Ohio Researcher Proves–Yet Again–That Test Scores Measure Primarily Family Income

Like the rest the country, Ohio is trapped in a test-and-punish education accountability system that castigates public schools when students’ test scores are persistently low. It is a system that punishes already vulnerable institutions—closing and charterizing schools in the places where scores are low, giving vouchers to help children “escape” so-called “failing” schools, and rating teachers by students’ scores. Ohio practices all of these policies.  The flaw inherent in such a system is that standardized test scores continue to correlate with the aggregate income of the families whose children are enrolled in schools and school districts.  Researchers have demonstrated again and again that, despite that poor children surely can learn and many do thrive academically, aggregate test scores are pretty much an economic indicator, not a measure of the academic quality of the school.  Test score gaps are in place before children enter Kindergarten, and they rarely close as children move through the grades.

Now, once again, Howard Fleeter of Ohio’s Education Tax Policy Institute, has documented that in Ohio, the schools that can brag of the highest test scores are located in the wealthy suburban school districts and the so-called “failing” schools are those that serve children living in poverty.  The ratings attached to school districts by the state based on test scores thus create further incentives for more families to abandon poorer and mixed income communities and move to expensive outer ring suburbs. This blog recently covered how school ratings by Zillow, the online real estate guide, contribute to segregation in the same way.

Here is how the Cleveland Plain Dealer‘s Patrick O’Donnell describes Fleeter’s findings: “State test scores continue to rise right along with a school district’s affluence, and fall as poverty rates increase….  Ohio may have changed academic standards and its state tests last school year, but the recurring relationship between test scores and poverty remains the same…. Fleeter has reported the relationship between test scores and family income on an annual basis the last several years…. He repeated that analysis this week using preliminary test scores from the spring on Ohio’s new math, English, science and social studies tests…. As he does each year, Fleeter compared the percentage of students scoring ‘proficient’ or better on state tests in each school district to the percentage of students considered ‘Economically Disadvantaged’….”  And just as he has found previously, aggregate test scores correlate with the income level of the families who reside in the school district.

Fleeter’s analysis, according to O’Donnell, demonstrates that last spring when Ohio administered PARCC tests of language arts and math and tests designed by the American Institutes for Research to measure science and social studies, the highest-scoring 20 percent of districts (serving only 17.4 percent of students who qualify for free-or-reduced-price lunch) had 86.2 percent of students scoring proficient in math.  The bottom-scoring 20 percent of school districts (serving 77 percent of students who qualify for free-and-reduced-price lunch) had only 44.3 percent of students scoring proficient in math.

The press release from the Ohio School Boards Association, the Buckeye Association of School Administrators, and the Ohio Association of School Business Officials (who sponsored the new study) quotes Fleeter: “This means that those districts performing the best on the tests have the lowest percentage of economically disadvantaged students, while districts with the lowest performance have the highest percentage of economically disadvantaged students.”  Posted with the press release are the tables that represent Fleeter’s analysis.

While Fleeter’s conclusions should sound an alarm in a state that uses test scores as the primary factor to determine “A” through “F” grades in a punitive accountability system for schools and school districts, Fleeter merely proves once again what researchers have been demonstrating for half a century.  Last year, for example, in Our Kids, after exhaustively examining the experiences of children in schools through the lens of wealth vs. poverty, Harvard’s Robert Putnam concludes: “(T)he gap is created more by what happens to kids before they get to school, by things that happen outside of school, and by what kids bring (or don’t bring) with them to school—some bringing resources and others bringing challenges—than by what schools do to them.” (p. 182)

Here are the words of the respected researcher David Berliner, in a 2014 article published by the Teachers College Record: “For reasons that are hard to fathom, too many people believe that in education the exceptions are the rule… (S)tories of triumph by individuals who were born poor, or success by educators who changed the lives of their students, are widely believed narratives about our land and people, celebrated in the press, on television, and in the movies. But in fact, these are simply myths that help us feel good to be American.  These stories of success reflect real events, and thus they are certainly worth studying and celebrating so we might learn more about how they occur.  But the general case is that poor people stay poor and that teachers and schools serving impoverished youth do not often succeed in changing the life chances for their students… (O)ut-of-school variables account for about 60% of the variance that can be accounted for in student achievement.  In aggregate, such factors as family income; the neighborhood’s sense of collective efficacy, violence rate, and average income; medical and dental care available and used; level of food insecurity; number of moves a family makes over the course of a child’s school years; whether one parent or two parents are raising the child; provision of high-quality early education in the neighborhood; language spoken at home; and so forth, all substantially affect school achievement.”

In the 2013, Closing the Opportunity Gap, two researchers—Christopher H. Tienken and Yong Zhao explain: “(A)s a group, students labeled as economically disadvantaged or poor never score higher on standardized tests than their non-disadvantaged peers in any state on any grade level currently tested under NCLB.” (p. 112)

When are we going to pay attention to what has been conclusively documented?  Instead in Ohio, public school districts serving poor children are punished in several ways by state policy.  Children in so-called “failing” schools qualify for EdChoice vouchers that extract money from a school district’s budget for every child who takes a voucher to a private or parochial school.  The new Youngstown Plan, quietly fast-tracked through the legislature in June, permits the takeover and charterization of so-called “failing” school districts. It is aimed at Youngstown schools this year, but any district with three years’ of “F” ratings will be targeted in the future.  And Ohio, like all the other states that applied to the U.S. Department of Education for a waiver from No Child Left Behind, had to incorporate students’ test scores as a significant part of state teacher evaluations. We don’t merely blame the teachers; these days we formalize our blame right into each teacher’s personnel file.

Describing Fleeter’s new data connecting Ohio students’ test scores to their families’ economic circumstances, O’Donnell quotes the press release from the study’s sponsoring organizations, who say they will use Fleeter’s new data as a way, “to push for more focus on the educational needs of these students by urging lawmakers to look for additional ways to address these significant disparities.”  O’Donnell notes that in the past these organizations have used Fleeter’s research to advocate (unsuccessfully) for changes in the state funding formula to help poorer school districts.

What, to my knowledge, has never been seriously contemplated by Ohio’s legislature or state department of education is changing the state policy that, based on students’ aggregate test scores, punishes poorer school districts (with low state ratings and rankings, vouchers, and threatened state takeover) and punishes their teachers for choosing to teach in low-wealth school districts.  In an ironic twist in Ohio, this month the charter schools Ohio dubs “dropout recovery schools” are campaigning to do just that.  Ohio judges charter schools by aggregate test scores as well, and these schools allege Ohio should change the system that punishes them for serving students who come from so-called “failing” schools in the poorest urban school districts.

I would certainly be satisfied if Ohio found a way to design a new evaluation system that accurately examined schools themselves apart from the income-correlated test scores of the students.  But any new system must decouple state evaluations of schools from standardized test scores in both traditional public schools and charter schools. Any new system also needs to put the spotlight on disparities in inputs, not just on test score outcomes.  How much revenue is each school district capable of raising from its tax base?  How well is the state compensating for inequity?  What about small classes, access to advanced curricula, guidance counselors, art and music?