Using Existing Program Data to Assess the Health of Mentor/Mentee Relationships

Figure 1 Photograph by Scott Wagers. Public Domain
Figure 1 Photograph by Scott Wagers. Public Domain

Submitted by: Michael Lyons


Data were collected from a large national sample of Big Brothers/Big Sisters mentors and mentees.
Assessment data can provide a signal for those who may be at risk for terminating the program relationship early.
Additional support can be offered in struggling dyads.

Medical doctors routinely order patient tests to identify potential health problems early. Elevated levels of cholesterol, for example, can signal if a patient is at greater risk for heart disease. Armed with this information, the patient and doctor may order lifestyle or pharmacological treatments that reduce the patients’ risk of future heart disease.

Mentoring programs may also benefit from similar “tests” that signal a mentoring relationship is “off-track.” If leaders want to prevent potential risks associated with the mentor-mentee relationships ending early, they may be able to offer additional support.

“Existing data collection efforts within Big Brothers Big Sisters of America (BBBS) provide a promising foundation upon which to begin to identify those at risk for prematurely terminating the service.”

How Did A Community Psychology Perspective Inform Your Work?

Community Psychology has deep roots in applied and translational research that directly impacts communities and individuals. As mentoring researchers and programs continue to seek new ways to improve services, this work informs how mentoring programs may use data to individuate services to the needs of the mentor-mentees in their programs.


This study examined how 1) mentor and mentee’s reports of their relationship quality and 2) shared racial or ethnic identities predicted the premature end of a youth-adult mentoring relationship. Data were collected from 82,224 mentors and mentees who participated in a youth mentoring program operated by BBBS between 2014 to 2018. Because BBBS expects that most mentoring relationships last at least 12-months, mentor and mentee relationships that did not last this long were coded as “prematurely terminated.”

We ran three types of tests. First, we assessed whether the measure of “relationship quality” used by BBBS accurately measured mentoring relationship quality for mentors and mentees from different racial or ethnic and gender identities. Second, we tested if racial or ethnic match and relationship quality predicted premature termination among the sample of BBBS matches. Third, we tested if a mentoring program could accurately predict if an individual mentor-mentee match would end prematurely based on their individual match characteristics (i.e., their demographic and relationship quality scores).


  • We found that the BBBS measure of relationship quality accurately measured relationship quality across mentors and mentees from different racial or ethnic and gender identities.
  • On average, mentors and mentees reporting a lower quality mentoring relationships tended to end early.
  • A shared racial or ethnicity background between mentors and mentees was associated with a lower likelihood of premature termination (as compared to mentees identifying as Black matched with mentors identifying as non-Black).
  • We did not find sufficient evidence that information about relationship quality and mentor-mentee identities could accurately predict the likelihood that individual matches would terminate prematurely.

What Does This Mean For?

Research and Evaluation: Researchers may consider how existing data may be used to directly inform programmatic decisions to support youth development.

Practice: Results suggest that mentoring programs may consider other information beyond quantitative indicators of relationship quality to inform decisions about the likelihood of a mentor-mentee match terminating prematurely.

Original Citation: Lyons, M. D., & Edwards, K., (2022). Strategies for monitoring mentoring relationship quality to predict early program dropout. American Journal of Community Psychology. 70, 127-138.

Download a .pdf of this page here


Contact Us