized test scores is subverting our best efforts to educate our children.

3 comments:

  1. Do you know anything about Alfie Kohn's kids and school? Where they go or went? Do you happen to know his thoughts on homeschooling?

    ReplyDelete
  2. He does not speak about his family too often. He is quite sure to keep his private life private, but I believe he has mentioned that his kids were home schooled. But don't quote me on that.

    ReplyDelete
  3. Thought provoking post. Thank you. I apologize for the longish comment in advance.

    I was struck by "Research has shown that 50% to 90% of the variation in student learning is not attributable to school factors" If that can really be supported as stated, it says to me that school, as presently organized, might be a huge waste of money.

    Given the amount of resources, energy and focus societies put on schooling it should be taken seriously. If true it undercuts much of public discourse and expenditures.

    As a global stat, I don't trust it. It's more likely that some schools do add to student learning and others don't. In the States 1500 high schools create over 50% of the High School pushout/dropouts. Taken at the scale of the nation, we have a "high school dropout" crisis.

    In fact, it's more like an epidemic with centers of virulence. Not surprisingly, it overlaps with poverty. Concentrations of recent immigrants and 2 or 3 generation poor have massive dropouts rates. I heard recently on twitter that the graduation rate in Baltimore is 35%. (have not confirmed.)

    My question is what does Kohn use as his measure of learning? No doubt standard high stakes tests are exactly as described. Tools for politicos who think they are doing the right thing, but deeply flawed. Fair enough.

    But at the system level we need numbers to help inform intelligent decisions. In the absence of tests, any thoughts about the best way to collect and analyze unambiguous numbers.

    I've been thinking about using attendance as a plausible real time indicator of success. The notion is that "learning" in principle can not be an unambiguous metric. Complexity does not lend itself to simple numbers.

    Attendance in poor middle and high schools is both unambiguous and reflects the students "vote" about whether a school is worth going to. If it is reframed as a indicator of school success instead of a judgment on the students and their parents I think it works.

    Any thoughts from you and other visitors to the blog would be very appreciated.

    ReplyDelete