Minnesota’s teacher pay option has only minor impact on test scores
Steve Brandt, Star Tribune, November 18, 2013 – Minnesota’s $450 million alternative approach to compensating teachers has had little effect on student achievement.
That conclusion about Q-comp was reached by academic researchers from the University of Minnesota and the St. Catherine University in St. Paul. They found a detectable impact in reading scores in participating Q-comp districts, but other researchers who reviewed the findings call that impact small. The effect on math scores was about the same level, but the researchers couldn’t confidently attribute it to Q-comp.
However, the increase in reading scores was more than twice as much for districts that had been in the program for five years, and was also stronger for teachers in their first five years.
Q-comp is the signature alternative teacher pay program championed by then-Gov. Tim Pawlenty as an alternative to compensating teachers for their number of years on the job and college progress. It’s one of the earliest such programs in the nation and was approved by legislators in 2005, although pilot efforts date to 2001. It is paid for with state aid and property taxes, and so far has received a public commitment that exceeds $450 million. About one-fifth of the state’s school districts and nearly half of its charter schools participate.
The idea behind Q-comp is that teachers should be compensated for taking on school leadership roles, for participating in professional improvement plans, or for meeting individual, team or school improvement goals.
The goal-achieving aspect, more widely referred to as pay-for-performance, has been controversial among teachers to the extent it relies on test scores for individual teachers. But it makes up a minority of Q-comp spending by districts, the researchers found. A 25-student classroom generates in Q-comp about $6,500 per teacher annually, but on average, only about $2,200 of that is paid to teachers and only $243 of that is for performance improvement; the remainder can go to district initiatives such as evaluating teachers or paying substitutes for times when teachers are out of classrooms for planning or training.
Intentions and results
The research was conducted by economics Profs. Aaron Sojourner and Elton Mykerezi of the U and Kristine West of St. Catherine. The researchers compared individual student achievement growth, as measured by two generally administered student achievement tests, before districts adopted Q-comp and then after it was adopted. They then compared that rate of growth to that of students in districts that never adopted the program.
Pawlenty touted Q-comp in his presidential campaign. He said through a spokesman this week that the program “improved student learning and did so in a very cost-effective manner. The results would have been even better had the DFL and teachers unions not watered down the program.” He cited the deletion from his original proposal by the DFL-controlled Senate of a requirement that the new system replace traditional pay scales, and of more prescriptive limits on qualifying activities.
The study authors calculate a 5-to-1 ratio of social benefit compared with cost for Q-comp. “It’s small, but it’s real,” West said of Q-comp’s impact, “and it’s inexpensive compared to the other policy options that are out there.”
But state Education Commissioner Brenda Cassellius argues that other investments, such as early childhood spending, appear to have a bigger payoff.
Former legislator Barb Sykora, who carried the Q-comp legislation, called the research’s findings on program impact “somewhat disappointing.” But she attributed that to the law’s requirement that local districts agree on a Q-comp plan with their teacher unions. “As a general rule, teachers have not wanted to differentiate pay for performance,” she said.
Hard to isolate factors
The researchers describe Q-comp as a mix of pay-for-performance incentives, on which research has found mixed results nationally, and other workplace changes such as mentoring, collaborative analysis of student data and adoption of new teaching techniques. They said they weren’t able to separate the impact of financial incentives from changes to work practices such as increased collaboration among teachers.
Or, as University of Missouri economist Michael Podgursky summarized the issue: “You don’t know which ingredients. All you know is the cake mix produced a cake.”
But researcher West, a former high school teacher in Minneapolis, suggested that teachers don’t keep a secret technique in their pocket to pull out when incentives show up. To her, that means that to succeed, incentives need to be accompanied by job changes such as collaborative planning by teachers or professional training aimed at analyzing teaching practices.
The research also found no evidence that the incentive plans are causing teachers to shift to districts where they can earn more or causing other workers to enter teaching.
The research appears shortly after the Minneapolis district entered Q-comp but St. Paul teachers refused to do so. Mary Cathryn Ricker, president of the St. Paul union representing teachers, said the research “gives our Legislature a chance to rethink how we allocate the resources, given that this isn’t moving the needle.”
Denise Specht, president of Education Minnesota, which represents unionized teachers, saw another message from teachers in the research. “They’re not motivated by performance pay like other professionals are. They know it’s a team effort,” she said.