Researchers Examine Contracts’ Effects on Policy Issues

Does the existence of collective bargaining help or hinder attempts to establish performance-pay programs? Do salary schedules improve teachers’ compensation or curb districts’ ability to hire top talent? And can contracts serve as an avenue to define the features of high-quality mentoring programs?

Although such questions are likely to remain open for some time, a new generation of education researchers has picked up magnifying glasses and begun analyzing teacher contracts for answers in greater depth than ever before.

At a conference held here March 26 by the National Council on Teacher Quality, a Washington-based group that seeks to improve state and local policies affecting teachers, five scholars presented what are among the first studies to use primarily quantitative research methodologies to connect collectively bargained teacher contracts to various education policy outcomes.

That research area, the scholars said, has been all but ignored, despite growing academic interest in teacher-quality issues over the past decade.

“The only two published books I found about collective bargaining and education said in their introductions that this is an area of neglect,” said Kristine Lamm West, who co-wrote one of the two papers that won awards from the council.

Members of an academic jury that reviewed the submissions acknowledged that none of the finalists’ papers yields conclusive results. But the papers do lay the groundwork for increasingly complex analyses, they added.

New Questions

The researchers who presented their findings last week competed for a cash prize sponsored by the council. Each researcher presented his or her paper and then defended it to a jury made up of top education researchers—a format that Kate Walsh, the president of the council, jokingly likened to that of the popular “American Idol” television program.

The researchers drew on the NCTQ’s extensive database of collective bargaining contracts for their studies. Known as Teacher Rules, Roles, and Rights, or TR3, the database details such features as salary schedules, benefits, and other pay incentives from contracts in place in the nation’s 100 largest districts. (“Online Teacher-Contract Database Launched,” Jan. 10, 2007.)

Underscoring the depth of the analytical potential of the database, each of the prize-winning research teams used a novel lens to examine the thorny topic of teacher compensation, and their papers produced lively debate among members of the jury.

Ms. West and her co-author, Elton Mykerezi, whose paper took the top prize of $15,000, examined the question of how teachers’ unions affect support for various aspects of teacher pay.

Supplementing the TR3 data with information from the federal Schools and Staffing Survey, a database maintained by the research wing of the U.S. Department of Education, the two University of Minnesota-Twin Cities researchers found that union influence exerted through a bargaining contract strongly correlates with support for traditional pay structures, such as higher starting salaries, increased compensation for teachers who earn advanced degrees, and increased pay for experience.

The paper also found evidence that the presence of a bargaining contract reduces the probability that the district employs an incentive-pay plan that defines teacher performance entirely through student test scores, although that finding wasn’t as strong.

Conventional Wisdom

Overall, Ms. West and Mr. Mykerezi’s findings align with the conventional wisdom that teachers’ unions typically support input-based approaches to teacher pay, but are indifferent or opposed to performance-based pay based on student-achievement outcomes.

“I ran this past some union friends, and they said, ‘Well, good, our union dues are going to something,’?” Ms. West noted.

But, she added, she was surprised by a finding that bargaining contracts appear to improve compensation for teachers regardless of their length of service. That finding conflicts with the notion that unions are under pressure to benefit their more senior members at the expense of novices.

“The unions do not appear to be trading off starting salaries or benefits in exchange for these increases,” the authors wrote. “Rather, unions are able to use their collective bargaining power to increase all aspects of teacher pay and to compress the salary schedule so that it takes fewer years for teachers to reach their maximum salary.”

The second-place finishers, D. Chris Ferguson and Greg Gilpin, both from Indiana University Bloomington, tried to determine how the “rigidity” of salary schedules based on experience and seniority influences schools’ ability to attract teaching talent. Such schedules are set statewide in certain states like North Carolina, and through local collective bargaining contracts in other states.

Like Ms. West’s team, the authors used both TR3 and SASS data for their study. They found evidence that salary schedules appear to depress schools’ ability to attract the best teachers, with the top 75th percentile of teachers in schools with salary schedules having scores on the SAT college-entrance examination that were 2 percent to 3 percent lower than peers in schools without salary schedules.

“The salary schedule is most binding for high-quality teachers,” Mr. Ferguson said.

The findings were most apparent for the best math and science teachers, whose SAT scores were 3.5 percent lower, on average, than their peers’.

“It is the highest-ability individuals who would face outside wage options that are most significantly different than those wage opportunities in the education sector,” the authors wrote.

Such differences in teacher quality could ultimately affect student outcomes such as high school graduation and college-attendance rates, Mr. Ferguson said.

Study Limitations?

Although members of the jury largely praised the papers for their innovative approaches, they also found several areas of concern.

Eugenia Kemble, the executive director of the Albert Shanker Institute, a Washington-based nonprofit research group affiliated with the American Federation of Teachers, criticized the Indiana University researchers’ use of SAT scores as a measure of teacher quality. Other studies have questioned that link, she said.

“I don’t know why you picked that [variable],” she said, “but I find it to be a fundamental flaw.”

Susan Moore Johnson, a professor at the Harvard Graduate School of Education, expressed reservations with the researchers’ definitions. She pointed out that Ms. West’s paper conflates the presence of bargaining with union influence, a distinction she said should be clarified.

And the existence of collective bargaining, Ms. Johnson cautioned Mr. Ferguson, is a different phenomenon from the existence of a salary schedule, which many private schools also use.

Several other participants raised questions about the scope of the TR3 database and its appropriateness for serving as a data source. While populous, the 100 districts covered by the database are just a sliver of the nearly 15,000 school districts nationwide, noted David Schlein, a senior policy analyst of labor economics in the National Education Association’s collective bargaining and membership-advocacy wing.

But Jane Hannaway, the director of education policy studies at the Urban Institute, a Washington-based think tank, praised the database for opening new avenues for study. As new contracts are inked, TR3 will presumably incorporate information on how districts’ contracts change over time, which will offer researchers more statistical power.

“This is ... a cross section of contracts that’s likely to yield some interesting results we won’t get elsewhere,” she said. “The database will become longitudinal, hopefully, and I think the analyses will get richer and richer.”