July 19, 1998
The State of Ohio, in an attempt to improve our schools, is spending millions of our taxpayer dollars proving what educational experts have known for years - children from high socioeconomic homes have a better chance of succeeding at school.
On June 11th, The Dispatch published the results of application of new state mandated uniform standards for schools in a survey of all central Ohio school districts. As a Granville parent, resident, taxpayer and homeowner, it did give me a little personal pleasure to know our district was the only one passing all 18 standards. As a former consultant to the Columbus Public Schools, it bothered me to see that district bringing up last place. As a researcher, however, it was obvious the new process of uniform state standards is terribly flawed.
Educational researchers have known for years that the best predictor of a child's success in school is the education level of that child's parents, particularly the mother. Although human beings are born with very similar ranges of intelligence, the different nurturing processes which take place in the formative years greatly impact a child's ability to learn. Different levels of protein in the diet and stimulation in the home have a great impact on the formation of synaptic contacts in the human cerebral cortex, that is, brain development. This was recently documented again by a major study of the Departments of Pediatrics and Neurology at the University of Chicago. The point of all this is that children who are born with similar ranges of intelligence arrive at school with a wide range of ability.
When the new Ohio data was released, we used our SchoolMatch databases to compare family resources (education level and income) against the number of standards met by each of central Ohio's school districts. The problem with uniform standards for dissimilar school districts may have never been better demonstrated than in the case of application of the new Ohio system to central Ohio school districts. We compared each of the district's family education level and income level to the number of standards met. The data depicted nearly a straight line correlation - those districts with combined very high family resources (Granville, Dublin, Olentangy and Upper Arlington) met the most standards. The districts in the middle (Circleville, Groveport, Lancaster) met about half the standards. Districts with the least amount of family resource in terms of parent education and income level, like Columbus, fared worst.
Over 10 years ago, our advisory board chairman, Dr. M. Donald Thomas, highly respected former superintendent of schools in Salt Lake City, Utah, and education advisor to several governors, was making a presentation to a joint session of the South Carolina legislature. Don explained to the policy-makers that goals should be based on the performance of "mean-matched" schools.
When asked to explain the term, "mean-matched," Don explained that schools should be compared with those schools which had similar characteristics. "You don't want to compare apples with oranges," he said.
The "Audits of Educational Effectiveness" which we have completed for the Columbus and Olentangy districts used a "mean-matched" process which considers socioeconomic and demographic characteristics of similar groups of students in similar school system settings throughout the country. Don Thomas has been working in a similar vein with the Newark City Schools, and collectively we have worked with over 700 school districts nationwide.
In their missionary fashion, senior SchoolMatch consultants such as Dr. Alonzo Crim of Atlanta and Dr. Richard Koeppe of Denver have converted a significant number of school leaders to the "mean-matched" concept. School systems have used the methodology to identify "sister districts" with similar student populations around the country with which they can compare programs and results.
Irrefutable research tells us that children can learn if they are provided adequate time and resources. The "mean-matching" process is intended to only compare groups of children in school settings, and obviously should never be used as an excuse for overlooking poor group performance or not expecting outstanding individual student achievement. Clearly all children are capable of success in school, given enough time and resources.
Because children do not start school with the same developmental levels, however, it is not a fair measure of the work of teachers, administrators or support personnel to compare schools unless the resources provided to the children at home are taken into account.
Regrettably, the new state reports are measuring family resources more than good teaching and effective school management.
is Distinguished Research Professor at the
University of Dayton and is President & Chief Executive Officer
of SchoolMatch®, a Columbus based educational auditing, research, data