In the ongoing struggle for school improvement, few groups get to play the role of bad guy as often as the teachers’ unions. Witness such recent books as: Power Grab: How the National Education Association is Betraying our Children and The Teacher Unions: How the nea and aft Sabotage Reform and Hold Students, Parents, Teachers, and Taxpayers Hostage to Bureaucracy.
Given such titles, readers might be surprised to find teachers’ groups wearing the white hat in a new study in this winter’s issue of the Harvard Educational Review.
The authors’ nationwide analysis examined student results on the SAT and ACT college-entrance exams and the proportion of teachers in each state who work under collective-bargaining agreements. What the study found was a significant positive correlation between the degree of unionization in a state and how well its students fared on the tests. Moreover, that link remained after controlling for such factors as family income and parents’ education.
“When we started working on this project, we thought there’d be no relationship,” said Brian Powell, a sociology professor at Indiana University Bloomington. “We were really, to be honest, shocked that we found a positive effect. And the more analysis we did and the more runs we made, it was surprising how robust the link was.”
Mr. Powell carried out the study with fellow IU professor Robert M. Carini and Lala Carr Steelman, a sociologist at the University of South Carolina-Columbia.
The study is not the first to question the view that what’s good for teachers’ unions is bad for students. In 1987, researchers Randall W. Eberts and Joe A. Stone found that students in unionized school districts enjoyed a small but statistically significant edge in mathematics scores. By contrast, a 1996 paper by Harvard economist Caroline Minter Hoxby suggested that unionized districts tended to have higher dropout rates.
Last week, Ms. Hoxby criticized the new Harvard Educational Review study for relying on state-level comparisons, to reach conclusions about unionization and performance. “You really cannot control easily with just five or six variables for all the differences between, say, New York and Mississippi,” she said.
Mr. Stone, who is now the dean of the college of arts and sciences at the University of Oregon, sees at least one way in which the Hoxby research can be reconciled with the new study. Since weaker students are the most likely to drop out, he notes, a higher attrition rate could yield a stronger pool of students taking college-entrance exams.
Still, Mr. Stone finds scant grounds to believe that teachers’ unions are anathema to better schools. “Collective bargaining is not the devil behind poor student performance,” he said.
