Keeping Teachers In School

WASHINGTON, JUNE 26, 2008--Does mentoring help improve new teachers? Researchers have long found that experience matters when it comes to teacher effectiveness. High teacher turnover means more teachers with less experience, resulting in a negative effect on student achievement. As a response, many school districts and more than thirty states require mentoring programs that pair veteran teachers with rookies to help ease the transition into the classroom. But while over 1 million teachers were mentored from 1993 to 2003, evidence on the overall impact of mentoring is inconclusive.

At a recent AEI event, Jonah Rockoff of Columbia Business School presented a paper examining a mentoring program in New York City, which adopted a program designed in partnership with the University of California, Santa Cruz, in 2004. Rockoff evaluated not simply whether the program worked, but instead measured "the impact of mentoring quality" by showing whether specific mentoring characteristics had an effect on teacher and student outcomes. Using teacher retention rates, absence rates, and student level achievement data, Rockoff's findings indicated that

• Student achievement improved for teachers who spent more time in mentoring
• The most consistent predictor of teacher retention is whether mentors had previous experience working in that particular school

"What I find," Rockoff said, "is that for teachers who get more hours of mentoring, their students do better on standardized tests." Studies in the past looked at the "raw relationship between hours and mentoring received" but failed to understand that teachers being given the most mentoring time are likely the ones that need the most help. He said that this finding has led to "overly negative conclusions" about mentoring programs, and he emphasized the need for new research methods that use mentoring characteristics to "get around that spurious relationship."

Rockoff also found that school-specific knowledge was the most important factor affecting teacher retention. Mentoring experience in that particular school helps provide "local knowledge about what goes on in your school [and can help a teacher] navigate local institutions . . . deal with local administrators," and understand "kids in that school with their particular characteristics and proclivities for how they learn."

But putting these findings into practice can be difficult. As Michael Petrilli of the Thomas B. Fordham Institute noted, "these programs cost money." And when schools are in hard financial times, a superintendent might see mentoring programs "as a big fat target" in a budget. In reality, while these programs may be the most effective support teachers have, there can be political problems with siphoning money from professional development programs to mentoring.

There are other hurdles for proponents of mentoring programs. Aminda Gentile of the United Federation of Teachers argued that remedying new teacher induction plans are easier said than done. Howard Nelson of the American Federation of Teachers acknowledged the rigor of Rockoff's paper but pointed out the difficulty of maximizing mentors with school-specific knowledge because of the tendency for mentors to become "evaluators" instead of supporters. "The last thing teachers need is another assistant principal."

With the increase in alternative teacher training methods, which can put under-prepared teachers into difficult teaching environments, mentoring programs for first-year teachers will continue to be a hot button issue. For now, "mentors are allocating their time where they see the greatest need," Rockoff said, which is a step in the right direction for decreasing teacher turnover.

--JON FLUGSTAD

For video, audio, and more information about this report, visit www.aei.org/event1749/.

For more information about AEI's Future of American Education Project, visit www.aei.org/futureofeducation/.

For media inquiries, contact Véronique Rodman at 202.862.4870 or vrodman@aei.org.

###