Monday, January 28, 2013

Oklahoma's A-F Grading System Get's An "F"

Recently I read, “An Examination of the Oklahoma State Department of Education’s A-F Report Card”, commissioned by the two largest professional school organizations in the state:  the Oklahoma School Boards Association (OSBA) and the Cooperative Council for Oklahoma School Administration (CCOSA).  If you kiddos out there haven’t gotten a chance to read it yet, don’t worry, I’ll bet it will show up on the Common Core State Standards reading list with all those other exciting non-fiction reads kids are going to enjoy so much, but that’s another story.

I read the paper after reading the Daily Oklahoman editorial “Oklahoma Policymakers Shouldn’t Retreat From A-F grading System”, because it seemed a very partisan assessment.  Now don’t get me wrong, I have an opinion on this issue as well, in opposition.  After much study of the No Child Left Behind waiver (first request, final request) for which our state applied, our organization sent the offices of the State Superintendent and the Governor letters explaining why we were in disagreement.  

Our concerns stemmed from a number of sources.  As a Conservative organization, we objected to President Obama’s notion that he could ignore the Constitution and go around Congress to make his own law (waiver) simply because he believed Congress would not act to reauthorize the NCLB.   We also highly objected to the fact that results of the A-F system would cause districts to lose control of their schools to the State Department of Education and whatever contractor they saw fit to administrate those ‘failing’ schools.

It should be said, however, that after years of writing and researching education issues, I have become a near-complete cynic of either political party’s actual interest in education for the sake of educating children.  Consequently, I began to assess the article by first researching its authors.  

This gave me no sense of satisfaction as both were clearly entrenched in the University system and one is currently employed by an organization with clear Progressive leanings.  I moved on to research the origins of the citations made by the authors in support of their thesis.  Yuck.  Linda Darling-Hammond.  Darling-Hammond is a close friend of former Weather Underground terrorist Bill Ayers, so anything she says is suspect to me.  Truly, from the beginning it didn’t seem as though I would find anything we could support in this assessment.

I have a Master’s Degree in Biology. In order to conclude my degree I was FORCED to take at least one statistics course.  Not only am I no statistician, I don’t even play one on TV, so reading the paper in its entirety was daunting.  Yet, once I began, I found most statistical jargon had been replaced by or supplemented with understandable correlations that could make sense to readers with even a basic understanding of statistical theory.

I was frustrated by my takeaway and here it is:  Any time authors of a research paper examining a particular statistical method articulate phrases such as, “perhaps a decision was made”, “no clear justification”, “basis undocumented”, “nonlinear relationship”, “basis assumes” and/or “threats to validity”, there is a problem – and not with the examiner, but the examinee.

It seems shocking, given the emphasis on STEM (Science Technology Engineering and Math) initiatives as a way for the Obama administration to pump more money into schools, that our own SDE doesn’t seem to understand the concepts of validity and replicability as part of scientific inquiry in any form – certainly in the creation of a statistical rubric the use of which can mean life or death of a school.

The rubric that determines the A-F letter category for each school is broken into four components; Student Achievement (33%), Individual Student Growth (17%), Growth of Bottom Quartile (17%) and Whole School Performance (33%). 

Let’s look at some of the most central concerns.
Calculation of Student Achievement

“The scores assigned to represent proficiency levels…do not seem to correspond to any recognizable metric”.  This goes to the necessity for a study to be replicable.  If a study cannot be replicated, it can’t be reliable.  In this case it was impossible for the authors to decide how the categories of Proficiency Levels were determined because there was “no clear justification” for why these categories were assigned.  This alone should be the nail in the coffin of this A-F system as categories used to assign specific value are apparently arbitrary. 

In addition, the “basis for letter grade conversion is undocumented”.  How is it that an “A” is designated as having the Performance Index Range of 90-120, but all other letter designations merit a 10 point range?  Again the sense of arbitrariness challenges the concept of validity.
       Individual Growth
      Proficiency Levels (Limited Knowledge, Satisfactory, Proficient and Advanced) are introduced into the metric to allow parents to make assumptions about student academic Growth within the school.  Proficiency Levels are assigned a numerical value and the subsequent changes in numerical value determine the degree of student academic Growth.  Because of the way Proficiency Levels are assigned in the rubric, there is no way to determine student grown within a proficiency level - “no change and negative change are accorded the same zero point value”.  
Students could go from the top of the Advanced category to the bottom, or from the bottom of the Limited Knowledge category to the top, yet no change would be reflected in the actual letter designation for the school.   Not only that, but because academic growth and proficiency levels are non-linear in nature, there can be no extrapolation from individual student to school or even classroom.  There is no way to make scientific conclusions about data that have no real cause and effect relationship.
Growth of Bottom Quartile
      Since this is a very small category, small sample size can introduce bias that may not be able to be corrected away by calculating a confidence interval, making the measure unreliable.
      Whole School Performance 
      Attendance alone determines this component in elementary schools.  This component is a full third of the entire rubric, yet there is absolutely no scientific study which shows that students learn simply by benefit of being present in a school.  How is this measure relevant?

I could continue critiquing the author’s critique, but frankly there is no reason.  As I said previously, the entire system should go the way of the Porcelain God based upon the fact that the majority of the categories and values assigned are neither replicable nor valid.

Though every informed citizen has (or should have!) an opinion about the practice of public education –education isn’t, nor should it be, a partisan issue.  It saddens us here at ROPE, however, that state establishment Republicans continually circle the wagons around our Superintendent and Governor as they push the big government ideals of the Obama Administration while marginalizing those of us who dare to contest this phenomenon.  Interesting, isn’t it?  One can only be educated by learning, and one can only learn by opening one’s mind to possibilities outside one’s own orthodoxy, yet, we see very little of that from Republican bureaucrats in Oklahoma dedicated to “education reform”.  We here at ROPE look forward to the day we are able to witness the Republican “education reform” establishment practice what they preach.  We can only ‘hope’.

BIBLIOGRAPHY (in order of use)

Linn, Robert Lee and Sternberg, Robert J.   (January 2013) 32 pp.  An Examination of the Oklahoma State Department of Education’s A-F Report Card.

Ravitch, Diane (Dec. 31, 2012)  Common Core And The Fiction Non-Fiction Question.

Editor, Daily Oklahoman Online. (Jan. 22, 2013)

White, Jenni.   Bibliography, Literature List for NCLB Research

Department of Education.  (Feb. 8, 2012) WWW2.Ed.Gov.

Oklahoma State Department of Education website.  (Feb. 8, 2012) Ok.Gov.

Restore Oklahoma Public Education.   (Sept. 8, 2011) Letter to Education Officials – Please Do NOT Apply For An NCLB Waiver For Oklahoma.

Kelsey, Mike and Grimard, Leslie.  (Sept. 27, 2011)  Obama Waives Congress Away.  Heritage Foundation.

Restore Oklahoma Public Education.  (May 2, 2012) A-F Grading System Disapproval.

Chumley, CC.  (May 1, 2007)  No Shock, Just Facts: Liberal Indoctrination on Campus.  Human Events.

Garner, Donna.  (January 6, 2013)  Linda Darling-Hammond and Bill Ayers are closely associated with Columbia Teachers’ College where the social justice agenda is heavily enmeshed throughout the College.  Education News online.  (Nov. 23, 2009)  President Obama Launches “Educate to Innovate” Campaign for Excellence in Science, Technology, Engineering & Math (STEM) Education.

Shuttleworth, Martyn.  (2008)  The Principles of Validity and Reliability are Fundamental Cornerstones of the Scientific Method.

White, Michael.  (Dec. 29, 2010)  Scientific Method in Decline?