June 13, 2008

Making the Grade with Mentoring

Jonah Rockoff examines data from New York City’s public schools to assess whether mentor programs — increasingly used in the private as well as the public sector — reduce employee turnover and improve productivity.
Print this story

Hiring new teachers is an expensive proposition. The hiring season is also highly concentrated: schools fill their positions in the summer, and it can be difficult to find a replacement in the off-season if a teacher quits during the school year.

Over the last 15 years, public schools have increasingly embraced the practice of mentoring — pairing experienced educators with new teachers — as one means to increase teacher retention and improve student achievement, a key measure of productivity in schools. Mentoring has also been steadily on the rise in the private sector, for similar reasons.

As with professionals in other fields, more experienced teachers tend to be the most effective and productive teachers. “Schools are no different than any firm: loss of human capital is detrimental. In industries where human capital is really important, retention becomes a major issue,” says Professor Jonah Rockoff, whose research has frequently examined teacher training in public-school systems. There’s no solid evidence that turnover among teachers is higher than that among professionals in other fields. “But,” he points out, “turnover is typically higher in schools that serve disadvantaged populations. So these schools constantly need to spend resources on hiring, and end up filling their classrooms with inexperienced teachers.”

“The stress level in the beginning of a teacher’s career is often highly influential in whether a teacher stays or goes,” Rockoff says. “Mentoring is a way of easing that transition.” It’s also a way for seasoned professionals to pass on knowledge and advice. But mentoring is expensive, and no conclusive data exist about how well it works.

“Two big questions are what makes the mentoring experience beneficial and how should we tailor mentor selection and assignment to ensure the best outcome,” Rockoff says. Any number of factors could influence the mentor-mentee relationship and the program’s outcome, including whether the mentor and mentee are similar in age or gender, the mentor’s years of professional experience and how much time the mentee and mentor spent together.

The burgeoning of mentoring programs presented Rockoff with the opportunity to examine which of those factors were most important and to consider one of the classic questions of labor economics: Does investment in human capital (in this case, training) increase productivity?

The New York City public-school system, the largest in the nation, adopted a state-mandated mentoring program in 2004. Rockoff used the substantial amount of data available from the New York City system to examine the effect of mentoring on teacher retention and student achievement. He supplemented the data by speaking with teachers, mentors and school administrators.

Rockoff’s most consistent finding was that the factor most related to whether a teacher decides to continue teaching in New York City public schools and in the same school is whether the mentor assigned to him or her had previous experience working in that school, either as a teacher or a mentor.

“One big reason teachers will stay or go is whether or not they get information and training and help that allows them to operate well within the particular school environment that they’re situated in,” Rockoff explains. “If I’m a new teacher in P.S. 101 and have a mentor who’s been teaching there for 15 years, that could be really helpful because that mentor can help me figure out how to deal with Vice Principal Jones, who’s really tough, or where to get the cheapest school supplies or even what bus to take to get there on time. A mentor who’s been teaching 20 years in another part of town may not understand the students or administrators at that school.”

Happily, Rockoff says, this finding coincides with some new policy decisions that the New York City public schools had already put into effect that allowed principals to exercise more decision making at the school level.

Rockoff also found that a mentor’s having had prior experience in the teacher’s school was more important than such factors as whether the mentor and teacher shared similar educational backgrounds, or were of the same race or gender — differences that didn’t seem to have any impact, good or bad. Nor did the mentor’s caseload appear to matter much, despite the fact that some mentors were assigned more than 20 teachers, while others were assigned less than 15.

Regardless of their caseload, some mentors put in many more hours than others. Rockoff found that the more time a mentor put in overall the better that mentor’s teachers performed in terms of student achievement: students’ test scores in reading and math were significantly higher for those teachers whose mentors worked more hours.

“The effects are not straightforward to uncover. If you look at the simple relationship between how many hours of mentoring teachers received and student achievement, you see negative effects,” Rockoff says. He suggests that it’s actually the anticipation of a poor outcome — lower student achievement — that causes more hours of mentoring. “Mentors know that among their group of teachers, maybe there are two or three teachers who really, really need a lot of help, and that’s where mentors spend a lot of extra time.”

Teacher-mentoring programs usually stipulate that a teacher be matched with a mentor who shares the same subject-area expertise, a requirement that wasn’t always strictly enforced in the New York City teacher-mentoring program. “Quite a lot of effort goes into tailoring assignments based on license area,” Rockoff says. “Yet teacher-retention rates, student outcomes and even the survey results [teachers' evaluation of their mentor] show that whether a teacher is assigned a mentor who is matched to their subject makes absolutely no difference.”

It turns out that teachers often thought their mentor matched their subject area even when their mentor’s license was in a different area, particularly if they rated their mentor highly overall. Rockoff suggests that the best mentors “are able to jump in and give advice that’s relevant, leading teachers to believe that the mentor is licensed in the same subject area. So pushing certain program requirements may actually lead to inflexibility and rigidity that is unproductive.”

When New York City’s teacher-mentoring program was developed, the program’s designers spent a lot of time debating programmatic aspects — caseload, subject-area expertise, years of teaching experience. “But those things didn’t seem to make that much of a difference. The model of formal mentoring advocated by many groups who design these programs is based on the idea of a central, full-time mentor who works in a number of schools. That may hold some advantages for a district’s administration,” Rockoff says. “But if it’s the school-specific knowledge that’s really important, that says that a district is better off trying to have principals run their own systems within their schools.”

Rockoff’s findings suggest that mentoring programs — whether based in the private or the public sector — should seek to assign mentors based on their experience in the local work environment and their ability to provide relevant advice and guidance, regardless of their field of specialty. “The information that needs to get to new employees,” Rockoff says, “is information about how to operate in their specific environment.”

Jonah Rockoff is associate professor of finance and economics at Columbia Business School.

Read the Research

Jonah Rockoff

"Does Mentoring Reduce Turnover and Improve Skills of New Employees? Evidence from Teachers in New York City"

Download PDF 
View abstract/citation