New Mexico's tiered teacher-licensing system, which grants large boosts in pay as teachers advance, doesn't appear to have much of a relationship to teachers' ability to improve their students' test scores, according to a report from the state's legislative finance committee. (h/t to Hailey Heinz of the Albuquerque Journal, who broke the news.)
New Mexico's tiered-licensing system was instituted in 2003. Beginning teachers, in Tier I, make a minimum of $30,000. To move up to Tiers II and III, they have to earn satisfactory evaluations at the local level for several years, and submit a "professional development dossier" for state review. They earn $10,000 more annually after advancing to a new tier.
The system has been cited as a reason for fewer unqualified teachers and fewer teacher shortages, as Education Week reported a few years ago.
For the study, the researchers analyzed the relationship between licensure tier and student achievement for some 4,600 teachers of students in grades 3-8. They found that teachers across licensure levels had students with similar patterns of growth, suggesting that the PPD process doesn't adequately screen for teacher effectiveness.
About 90 percent of teachers who submit a portfolio to advance to a higher pay tier are successful, the report says.
Overall, the study states, the three-tiered system "acts as an expensive proxy for paying teachers based on their education and experience."
The study acknowledges that the system began before year-to-year achievement data linked to teachers were available, but has not since been updated to include "robust" measures of student achievement. The portfolios, it notes, are not comparable because each teacher selects the work samples and assessments they use.
It recommends that the state education department and the legislature do more to align the tiered system with evidence that teachers who advance are actually helping their students learn more.
However, the report sounds a caution on the use of "value added" modeling, which looks at student test-score data over time, for evaluating teachers, noting that they can identify top- and bottom-performing teachers over time, but are not good at distinguishing those in the middle.
It gives several examples showing that individual teachers' value-added scores could change depending on which specifications were chosen for the VAM. A gifted education teacher, for example, had top scores on a model that doesn't include student demographic factors, but would be deemed ineffective if those characteristics were taken into account.
Rather than mandating VAM for individual teacher evaluations, the report recommends that the state provide the data privately to schools and teachers. That way, individual districts could choose to tap the very best teachers as mentors. It also recommends permitting such teachers to bypass the portfolio process, and to move up a tier.
Interestingly, the idea of tiered licensure has popped up elsewhere in the nation recently.
A report issued this week by a specially convened group of California teachers proposed a system in the Golden State not unlike New Mexico's. Teachers would submit portfolios of student work to advance to a higher licensing tier, though they'd also take on new roles, such as mentoring and sharing best practices with their colleagues, something not in New Mexico's system.
The California teachers, though, said they could find "no productive use for VAMs in teacher evaluations or professional development."
Their report, a project of the National Board Resource Center at Stanford University, won praise from Stanford professor Linda Darling-Hammond, who sits on the board of the state's teacher-credentialing board.