Research has been an integral part of YWLP since day one.
To date, research carried out by the Curry School of Education staff is based on the self-reported experiences of the middle school participants and the college women mentors through interviews, focus groups, as well as quantitative measures. We supplement this information with middle school participants’ school records and teacher reports. Overall, our research has shown that YWLP has multiple benefits for the mentees as well as the mentors, and it has also begun to reveal mentoring mechanisms that promote positive change.
Middle School Girls
In 2007, YWLP began a three-year evaluation using a randomized assignment experimental design that included 333 seventh grade girls in the program and control conditions. Three quarters or more of the program girls reported that YWLP helped them improve the way they listen to people with views different from their own, talk with their friends, support their friends, deal with their problems, talk with other kids at school, interact with people who are different from them, and think about their future (Levy, Deutsch, Henneberger, & Lawrence, 2010). In addition, two thirds or more of girls reported that YWLP helped them improve the way they think about themselves, get involved in school as a leader, make decisions about their behavior at school, and deal with sticky situations. From 2007-2010, the average match retention rate was 83% for the full academic year. Within this group, more than half of the 7th grade girls were matched for a second academic year (59%), and 22% were matched for 24 months or longer.
While most girls (program and control) reported declines in grade-point average, these declines were attenuated for program participants in 2009-2010; this cohort also had stable school bonding, which decreased for their peers over the school year (Williams, Henneberger, Lawrence & Deutsch, 2012). Interviews with mentees reveal that girls reported changes in a number of areas as a result of participating in YWLP. These included social and relational skills (e.g., respecting others, trust), self-regulation (e.g., controlling their behavior and speech, improving their attitude), and self-awareness and understanding (e.g., taking on new social roles, becoming less shy) (Deutsch et al., 2012).
By integrating academic material, service, and reflection for its mentors, YWLP embodies a service-learning model that is associated with positive outcomes for its participants. Compared to their peers, YWLP mentors reported stronger outcomes in ethnocultural empathy, competence, social acceptance, and autonomy after a year of mentoring. Mentors’ degree of perceived peer support moderated this relationship, where mentors who felt more supported had stronger outcomes compared to those who experienced less support from their YWLP peers (Marshall, Peugh, Lawrence, & Lee Williams, 2012).
In another study, we used a mixed-methods approach to investigate how YWLP supports mentor commitment, prejudice reduction, and increased understanding and acceptance of diversity among the undergraduates serving as mentors (Lee, Germain, Lawrence, & Marshall, 2010). Results suggest that this model not only supports the longevity of mentor-mentee relationships, which is a critical aspect of effective mentoring, but also improves the mentors’ ability to interact with others across boundaries of difference. Over the four years of the study the average completion rate for the college women mentors in YWLP was 91%. In terms of cultural competence, YWLP mentors made greater gains in their tolerance of people from different backgrounds compared to non-YWLP college women. Through interviews, over half of the mentors commented on the importance of structural diversity and learning in YWLP (Lee, et al., 2010).
Results from another mixed-methods study using survey and observational data reveal that YWLP mentors and their seventh grade mentees report moderate-to-high satisfaction with the group experience, with no mean differences between groups (Deutsch, Henneberger, & Lawrence, 2012). Yet there were significant differences between groups in the group members’ (mentees) satisfaction with their one-on-one mentoring relationships. Although all groups demonstrated high levels of some positive social processes related to connectedness (e.g., fun), groups in which mentees’ reported higher levels of satisfaction with their one-on-one relationships engaged in more higher level positive social processes (e.g., caretaking). Groups in which mentee’s reported lower satisfaction with their one-on-one relationships demonstrated more negative social processes (e.g., disengagement).
We have also found that certain mentor factors are associated with mentee outcomes. Using data from 142 mentor-mentee pairs from 2005-06 through 2008-09, mentor’s reported academic self-worth, parent relationship, and not being too autonomous were important pre-existing characteristics related to mentee satisfaction (Leyton-Armakan, Lawrence, Deutsch, & Henneberger, in press). While these findings have implications for mentee selection among the college student population, they only accounted for a small amount of the variance in mentee satisfaction, suggesting that ongoing mentor training and support may be a more critical predictor of mentee outcome that mentor pre-existing characteristics.