The researchers compared what happened to the 3,000 students who had mentors at school with the roughly 5,000 students who said they had no mentors at all. The ones with school mentors did moderately better in high school with slightly higher grades – for example, a B- versus a C+ – and failed fewer classes.
But what was really striking was what happened after high school. Those who had formed a positive relationship with a teacher, a counselor or a coach increased their chances of going to college by at least 9 percentage points. That’s a substantial boost given that only 51% of students without a mentor enrolled in college.
Kraft and his colleagues brought the tools of modern applied economics to answer the question of a teacher’s worth outside of the classroom. There are many confounding factors and perhaps the teens who form these relationships with caring adults are different in other ways – maybe they are more ambitious or have more self-confidence – and they would have gone to college in higher numbers even if they hadn’t had a mentor at school. Though it’s impossible to account for all the possibilities, the researchers crunched the numbers in various ways, arriving at different numerical results each time, but consistently saw strong benefits for students who had mentors at school. This was true even between best friends, romantic partners and twins. For example, the twin sibling with a mentor did better than the one without, even though they were raised by the same parents and attended the same high school.
Kraft and his colleagues didn’t detect a big difference in college graduation rates between those with and without mentors. The largest difference seems to be the decision to apply and enroll in college. For students who are undecided on whether to go to college, having a school-based mentor seems to carry them over the threshold of the college gates.
Students from low-income and less educated families were less likely to have a mentor, but having a mentor was even more beneficial for them than it was for their higher income peers. Their college going appeared to be dramatically higher. The mentoring itself also seemed different for poor and rich students. Lower income students were more likely to report that their mentors gave them practical and tangible help, along with advice on money. Higher income students were more likely to report receiving guidance, advice and wisdom.
Being mentored by a sports coach was just as effective as being mentored by a teacher; these young adults experienced the same short-term and long-term benefits. However, female students were more likely to gravitate toward teachers while male students were more likely to bond with a coach.
Formal mentorship programs, such as Big Brothers Big Sisters, have also produced benefits for young adults, but Kraft said the benefits from the informal relationships studied here appear to be larger.
“We know how to set up formal mentoring programs but not all the relationships are going to pan out,” said Kraft. “We know far less about how to support and cultivate the formation of these voluntary relationships. And we have no control over whether or not it’s the students who might most benefit from them who are able to successfully seek out and form these mentoring relationships.”
But there are some clues in the study as to what schools can do to create the conditions for serendipity. “There is no magic wand for exactly the best way to do it,” Kraft said. “It’s not something we can say, do this and relationships will form. But schools are social organizations and can create environments where they’re more likely to happen.”
The researchers noticed that high schools with smaller class sizes and those where students said they felt a greater “sense of belonging” tended to produce twice as many of these mentoring relationships than schools with larger classes and a less hospitable school environment. “When students say that school is a place where they feel welcome and part of the community,” said Kraft. “you’re much more willing to open up to a teacher or counselor or a coach, and reciprocate when they reach out and say, ‘Hey, I see you’re looking a little down. Do you want to talk about it?’”
Kraft offers two additional suggestions for schools:
- Hire more Black and Hispanic teachers
White students were substantially more likely to report having a school mentor than their Black and Hispanic peers. That’s likely because the U.S. high school teacher workforce is 79% white and 59% female, and from middle and upper-middle class backgrounds. “Shared common life experiences increase the likelihood that you’ll develop an informal mentoring relationship because you can talk about things in a common way,” said Kraft. “This adds weight to the pressing need to diversify the teacher workforce.”
The researchers do not know why so many Asian males (more than 20 percent) sought out and built strong relationships with adults at school. Seventeen percent of Asian females had school mentors. Only 10% of Black and Hispanic female students had mentors at school while Black and Hispanic males reported slightly higher rates of about 12 percent. Fifteen percent of white students reported having school-based mentors.
- Create small group moments
Kraft suggests that school leaders can promote these student-teacher relationships by creating more opportunities for students to have multiple, sustained interactions with school personnel in small group settings. This doesn’t necessarily require smaller class sizes; small groups could be advisory periods, club activities or tutoring sessions during the school day.
Is the implication of this study that teachers should be taking on even more responsibilities? Kraft says that’s not his intention. Instead, he wants to recognize what many teachers and other school staffers are already doing. It’s another way, he said, “in which teachers are incredibly important.”