Tag Archives: testing

Testing, Testing…

It’s been a while since I posted…is this thing on?

This caught my eye: “States Listen as Parents Give Rampant Testing An F”

It says “States”, but it’s mostly about Florida, an early and eager adopter of high-stakes testing under the aegis of “accountability”, which is now joining the national movement (FairTest.org, UnitedOptOut.com, Facebook) to push back against standardized testing.  And no wonder:

In Florida, which tests students more frequently than most other states, many schools this year will dedicate on average 60 to 80 days out of the 180-day school year to standardized testing. In a few districts, tests were scheduled to be given every day to at least some students.

Of course, Montessori schools don’t have much use for tests and grades:

The bad marks with which teachers weigh up the work of girls and boys is like measuring lifeless objects with a balance, measured like inanimate matter, not judged as a part of life.  (Montessori, From Childhood to Adolescence)

And they are rarely to be found in Montessori schools.

But Montessori was a scientist, and we can’t turn our back on the science of standardized testing.  The math is unassailable, and it really does give you a detailed, unbiased picture of what you’re testing.  It just depends on what you’re measuring.  Montessori students are hard to measure accurately, for reasons discussed here.  One of the few solid studies we have is Rathunde and Csikszentmihalyi’s 2003 work, which accurately compared Montessori and non-Montessori middle school students while filtering out demographic biases such as race and income.  And what did they find?  Montessori students:

  • liked school better
  • felt more active, strong, excited, happy, sociable, and proud
  • enjoyed themselves more, were more interested in what they were doing, and wanted to do schoolwork
  • reported higher interest and motivation
  • experienced more challenge

In the school I work for, at Elementary conferences, parents get a list of lessons the student has had.  Our guides use software which can record a “status” for each lesson—something like “presented, practiced, proceeding with guidance, mastered”.  We recently decided to use only “presented”, for several reasons.  First, the volume of data:  30 children averaging 5 choices a day gives 150 “work events” per day to record. The laptop would have to be out and humming every minute, and we just didn’t want that to be the center of the classroom.  No doubt guides would record less than that, but how could they be sure of a random sample?

Second, we wanted to represent to parents the reality of what our guides do, which is give lessons.  The work afterwards is the child’s responsibility.  Guides know each child’s work intimately, of course, and they would be happy to tell you how Jason followed up on his Viscosity lesson, or sit down with you and Amber’s work journal. That’s what the conference is for.

Some parents were concerned about what goes on to middle school.  And that got me thinking about the difference between what we do, and grades and test scores.

A grade or score says this:  We warrant that, on this day at this time, this child was able to say or do or calculate these things. It doesn’t say anything about what the child knows or can do today, because no one can know that.

What we do is this:  We warrant that, over the last three years, this child has had all these lessons! And we warrant that we gave them in such a time and in such a way that she could not help but to be inspired and awakened.  Don’t believe us?  Just ask her.

And that’s why content testing is always going to be a challenge for Montessori schools.  Sure, we teach content, and we have the public school requirements as part of our classroom environments.  If the testing requirements will recede to a saner level, we can probably still do most of what we do and our children will pass the tests because they love to know things

But our idea of the whole purpose and program of education is radically different.  In the conventional model, education is about delivering content.  Adults decide what content, and when, and how, and how much, and children stand by to receive and be tested for their successful retention of content.

In Montessori, education is about human development.  That’s it, really: the understanding that humans have an inherent optimal path of development, which includes intellectual curiosity and the acquisition of culture, and that development can be helped or hindered.  Devise a test that can measure that, and we’ll be sure to excel.

Testing, Testing … (Is This Thing On?)

You may have heard about the gigantic testing scandal in Atlanta public schools.  Beginning in 2008, old-fashioned investigative reporting by the Atlanta Journal Constitution uncovered widespread cheating by adults on standardized tests in Atlanta Pubic Schools. The tests, instituted by the No Child Left Behind law, were closely tied to school funding, performance bonuses, and continued employment for teachers. About 150 educators have resigned, retired, or lost their jobs.  This week, 35 educators including former superintendent Beverly Hall were indicted for racketeering, theft by taking, influencing witnesses and making false statements.  

What did we expect?  If we tie funding and compensation to test performance, and then ratchet up the economic and political pressure, this “outcome” is entirely predictable. The Constitution’s investigation indicates that “196 districts throughout the U.S. exhibit suspicious patterns of test scores that, in Atlanta, indicated cheating.”  A sampling of opinion from around the web:

Eugene Robinson in the Washington Post:

It is time to acknowledge that the fashionable theory of school reform — requiring that pay and job security for teachers, principals and administrators depend on their students’ standardized test scores — is at best a well-intentioned mistake, and at worst nothing but a racket.

CheatingCulture.com editor David Callahan in the Huffington Post spells out the role of No Child Left Behind and ties the scandal to our larger cheating culture in this country.

And Erika Christakis at Time.com uncovers a hidden dimension to the scandal:

faked scores prevented some schools from accessing three quarters of a million dollars in federal money to support struggling learners because they no longer qualified for help. 

Christakis goes on to challenge the testing itself:

Even if we eliminate all the cheating, what remains is a broken system built on the dangerous misconception that testing is a proxy for actual teaching and learning.

Definitely worth a read.  In fact, from looking at her blog, this sounds like someone who might be interested in learning a little about Montessori.  I think I’ll send her a link.