Evaluation Offers Insight into Mentoring Older, Juvenile Justice-Involved Kids

Though the construct of mentoring is somewhat nebulous, everyone understands the concept of mentoring in youth work to mean a positive relationship between an adult and a youth. By and large, mentoring programs tend to start these relationships with younger children.

But how does the dynamic of that relationship differ when you are targeting teens who often find themselves in a mentoring match because of court involvement or other risk factors present in their lives?

A recently published federal evaluation of the services rendered by Youth Advocate Programs (YAP) – which provides mentors to youth who would otherwise be locked up – offers some new insight into what works best in having an effective model for adolescents.

Researchers Michael Karcher and David Johnson assessed data from four YAP locations to measure two things: the effectiveness of the program overall, and the “ways in which advocacy and specific types of mentoring interactions contribute to youth outcomes through program participation.”

The overarching goal:

To better understand the viability of advocacy as an intervention for youth at high risk for future criminal activity, to identify critical practices that may be relevant to YAP and other programs using individualized treatment approaches to reduce delinquency and recidivism through advocacy efforts.

On the first count, the study finds YAP to be effective in lowering the self-reported criminal behavior of youth who participate. For those whose instinct is to cast a wary eye at the “self-reported” aspect, remember that gauging effectiveness by rearrest or conviction discards all of the crimes for which youth are not caught.

Karcher, speaking with Youth Services Insider, was direct in his praise for YAP: When it comes to mentoring older youth, a “YAP program with staff who know how to support mentoring matches is as good as it gets.”

The evaluation’s further findings will likely be valuable to YAP in reaffirming some things and suggesting others about its model. But the findings should also be of interest to any program wanting to reach older youth.

You can click here to read the entire study. There were several critical factors of the mentoring relationship that Karcher and Johnson identified as being correlated with self-reported behavior.


Advanced levels of education among mentors was associated with lower levels of self-reported behavior, the researchers found. They also found that those mentors with teaching experience had an even bigger positive impact, but Karcher said the fact that only 7 percent of the mentors had such experience means this could be “a fluke.”

This somewhat challenges the bedrock YAP principle that the best mentors for at-risk youth are adults from their own neighborhood, or at least who share some similar personal experiences. Surely every community has its share of well-educated adults, but YAP’s focus has always been recruiting those who will have authentic currency to reach kids.

“YAP views similarity between youth and their Advocates as a primary way in which YAP Advocates are able to reach youth and help them,” the evaluation said. “Assuming most of the youth in YAP are from families with lower than average educational attainment, which may be incorrect, Advocates with college degrees may be expected to be less effective by virtue of their educational dissimilarity from their mentees and their mentees’ family members.”

Instead, the evaluation said, “We found a direct positive effect of Advocates’ levels of educational attainment on youth outcomes. … Recruiting more educated Advocates and those with teaching experience may provide one way to boost program impact, even though it conflicts with YAP’s logical model which places a premium on hiring Advocates who are most like those they serve.”

YSI‘s two cents: the finding doesn’t necessarily suggest recruiting should be based on someone’s educational background. It does mean that further research should seek to distill what it is about the educational difference that is relevant in this space. Surely it is not that one mentor aced her GRE while another went straight to the workforce after high school.

Karcher said he and Johnson also intend to go back and attempt to study the impact of cultural similarity between mentor and mentee, which as mentioned is a principle of the YAP model. He said they just ran out of time on this project, and that he’d need to better hone the definitions and boundaries of “sameness” (neighborhood, race, either, both, etc).


Think of “play” as sports and recreation activities shared by mentor and mentee, as opposed to deep conversation, academic work or problem-solving actions.

Karcher and Johnson found that mentors who, early in a match, spent an abnormally high amount of time engaged in “play” with a youth were associated with higher levels of self-reported negative behaviors.

They found the diametric opposite in examining the level of play later in a mentor relationship. Greater time spent playing in this time frame was associated with lower levels of self-reported negative behavior.

From the evaluation:

Therefore, attending to when a problem-focused approach is taken in a match may be very helpful in maximizing the benefits of mentoring this type of youth. These findings do not support the use of what Morrow and Styles’ called the developmental style. It does not appear best, when mentoring court-referred youth, to emphasize cultivating a friendship first and allow problems to arise naturally over time. The findings regarding the role of play early in the match concur with this view.

This, to YSI, is the most interesting part of the evaluation. A mentor’s instinct might be to make inroads with a youth by having fun with them, trying to build a positive image in their mind, and then getting into the more serious issues.

And with younger children, that might very well be the case. But this evaluation suggests that older teens might not be impressed.

On the flip side: mentors who get down to business, and gradually grow into a friendship with the youth, seem to yield better results than those who keep it professional throughout.

This speaks to the preconceived notion held by youth involved with the court system and, frankly, many other teens: that adults, writ large, suck.

“There’s not a lot of respect for the conventional world of adults,” said Karcher. “They’re prone to think, this guy’s a sucker. You’re just a sucker because you want to help people.”

The findings suggest that mentors who strive for immediate connections with teens might play into this notion.

“I was expecting another pattern, that kids would be wary – ‘Do I trust this guy?” Karcher said. “But as we looked at the data, we thought, you can’t deny this kid’s in trouble. He just skirted going to juvie. To come in and pretend they’re not really on the edge of crime would be disingenuous.”

It’s worth noting that the findings about play strangely intersect with the findings about mentor education levels, because higher levels of education were associated with lower engagement in play.

So the sweet spot for mentor performance with older youth seems to be here: more education, early engagement in problem-solving, steady growth into friendship.

Print Friendly, PDF & Email

John Kelly, Editor in Chief, The Chronicle of Social Change
About John Kelly, Editor in Chief, The Chronicle of Social Change 1212 Articles
John Kelly is editor-in-chief of The Chronicle of Social Change. Reach him at jkelly@chronicleofsocialchange.org.