Introduction
Research proposals are plans on how you intend to conduct research. They allow you to explain what you want to study, what methods you will employ to conduct the study, importance and relevance of the research, and all other key areas that will demonstrate feasibility of your research plan. Writing an excellent research proposal requires extensive planning and organizing your ideas coherently. Therefore, the purpose of this research proposal is to explore mentorship programs and their effectiveness as a means of supporting vulnerable youth populations.
Background and significance
Youths from disadvantaged backgrounds face numerous challenges that put them at risk of poor life outcomes related to education, employment, mental health and criminal involvement (Harvard University, 2014). Factors such as poverty, family dysfunction, health issues and community violence profoundly impact youth development. There is an urgent need to provide appropriate support and intervention programs for at-risk youth. Mentorship programs aim to address this need by pairing vulnerable youth with responsible adult role models who can offer guidance, encouragement and assist with overcoming barriers. Mentorship has been shown to positively influence youth attitudes, behaviors and relationships (Dubois et al., 2011). Studies suggest effective mentoring relationships improve educational outcomes including attendance, academic performance and high school graduation rates (Cavell et al., 2009). They also boost self-esteem, emotional wellbeing and reduce engagement in risky activities associated with substance use, violence and criminal behavior (MENTOR, 2015; Rhodes et al., 1999).
Despite the promise of mentorship as a youth support strategy, not all programs yield optimal outcomes. Elements like poorly screening or training mentors, weak guidance on mentor-mentee interactions and lack of structured activities negatively impact program effectiveness (DuBois et al., 2002; Grossman & Rhodes, 2002). Researchers have called for randomized evaluations of well-designed programs using sound implementation science. In addition, moderators of mentorship effects on various youth populations merit closer examination to refine program models catered to specific target groups (Karcher, 2009). With over two million youth in the US seeking mentors annually yet many remaining on waiting lists, optimizing mentorship efforts through empirical research is critical to address pressing social issues (MENTOR, 2017).
Problem statement
This proposed study aims to evaluate the effectiveness of a community-based mentorship initiative for at-risk youth called the “Learning through Mentorship Program” (LTMP) administered by a local non-profit organization in a large metropolitan area. Specifically, the study seeks to:
Examine differences in academic performance, behavioral adjustment, social skills and self-esteem between youth in the LTMP program versus those on the waiting list through a quasi-experimental design with assessments pre and 6 months post mentor matching.
Investigate which elements of program implementation (e.g. mentor training, structured mentor-mentee activities) potentially moderate impacts on various youth outcomes.
Explore moderators of program effects pertaining to youth demographics (gender, ethnicity), family characteristics (family structure, SES) and level of risk factors.
The main hypotheses are: 1) Youth participating in the LTMP will demonstrate more positive changes compared to waiting list youth across the assessed domains. 2) Stronger program implementation quality will correlate with enhanced youth impacts. 3) Effects may vary based on youth/family attributes with more vulnerable groups possibly benefiting greater. Understanding which at-risk populations best respond can optimize mentorship approaches.
Literature review
Theoretical frameworks applicable to mentorship include social learning theory emphasizing how youth model mentor behaviors (Bandura, 1977), attachment theory regarding benefits of supportive relationships (Bowlby, 1969), and developmental context perspectives on positive adult role models navigating adolescence (Eccles & Gootman, 2002). A wealth of quantitative research shows participation in community-based mentorship programs leads to various social, emotional and academic gains over control/nonmentored youth. In a meta-analysis of 55 program evaluations between 1970-2005, significant improvements were found in areas such as social competence, academic performance and misconduct across diverse samples (DuBois et al., 2002). Other rigorous studies found equivalent or larger effects in experimental designs with random assignment (Grossman & Rhodes, 2002; Herrera et al., 2007, 2011). Subsequent meta-analyses have consistently confirmed small to moderate positive impacts on multiple domains (Raposa et al., 2019).
Research also indicates considerable variability in mentorship outcomes depending on program quality, implementation practices, and youth characteristics. Stronger effects emerge from structured programs with trained mentors, ongoing support, and focus on personal/social development rather than unstructured casual matching (DuBois et al., 2002; Karcher & Herrera, 2007). Matching on common interests and activities is important, as is mentor’s empathy, warmth and commitment (Grossman & Rhodes, 2002). Certain subgroups such as youth with family risk show greater benefits in areas like improved family relations (Karcher, 2009). Moderating influences highlight the need for tailored interventions based on factors moderating mentorship impacts.
Methodology
This quasi-experimental study with pre and posttest assessments will involve 80 LTMP participants and 60 youth on the program’s waiting list serving as the comparison group. Youth will be randomly assigned to groups by the community center. Participant demographics and level/type of risk exposure will be recorded from intake files to examine potential moderators. Participants’ academic achievement, school adjustment, behavioral conduct and self-esteem will be measured pretest one month before matching and posttest 6 months later. Academic and conduct scores will come from school records. The Self-Perception Profile for Adolescents (SPPA) and Youth Self-Report (YSR) will capture self-esteem and behavioral adjustment.
Program implementation quality will be assessed via mentor interviews, activity logs and site observations using a checklist derived from literature criteria (mentor training, supervision, activity structuring, contact consistency). Semi-structured interviews with a subset of matched mentors and mentees will explore experience quality and relationship factors. Qualitative data will enhance quantitative results. Data analysis will employ ANOVA, ANCOVA and multiple regression to compare pretest-posttest changes and examine moderators. Qualitative responses will be coded for themes. Effect sizes will provide indication of practical significance. Non-significant results will be interpreted cautiously given the quasi-experimental design limitations.
Potential limitations and delimitations
Limitations include lack of random assignment limiting internal validity, inability to verify self-report measures, possible missing outcome data if dyads discontinue participation, and not accounting for all variables impacting youth development. Delimitations are focusing only on youth 12-16 in the community center catchment area; not including youth receiving other services concurrently; and assessing a selective set of outcomes. Lack of long-term follow-up precludes determining if effects are sustained over time. Generalizability may be limited to similar structured community-based mentoring programs for at-risk youth. Social desirability response bias is possible on measures completed by mentors/mentees. Statistical power decreases for subgroup analyses.
Andrew Stroud