Are We Failing At Grading Schools? Ctd

Andrew Sullivan —  Aug 3 2013 @ 2:44pm

A reader writes:

The recent letter you published in (qualified) defense of Tony Bennett relies on some claims that are pretty well debunked in this Slate article by Jordan Ellenburg.  The state already had a system in place for dealing with the “non-traditional” schools in Indiana, one that would not make the grade for Christel House inherently unfair.  As Ellenburg writes, Bennett intervened after the odd “grade 3-10” issues of measurement had been worked out, and he created an entirely new loophole for the school.  Basically, rather than using a different weighting system, Bennett just got rid of the bad scores of 9th and 10th graders, and then gave the school an A even though with these scores removed it STILL earned a B.

In other words, while your education-employed correspondent may be right in general, in this case Bennett was intervening after the statistical work had been done to account for the factors that made Christel House “unique.”  This is a pretty clear-cut case of politically motivated intervention, not just an accident of a complex school accountability scheme.  These systems may have honest problems, but Tony Bennett is a dishonest one.


NPR yesterday had a fairly good report on Tony Bennett. Showing “both sides”, as they do, they gave Bennett’s side of the story:

After his system gave the friend’s charter school a “C”‘ he quietly ordered corrections (the school doesn’t graduate students, since it only handles two grades under 12th grade). With the “fix” made, the system upped the grade to an “A”. Good story, but it doesn’t cover everything. NPR then got a rundown of the situation from Mike Petrilli, VP at Thomas B. Fordham Institute who’s been critical of Bennett policies in the past. Petrilli slammed the school and Bennett, saying the school was failing Indiana’s state standardized tests.

Another reader, who agreed to forgo the Dish’s default anonymity policy, writes:

My name is Lili Lutgens and I am President of the Board of Directors of Community Montessori Charter Public School in New Albany, Indiana. We are a pre-K through 12th grade charter public school that uses Montessori teaching techniques to offer families in our community an alternative to traditional public education. Like Christel House Academy, we added one grade per year until 2010 when we graduated our first class of seniors. We also take children who have struggled academically in the traditional public education environment and so we take a lot of kids who are not very good test takers.

But unlike Christel House Academy, our accountability scores have never taken any of this into consideration.  Instead, we are judged solely on our students’ test taking performance and graduation rates.  Never mind that we are offering an alternative for kids struggling in a traditional academic setting and that historically in any given year we have had a significant number of new students.

I guess we just don’t have the kind of money to donate that it takes for the state to give us the benefit of the doubt.

By the way, if you want more info on exactly why the standardized tests underlying these accountability scores are problematic, you may want to look at the work of W James Popham, professor of academic testing at UCLA.  In a nutshell, the tests do not measure instructional quality but are used to judge it anyway.  Popham delivered an important paper at the annual meeting of the National Council on Educational Measurement last year, and it can be found here [pdf].