Fresh Look at Coleman Data Yields Different Conclusions
What would the landmark Coleman Report show if the numbers were reanalyzed today using more sophisticated statistical techniques? According to Geoffrey D. Borman, a researcher from the University of Wisconsin-Madison, the results would be markedly different.
Mr. Borman, an associate professor of educational leadership and policy analysis, educational policy studies, and educational psychology, got the idea to reanalyze the report’s findings in the mid-1990s, when he was a graduate student at the University of Chicago. James S. Coleman, the famed author of the 1966 report, was on the university’s faculty before his death in 1995, and Mr. Borman had been a student in the last class he taught there.
The idea continued to haunt the young researcher a few years later, when he began work as an associate research scientist at the Center for Social Organization of Schools, at Johns Hopkins University in Baltimore.
Mr. Coleman founded that center in 1966 partly to house the federally commissioned study, known more formally as “Equality of Educational Opportunity.” And Mr. Borman’s office at Johns Hopkins overlooked a building that had been used to store the IBM 1401-model mainframe computer that analyzed the Coleman data.
Mr. Borman found the original data files from the report at the Inter-university Consortium for Political and Social Research at the University of Michigan in Ann Arbor. With N. Maritza Dowling, a colleague at the University of Wisconsin, he weeded out and reanalyzed data collected on a subset of 56,000 9th graders who took part in the original survey.
Contrary to the findings of the original report, Mr. Borman concluded that school factors matter a lot in determining how students fare academically. The researchers for the reanalysis found they could attribute up to 40 percent of the variation in achievement differences between students to such factors, rather than to differences in students’ family backgrounds, such as how educated their parents were or how many books were in their homes.
“This really contradicts some of Coleman’s findings,” Mr. Borman said. “He attributed so much of the difference in achievement outcomes to what students brought with them to school.”
The reanalysis also suggested that whom students went to school with was more important than the color of their own skin or their families’ income levels.
“Being poor and being African-American is important for understanding achievement outcomes, but attending a predominantly African-American school or a predominantly poor school is one-and-a-half times more important,” Mr. Borman said.
Mr. Coleman found a similar link between achievement and the demographic makeup of schools, but the effects that he documented were smaller and linked more to social class than to race.
“It’s clear from these analyses that racially segregated schools compromised African-American students’ ability to achieve educational outcomes comparable to their white peers’,” Mr. Borman said.
“This kind of evidence may have been sufficient to hasten the implementation of Brown v. Board of Education,” the U.S. Supreme Court decision that struck down school segregation, he said of the findings from the reanalysis. “Who knows?” he added. “Maybe these findings would’ve hastened the advent of voluntary desegregation and avoided mandatory busing.”
Teacher Bias Explored
In addition, the researchers rehashed some of the teacher-survey data from the Coleman study and concluded that teacher biases may also have contributed to achievement disparities among students.
Student test scores were lower on average in schools where the typical teacher had expressed a preference for teaching middle-class students. The new researchers also found that achievement gaps between black and white students in the same schools tended to be larger in the schools where most teachers had expressed preferences for teaching college-oriented students or the children of professional and white-collar workers.
“That was overlooked in the Coleman Report because the main approach was on exploring differences that occurred across schools rather than specific differences within schools,” Mr. Borman said.
James M. McPartland, a Johns Hopkins professor who was part of the original research team led by Mr. Coleman, said in an interview last week that he considered Mr. Borman “a very good researcher,” and that his findings on teacher attitudes were indeed new.
Yet he said he regarded much of the reanalysis as principally reflecting “nuance differences.”
“This is a matter of degree rather than a contradiction,” said Mr. McPartland, who is the director of the Center for Social Organization of Schools.
Mr. Borman presented his findings at the annual meeting of the Washington-based American Educational Research Association in 2005. A revised version of the study is currently being reviewed for publication in a peer-reviewed journal.
Vol. 25, Issue 41, Page 21Published in Print: June 21, 2006, as Fresh Look at Coleman Data Yields Different Conclusions