Content uploaded by Simon Larose
Author content
All content in this area was uploaded by Simon Larose on Jan 27, 2015
Content may be subject to copyright.
Academically at-risk students
1
ACADEMICALLY AT-RISK STUDENTS
Simon Larose and George M. Tarabulsy
Laval University, Québec, Canada
In David L. DuBois and Michael J. Karcher (Eds). Handbook of Youth Mentoring (2nd edition).
New York : Sage Publications.
Simon Larose, département d'études sur l'enseignement et l'apprentissage, Faculté des sciences
de l'éducation, Université Laval. George M. Tarabulsy, École de Psychologie, Université Laval.
This chapter was supported by grants from le Fonds Québecois de la Recherche sur la Société et la
Culture and from the Social Humanities Research Council to the first author. Correspondence
concerning this chapter should be addressed to Simon Larose, département d'études sur
l'enseignement et l'apprentissage, Faculté des sciences de l'éducation, Université Laval, Qc Canada
G1K 7P4. E-mail: simon.larose@fse.ulaval.ca
Academically at-risk students
2
INTRODUCTION
Many educators consider formal and natural mentoring to be unique experiences that may
prevent school dropout. It is widely held that pairing academically at-risk students (AARS) with a
well-intentioned adult, “competent” peer or experienced teacher typically may improve AARS’
adjustment. The proliferation of mentoring programs implemented in schools and communities
around the world without prior careful evaluation illustrates the predominance of this view
(Randolph & Johnson, 2008). Yet recent studies and meta-analyses suggest that the impact of
mentoring on the development of AARS is modest at best and can even be detrimental under certain
conditions (Blinn-Pike, 2007; Eby, Allen, Evans, Ng, & DuBois, 2008; Wood & Mayo-Wilson,
2011). A comprehensive evaluation of the benefits of mentoring for AARS first requires a clear
understanding of the prevalence, determinants and impact of the risk for academic failure. Without
such insight, conclusions regarding the effectiveness of mentoring among this population will
remain unclear. This chapter constitutes an important update of the chapter first published in 2005.
It includes several novel elements, notably an analysis of recent theoretical perspectives on
mentoring AARS, a review of empirical studies published since 2000 and a proposition of key
reflective actions for mentoring practitioners.
Who are Academically At-Risk Students?
Academically at-risk students are a heterogeneous population. There are many ways for students
to be considered “at-risk” for lower academic achievement and school dropout. AARS may present
negative internal forces acquired over the course of their development and/or may have experienced
negative external influences that undermine their school adjustment. For instance, negative internal
forces may include behavior and/or emotional problems in early childhood (e.g., aggressiveness and
Academically at-risk students
3
hyperactivity), poor study skills in early adolescence and career indecision in late adolescence.
Indicators of negative external influences may include family or community-based poverty,
unstable family circumstances (e.g., divorce), peer rejection, and academic challenging situations
(e.g., school transitions). Such negative forces and influences are recognized as having the potential
to alter the quantity and quality of academic support available to students by challenging parental
school involvement and academic relationships with peers and teachers (Tyler & Lofstrom, 2009).
Within this context, it is not surprising that AARS are more likely to drop out of school.
High school dropout continues to be an important social and economic problem in western
nations. In 2005, 10% of Canadians between the ages of 20 and 24 had not graduated from high
school or were not enrolled in high school (Bowlby, 2005). In 2007, this same statistic stood at 16%
in the United Sates among those 16 to 24, for a total of 3.3 million young adults (Cataldi, Laird,
KewalRamani, & Chapman, 2009). In countries of the European Union, 15% of young adults aged
18 to 24 had failed to obtain a high school diploma in 2007, although rates vary significantly
between countries (e.g. 5% in Poland and 36% in Portugal; Pour la Solidarité, 2009). In most of
these countries, dropout rates are higher among boys, ethnic minorities -especially African-
Americans and Hispanics in the United States- and youth from lower socioeconomic status. School
dropout has been associated with a number of individual and social consequences, such as receiving
social aid, having physical or mental health problems, involvement in illegal activities, and
becoming the parents of children more likely to drop out (Tyler & Lofstrom, 2009).
The next section focuses on some of the different theoretical perspectives on which recent
AARS mentoring research has been based. These perspectives help document research that has
addressed the effectiveness of mentoring, and elaborate on potential underlying mechanisms at the
heart of the mentoring process.
Academically at-risk students
4
THEORY
Approaches to mentoring research that involve AARS may be grouped into three categories. The
first category comprises atheoretical research. It justifies the relevance of mentoring through
empirical evidence (e.g., the fact that mentoring relationships should be sustained over a period of
at least 6 months), but does not rely on any theories or perspectives to describe mentoring
processes. The second category includes research based on established theory or general models
used in fields outside mentoring (e.g., attachment theory) without addressing issues specifically
related to the mentoring experience (e.g., Larose, Bernier, & Soucy, 2005). Finally, the third
category includes recent models inspired by mentoring research that sheds light on the mechanisms
involved in the experience of being mentored (specific models) (e.g., Keller, 2005). In the next
section, we describe the general and specific models.
General Models
Attachment theory (Bowlby, 1982) is often cited in research on mentoring AARS (Gormley,
2008). It suggests that attachment patterns which developed over time (secure, ambivalent or
avoidant) influence the likelihood of seeking a mentor or agreeing to be mentored, and thereby
moderate the outcomes related to the mentoring process. Youth raised in fragile environmental
contexts characterized by low parental sensitivity may have developed negative perceptions of self
and others that are believed to undermine their faith in others as sources of support and in the
usefulness of using such support in crisis situations. Certain youth may actively seek help from
others (i.e., ambivalent attachment) only to feel unsupported as a result of their intense focus on
their emotional turmoil or inability to detach from the conflict of their family of origin. Other youth
do not actively seek out support since they have come to expect dysfunctional interpersonal
Academically at-risk students
5
reactions from others and prefer to rely on themselves to overcome their difficulties (avoidant
attachment). In a theoretical paper, Gormley (2008) suggests that insecure attachment styles of both
mentees and mentors can significantly limit the capacity of mentoring to improve youth
development (moderating effect).
Prevention theories are often mentioned when attempting to identify the degree of mentoring
required to promote school completion (Randolph & Johnson, 2008). Mentoring can be based on a
selective prevention approach targeting students who display moderate risk. In such cases, students
are identified using a series of risk and contextual factors and mentoring is provided before they
encounter academic failure. Mentoring may also be offered to those who have experienced repeated
academic failure (indicated prevention). The mentoring activities are then more targeted,
prescriptive and often combined with other approaches, such as tutoring, social skills development
and remedial courses. Finally, mentoring may be used to promote academic success of all students
regardless of their risk level (universal prevention). Intervention in this case is often less intense and
aims mainly at encouraging the vocational development and institutional bonding of students.
Given the highly heterogeneous profiles of the AARS population, we propose that the
application of selective prevention strategies constitutes the most promising avenue for mentoring
AARS. Such an approach helps target negative internal and external influences prior to high-risk
situations becoming overly complex and more difficult to address by mentors and non
professionals. Further, selective prevention strategies may provide mentors guidance in targeting
the kind of support and the activities they might engage in with mentees. This hypothesis is
consistent with the concept of resilience, often used in prevention research, which posits that youth
raised in fragile environmental contexts characterized by economic and social adversity can
Academically at-risk students
6
nevertheless fare well when exposed to protective factors such as natural and formal mentoring
(Randolph & Johnson, 2008).
Another perspective that has shed light on AARS’ mentoring processes is derived from
Mentoring Functions Theory. This theory, which was adapted for mentoring college students (Crisp
& Cruz, 2009), posits four interconnected functions: 1) psychological and emotional support, 2)
support for setting goals and choosing a career path, 3) academic subject knowledge support, and 4)
role modeling. Proponents of this view believe that mentoring serves as a psychosocial resource that
enables AARS to fulfill their security and affiliation needs (function 1). It entails providing
emotional and social support, such as listening and taking part in shared social activities. Mentoring
is also thought to provide AARS with a context that allows them to evaluate their strengths and
aspirations (function 2). Coaching, challenging current plans and progress and goal setting are
examples of actions that fall under this function. For others, mentoring provides AARS with a way
to enhance their knowledge and acquire strategies useful to academic success (function 3). Teaching
study strategies and tutoring are examples of actions stemming from this function. Finally,
mentoring can serve as a locus of identification (function 4). By promoting institutional values and
sharing their experiences about the school, mentors provide mentees with opportunities to bond and
develop attitudes and behaviors that will help them progress.
Specific Models
Three specific models offer interesting research avenues for understanding the mechanisms
involved in mentoring AARS. First, the Youth Mentoring Model (Rhodes, 2005) suggests that
mentoring relationships built on reciprocity, trust and empathy promote the development of AARS
by increasing their social skills and emotional well-being (e.g., asking teacher for help), fostering
Academically at-risk students
7
their cognitive skills (e.g. critical thinking and self-awareness) and exposing them to a positive role
model with whom they can identify. This theoretical approach also suggests that these
“developmental gains” gradually change the dynamics between AARS and the important people in
their lives, such as parents, peers and teachers, and that these changes account for the effects of
mentoring on the subsequent academic and social adjustment of AARS.
Second, the Systemic Model of Youth Mentoring (Keller, 2005) posits that mentor-mentee
relationships must be understood within the larger context of all possible relationships, including
those among mentors, mentees, parents, and teachers. It stipulates that these relationships are
interconnected and that the activities, discussions and emotions experienced during mentoring both
influence and are influenced by the mentee’s other relationships. The model also suggests that the
actions of mentors can indirectly influence parent-adolescent and teacher-adolescent relationships.
When interacting with their mentees, mentors can validate and reinforce the views of parents and
teachers, provide them with autonomy support, which, in turn, can help them better manage their
relationships with authority figures, or simply brighten their lives, generating positive emotions that
can be transposed to other relationships.
Third, unlike the two previous models, the Mentoring Sociomotivational Model proposed by
Larose and Tarabulsy (2005), which was inspired by Self-Determination Theory (Deci & Ryan,
1985), places additional importance on mentor behaviors. According to this model, four sets of
mentor behaviors are critical for improving AARS development: structure, engagement, autonomy
support and competence support. A mentor who establishes clear guidelines in terms of mentoring
objectives, activities and functioning (structure), who openly and respectfully discusses personal,
academic, and career issues with the AARS (engagement), who accepts and validates AARS’s
personal choices without exercising any control or pressure (autonomy support) and who is able to
Academically at-risk students
8
increase the AARS’s feelings of competence following negative experiences (competence support)
should develop a more productive relationship with the student. The resulting positive relationship
is believed to foster AARS’s feelings of competence, relatedness, autonomy, and support which
might improve their social and academic adjustment.
Both general and specific models provide promising avenues for research and intervention with
AARS. First, they suggest the existence of several interactive psychological processes whose effects
may influence relationships other than those between AARS and mentors. These processes include
identification, attachment, critical reasoning, self-reflection, perceived social and instrumental
support, resilience, and satisfaction of motivational needs. Such processes should be part of targeted
experimental and longitudinal investigations in future mentoring research.
Second, several models suggest the importance of multi-level interventions (with at-risk
students, but also with their parents, teachers and peers) and organized actions that reflect the rules
and values of the significant individuals in the lives of AARS. They propose that mentors’ actions
within the mentoring relationship could generate positive collateral effects, which could result in a
less fragile environmental context for AARS. For example, promoting the expectations of parents
and teachers within the mentoring context may improve relationships between AARS and their
parents and teachers, and possibly, encourage parents to become more involved in their youth’s
education, and teachers to build stronger interpersonal bonds with the mentored youth.
RESEARCH
Here, we review findings of empirical studies that: 1) focus on the determinants of student
involvement in mentoring, 2) address theoretical predictors of mentoring relationship quality for
AARS and other student populations; and 3) explore several hypothesized effects and processes
Academically at-risk students
9
driving the mentoring experience of AARS. The studies reviewed have, for the most part, been
identified through the usual data bases, PsycLit and ERIC, with the following key words: youth,
mentoring, student, at-risk student, high school, college, academic adjustment, performance. All
empirical studies published since 2000 that dealt with formal or informal mentoring and included
samples of elementary, high school and/or college students were considered.
Determinants of Participation in Mentoring
In a prospective study, Larose and colleagues (2009) showed that low-risk students with the most
personal resources (i.e., positive help-seeking attitudes, perceived support from friends and school
motivation), but fewer environmental resources (low maternal income and education, leaving home
to attend college, belonging to a family of recent immigrants, not having any siblings who
previously studied in college) were more likely to accept the support of a formal mentor upon
entering college. These observations are in part similar with those reported in a longitudinal study
on natural mentoring involving over 12,000 American youth (Erickson, McDonald, & Elder, 2009),
which found that students with the most personal (e.g., personality, physical appearance, and
college aspirations) and environmental resources (e.g., parental income and education, friends, peer
network centrality) were more likely to have had a mentor in their lives than other students.
These findings suggest that youth who are the least socially and academically equipped have a
lower chance of crossing paths with a natural mentor or of accepting formal mentoring
opportunities. They also propose that youth with modest personal resources living in difficult
circumstances may be open to formal mentoring program participation. Such findings give us a
better sense of the youth who are most likely to form natural mentoring relationships and helps
identify those who will need assistance to establish a formal mentoring relationship through a
Academically at-risk students
10
program. The challenge, then, is to identify strategies to attract to mentoring those with limited
personal resources who might also benefit from mentoring.
Determinants of Mentoring Relationship Quality (MRQ) for AARS
We know that mentors’ background in a helping profession (DuBois, Holloway, Valentine, &
Cooper, 2002), mentor perceptions of their own competence as mentors (i.e., mentor efficacy)
(Karcher, Nakkula, & Harris, 2005), and the previous attachment patterns of AARS (Larose,
Bernier, & Soucy, 2005) are personal factors that affect the MRQ. The quality of training and
supervision offered to mentors (Rhodes & DuBois, 2006), the possibility of choosing one’s mentor
(Kendall, 2007), and the duration and frequency of relationships (Grossman & Rhodes, 2002) are
significant organizational parameters that positively influence MRQ.
The content and manner of negotiating mentoring activities also contribute to MRQ but the latter
process seems more critical. A recent study on mentoring of low risk college students (Larose,
Cyrenne, et al., 2010) found that MRQ was positively influenced by a set of mentor behaviors that
combined emotional engagement and reciprocity with some directivity (i.e., a more authoritative
style). In this study, the MRQ was not found to be affected by specific mentoring content (i.e.,
discussing academic issues, doing specific activities or trying to solve specific problems). Similarly,
in a study of BBBS school-based mentoring students, Karcher, Herrera, and Hansen (2010) reported
unique contributions made to MRQ by both relational and goal-directed activities. They report that
collaboratively negotiated activities of any kind were best for student outcome. Taken together,
these studies suggest that what may matter most for establishing positive bonding with AARS is not
so much the specifics of the mentoring activities per se, but rather the manner in which mentors and
mentees choose the experiences they engage in.
Academically at-risk students
11
The similarity of interests between mentors and AARS also appears to be a determinant of MRQ.
In a research synthesis of mentoring programs intended for different categories of high-risk
adolescents, it has been shown that AARS who perceived high levels of similarity with their
mentors in terms of vision, perspective and values, but not in terms of demographics, reported
greater liking and satisfaction with their mentors (Sipe, 2002). Alternatively, one study has shown
that students considered to be at risk because of low high-school grades displayed more adaptive
behaviors and perceptions in mentoring and earned higher grades when their attachment orientation
to parents was in contrast to their mentor’s relational style (Bernier, Larose, & Soucy, 2005). AARS
presenting dismissing attachment tendencies benefited more from working with mentors who
valued dependency, relationships and closeness, whereas AARS presenting preoccupied/anxious
attachment tendencies hold more gain from working with mentors who valued self-reliance,
achievement and autonomy. Providing the AARS with a challenging relational stance that is not in
line with the student’s own seems to reinforce a process of exploration and change which may
positively affect the course of the relationship.
Effects and Explanatory Processes of Mentoring
Formal mentoring. In the last 15 years, there have been many independent – sometimes small-
scale – experimental and quasi-experimental studies suggesting that formal mentoring of AARS can
lead to changes in a number of cognitive, emotional, and behavioral outcomes. For example, formal
mentoring has been found to improve attitudes toward school and helping, academic confidence,
school connectedness, perceptions of parental and teacher relationships, vocational and reading
skills, participation in college preparatory activities, and persistence in college (see Blinn-Pike,
2007; Eby et al., 2008 for reviews). Yet, conclusions from meta-analyses suggest that the above
Academically at-risk students
12
findings must be interpreted with caution. In fact, the reported effects of formal mentoring were
modest, with Cohen’s d coefficient rarely exceeding .20 (Blinn-Pike, 2007; DuBois, Holloway et
al., 2002). Two such analyses even suggest that the outcomes of formal mentoring for youth are
much more limited than those associated with workplace mentoring and volunteer tutoring (Eby et
al., 2007; Ritter, Denny, Albin, Barnett, & Blankenship, 2006).
Most studies on the impact of formal mentoring involve short-term evaluations. Only three
studies provide exceptions by following up samples some time after the end of the mentoring
program. The first evaluated the Quantum Opportunity Program (QOP) (Rodriguez-Planas, 2009).
QOP was an intense case management and mentoring after-school program for low-achieving
students. The mentoring component involved assigning 15 to 25 youths to case managers with
whom they were expected to develop trusting relationships. Case managers were also required to
establish links with schools, families and friends. In addition to mentoring, QOP offered students
developmental activities, community and educational services and financial incentives. Using a
randomized control trial, the study showed modest short-term gains on high school graduation and
on the pursuit of postsecondary studies, especially among younger adolescents, girls and lower
achieving students. However, these effects decreased over time, and 5 years after the program, the
QOP had no detectible impact on the educational or employment outcome of its participants.
The second study evaluated the long-term effects (15 months after the match) of the BBBS
school-based mentoring program using a randomized control group design and involving the
participation of 10 agencies in the United States (Herrera et al., 2007). Researchers followed AARS
in primary and high school, where 52% of the experimental group students continued to receive
mentoring after one-year. The study showed no long-term impact of the program on the majority of
Academically at-risk students
13
the academic (e.g., general academic performance, GPA, quality of class work) and non-academic
(e.g., substance use, social acceptance, and relationship with parents) outcomes examined.
The challenge of proving long-term effects was also documented for AARS at early elementary
levels. In a prevention program for aggressive elementary school students, Hughes et al. (Hughes,
Cavell, Meehan, Zhang, & Collie, 2005) compared the impact of two mentoring programs, which
lasted 3 semesters and involved college students as mentors. The first program (Prime Time)
combined community-based mentoring (weekly visits) with a focus on child skills training and
consultation for parents and teachers. The second program (Lunch Buddy) was a stand-alone,
school-based mentoring program that involved two visits a week during lunch times in the presence
of non-mentored peers, with a different mentor being assigned each semester. Using a randomized
control trial, authors found that in the short term, both programs reduced children’s externalizing
problems and increased their level of academic and behavioral skills. Surprisingly, at the one- and
two-year follow-ups, the effects observed among participants of the Prime Time program had
completely faded, while Lunch Buddy participants were evaluated by their teachers as displaying a
higher skill level and fewer externalizing problems. This was found to be the case even though the
mentoring relationship of the Lunch Buddy program was of lower intensity than that of the Prime
Time program (i.e., multiple shorter matches rather than one longer match). Factors related to the
quality of interactions with peers were used to explain these findings.
To summarize, the studies that have examined the long-term effects of formal mentoring suggest
that acquired progress is difficult to maintain. Various factors, including the initial characteristics of
students and mentors, the intensity and degree of contact or the nature of the relationship forged
between the mentor and the mentee, may explained this situation. In this context, it is pertinent to
Academically at-risk students
14
examine the factors that potentially buffer or emphasize the effects of formal mentoring on AARS
(i.e., moderating factors).
Moderating processes. Although risk level has often been considered an important moderator of
mentoring effects, research findings in this area are mixed. Stronger effects for high-risk students on
social and behavioral adjustment indicators were reported when student risk was estimated using
academic indicators (e.g., low GPA, high absence rate from school; Whiting & Mallory, 2007;
Rodriguez-Planas, 2009). Some studies also suggest that boys with low academic scores are
especially responsive to formal mentoring (Whiting & Mallory, 2007). Other studies have found
that mentoring was helpful for all students, but that risk moderated mentoring outcomes. For
example, in a study evaluating a program that used both peer leaders and adult mentoring, results
showed that high-risk students (grades below 70% and/or more than 8 school absences) improved
their ability to resist peer pressure during the transition to high school, while low-risk students
enhanced their capacity for making friends (Holt, Bry, & Johnson, 2008). Conversely, Morrow-
Howell and colleagues (2009) showed that the effect of formal mentoring on reading
comprehension was less pronounced for a higher risk, special education group of students than for
non-special education students, suggesting that the impact of mentoring may be less evident when a
student’s risk level presents greater psychological or learning challenges. Finally, one study has
found that when the risk was based on the student’s feelings of being disconnected from school,
having a mentor holding positive views regarding youth behaviors lead AARS to be more
emotionally engaged in the mentoring relationship and, subsequently, to report stronger
relationships with their teachers (Karcher, Davidson, Rhodes, & Herrera, 2010).
The hypothesis that greater psychological or learning difficulties moderate the youth mentoring
outcomes is consistent with the studies that include emotional indicators in their definition of risk.
Academically at-risk students
15
Among students with lower high school grades, but sufficient to be admitted to college, those who
reported more insecure attachment relationships to their parents were less likely to hold positive
perceptions of mentoring and, subsequently, more likely to report high levels of conflict with their
teachers (Larose, Bernier, & Soucy, 2005). This latter finding underlines the challenge of
mentoring AARS when emotional factors are considered.
Certain characteristics of mentors themselves also seem to play an important role, in addition to
mentee levels of risk. The effects of mentoring AARS appear to be greater when mentors feel they
have the competence to help (Parra, DuBois, Neville, Pugh-Lily, & Pavinelli, 2002), hold highly
positive attitudes towards youth (Karcher, Davidson, Rhodes, & Herrera, 2010), express high
motivation for self-enhancement (Karcher, Nakkula, & Harris, 2005), and when they have a
background in a helping profession (DuBois, Holloway et al., 2002), are not poor, and are not
married (Grossman & Rhodes, 2002). Studies also suggest that high school and college student
mentors have more difficulty generating positive results among at-risk mentees than adult mentors
or teachers (Whiting & Mallory, 2007; Hughes et al., 2005; Converse & Lignugaris/Kraft, 2009).
Some contextual factors also play a role in program effectiveness. These factors include initial
and ongoing, structured training for mentors, monitoring of program implementation and parental
involvement (DuBois, Holloway et al., 2002). In addition, programs viewed as effective allow for
more extended contact between mentors and AARS, engage AARS in social and academic
activities, and structure interactions in such a way as to favor the development of AARS autonomy
in decision making (Morrow-Howell et al., 2009; Blinn-Pike, 2007).
Mediating processes. While several models suggest the presence of different mediating processes
to explain mentoring outcomes (see THEORY), such processes have rarely been studied. Rhodes
and colleagues (Rhodes, Grossman, & Resch, 2000; Rhodes, Reddy, & Grossman, 2005) found that
Academically at-risk students
16
improvements in the quality of relationships with parents mediated the impact of the BBBS
community match programs on youth self-worth, school value, and grades, as well as on peer
relationships, but only for youth involved in relationships that lasted more than 12 months. This
study included 959 adolescents and used a randomized control trial design. Authors claim that this
mediating process may have been the result of a gradual change in mentee representation models of
relationships with parents or a decrease in the tension experienced between youth and their parents.
They also suggested that short-term mentoring does not provide sufficient opportunity for the
development of secure relationships. Similarly, other reports revealed that perceived connectedness
to parents and perceptions of support from significant adults outside mentoring act as mediators of
the link between formal mentoring and mentee academic adjustment (Karcher, Davis, & Powell,
2002; DuBois, Neville, et al., 2002).
Two quasi-experimental studies explored the role of interpersonal processes in mentoring as
mediators of AARS outcomes. The first study was conducted with students who had failed more
than half of their first semester courses. Results showed that mentees who had a positive working
alliance with their mentors (i.e., agreement on goals, positive bonding) were more likely to improve
their academic competence, participation in class, tendency to seek help from teachers, and
academic perseverance than were mentees in less collaborative mentoring relationships or students
in a control group (Larose, Monaghan, Chaloux, & Tarabulsy, 2010). The second study, conducted
with low-achieving students, indicated that youth-perceived autonomy and relatedness support in a
teacher-student mentoring relationship led to better academic adjustment (Larose, Tarabulsy, &
Cyrenne, 2005).
Academically at-risk students
17
Such studies support the argument laid out in several models, suggesting that the impact of
mentoring on dropout prevention may be explained in part by the satisfaction of motivational needs,
the quality of the mentoring relationship, and an improvement in youth relationships.
Natural mentoring. Studies on the impact of natural mentoring of AARS are much less common.
Here, we present the conclusions of one of the largest longitudinal studies conducted in the United
States on natural mentoring and education (Erickson et al., 2009). Beyond the effect of personal and
environmental resources (e.g., parental income and education, number of friends, school size,
physical appearance, and personality), natural mentoring was positively associated with academic
performance in high school and with subsequent youth’s educational status (i.e., highest degree
achieved). For youth with existing resources, natural mentoring with relatives (e.g., brother, sister,
grandparent) had a positive effect on educational status. When these resources are limited,
mentoring by a teacher was identified as having the most positive impact. This last finding is
consistent with observations made by DuBois and Silverthorn (2005), indicating that at-risk youth
with non-familial mentors were more likely to complete high school than those who identified a
familial mentor. In addition to having an impact on AARS in the classroom, teachers appear to be
meaningful models for preventing student dropout among this population.
The number of natural mentors, their characteristics and the type of relationships they build with
their mentees also play a significant role in preventing academic difficulties. In a correlational study
among 140 Latino from an urban public high school deserving a predominantly low-income student
population (Sanchez et al., 2008), the presence of a natural mentor was found to be associated with
fewer absences in class, greater educational expectations, and strong sense of school belonging.
Further, the number of reported mentors was positively related to academic outcomes. Finally,
Academically at-risk students
18
mentors’ education, frequency of contact, relationship length, and total form of support provided by
mentors were positively related to student academic outcomes.
In sum, research on mentoring AARS over the last 15 years has mainly addressed the short term
impact of mentoring and its moderating processes. Research on natural mentoring is clearly less
abundant than that on formal mentoring. As such, many issues will require further careful analysis
in future works: What factors attract youth to formal mentoring? How can mentoring reach the most
emotionally at-risk youth? Who is best positioned and skilled to support AARS? What are the long-
term effects of formal and natural mentoring? What developmental mechanisms account for these
long-term effects? On the methodological front, longitudinal follow-up, the systematic
quantification of size effects, the use of randomized designs, including various “mentoring”
conditions, and controlling for confounding factors before and during the mentoring experience are
important parameters that must be accounted for in future research.
PRACTICE
This review draws out the need to implement mentoring programs with great care and by
considering an important number of parameters. The following are a series of key reflective actions
which we perceive to be important in designing, implementing and managing mentoring programs
for AARS (see Table 1). One such action specifically concerns natural mentoring.
Program Promotion, AARS Screening and Mentor Selection
It has been shown in the research section (e.g., Larose et al., 2009) that students who lack
personal resources are not as likely to have the opportunity to be mentored or to agree to take part in
a formal program. Thus, program managers may wish to implement promotional approaches that
Academically at-risk students
19
will help AARS understand that mentoring is not a threat, but rather a useful tool that may fulfill
personal and academic needs. Furthermore, advertising the instrumental (e.g., help with school
work) and vocational functions (e.g., enhanced knowledge of the job market) of mentoring may
attract more AARS, particularly some boys who may feel uncomfortable in close relationships.
Few programs establish diagnostic profiles of the youth they seek to help. Yet doing so might be
helpful to identify the degree and origin of an AARS’s difficulties and therefore better prepare
mentors adapt the quantity of mentoring provided. Even major national mentoring programs, such
as the QOP or the American BBBS school-based mentoring program, were not able to generate
meaningful long-term outcomes on the academic success and educational status of AARS,
indicating the need to better understand and be responsive to the needs of participating students,
potentially improving the long-term effect of intervention.
The reviewed studies also underline the importance of choosing teachers as mentors for AARS
(e.g., Converse & Lignugaris/Kraft, 2009) and questions the effectiveness of programs that call on
high school and college mentors (e.g., Whiting & Mallory, 2007), except those who are especially
well suited to serve as mentors. While it cannot be concluded that younger mentors should be
excluded from mentoring programs devoted to AARS, they should be carefully selected based on
maturity, social skills and interpersonal abilities. In our view, young mentors should also be
provided with substantial training and supervision. It may be helpful that they be mentored
themselves by more experienced adult mentors.
Mentor Training and Supervision
Ten years ago, some studies estimated that nearly half of the mentoring programs implemented
in schools contained no more than two hours of mentor training (see Larose & Tarabulsy, 2005). A
Academically at-risk students
20
recent national survey on practices of BBBS programs in the United States suggest that the situation
has not significantly progressed since that time. In fact, only 1% of the agencies surveyed reported
providing their mentors more than 4 hours of initial training (BBBSA National Office, 2009). Initial
training and mentor supervision are key practices that should not be neglected.
Fortunately an increasing number of programs devoted to AARS are investing equal amounts of
resources in training and supervising mentors. For example, the MIRES program (Larose et al.,
2011) which aims at helping students complete science programs at the college level, offers two
days of initial training, as well as several individual and group meetings throughout the mentoring
program. It also features a hands-on component, whereby mentors practice interacting with a
mentee through simulated videotaped scenarios under the supervision of instructors. The program
and mentoring meetings are structured on the basis of the Sociomotivational Model of Mentoring.
Mentors learn how to structure mentoring relationships on an instrumental and emotional level, and
to support mentee autonomy.
While the content of training sessions might vary significantly from one mentoring program to
another, research suggests that certain goals should be prioritized when intervening with AARS
(Tyler & Lofstrom, 2009; Keller, 2005; Crisp & Cruz, 2009; Larose, Cyrenne et al., 2010). These
are: to identify the protection and risk factors of school dropout, to enhance the mentors’ abilities
for empathy, authenticity, and collaboration, to learn how to practice efficient academic problem
solving, to understand the potential effects of the student’s academic and interpersonal profile
regarding their expectations, emotions and behaviors throughout mentoring, and to intervene with
parents, peers and teachers. We also believe that supervision should both equip mentors to respond
to the timely needs of mentees and aim at consolidating their perceived competence in their ability
Academically at-risk students
21
to be helpful. Modeling and constructive feedback may be used by supervisors to achieve these
goals in their meetings with mentors.
Matching Mentors and Mentees
Research suggests that matching AARS and mentors based on similarity of views help to
strengthen the quality of the bond (Sipe, 2002). Allowing AARS to select their mentor appears to
be a promising strategy to achieve this goal (Kendall, 2007). However, freely choosing a mentor
implies complex logistics and a level of resources that most organizations do not have. Matching
AARS and mentors must therefore be based on criteria that respect both the mentee and the
resources available to the mentoring organization, in a way that will maximize the quality of the
future relationship. Such criteria may include personal, social, and vocational interests, mentoring
expectations, and views about the implication of parent and teacher in the mentoring process.
The Mentoring Relationship
Ensuring frequent, high-quality contact is a recommendation that applies to all mentoring
programs, regardless of the specific clientele targeted. In light of the findings on the long-term
effects of mentoring AARS (e.g., Herrera et al., 2007), the issues that must be raised concern the
intensity and responsiveness of the specific mentoring experience. Should a youth experiencing a
difficult single-parent context and one who has had behavioral problems since kindergarten be
offered the same type of mentoring? At the elementary level, does a student with reading and
writing difficulties not require a more specialized mentor than another experiencing anxiety about
the transition to high school? Is one year of mentoring sufficient to lessen the impact of a high-risk
situation, which in certain cases, began in early childhood? In our opinion, the quantity of
Academically at-risk students
22
mentoring must be adjusted to the level of risk for academic failure and, in certain cases, an offer
of mentoring over several years might be considered.
Reinforcing the link between mentor and the AARS’ teacher is an aspect of mentoring
practice that merits further attention. This type of action is at the core of several intervention and
mentoring programs, such as The Check and Connect Program (Sinclair, Christenson, &
Thurlow, 2005) and The Achievement Mentoring Program (Holt et al., 2008), intended for
primary and high school AARS. Within such programs, mentors are not only expected to conduct
a systematic follow-up of mentees’ academic behavior and achievement, but also of being in
constant contact with the mentees’ teachers. In this context, the mentor becomes a kind of
mediator, informing teachers of mentee progress, so that teachers may be sensitive to these
aspects and reinforce this progress within the classroom setting.
Natural Mentoring
Natural mentoring is a unique experience that benefits the youth involved. Unfortunately, those
who need it most have fewer opportunities for accessing it. This knowledge alone justifies the
importance of implementing mentoring programs. It also calls upon schools and communities to
consider the predominant culture in their environments and adopt measures that will foster the
creation of natural mentoring relationships between adults and youth.
CONCLUSION
In conclusion, many scientists and practitioners view mentoring as a preventive action having
the potential to support academic functioning in AARS and reduce the risk of dropping out of
Academically at-risk students
23
school (Tyler & Lofstrom, 2009; Wheeler et al., 2010). However, the causal relation between
taking part in mentoring programs and the reduction of high school dropout are not as convincing
as it might be. The body of knowledge reviewed in this chapter is vast, though highly focused on
the short-term effects of programs implemented in schools and communities. Future research
would do well to address the long-term outcomes of mentoring relationships and examine the
dynamics within these relationships (e.g., mentor strategies, mentee behaviors, conflicts,
activities). In pursuing both of these objectives, the highly heterogeneous profiles of at-risk
students should also be considered. This will no doubt lead to more comprehensive and nuanced
intervention models that will better meet the needs of the AARS population.
REFERENCES
BBBSA National Office (2009). Analysis of responses to agency practices survey for Big
Brothers Big Sisters of America’s community-based mentoring program. Compiled by Wheeler M.,
& DuBois D.L., Unpublished manuscript.
Bernier, A., Larose, S., & Soucy, N. (2005). Academic mentoring in college: The interactive role
of student’s and mentor’s interpersonal dispositions. Research in Higher Education, 46, 29-51.
Blinn-Pike, L. (2007). The benefits associated with youth mentoring relationships, In T. D.
Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspective approach
(pp. 165-187). Malden: Blackwell Publishing.
Bowlby, G. (2005). Provincial drop-out rates - trends and consequences. Education Matters:
Insights on Education, Learning and Training in Canada, 2.
Bowlby, J. (1982). Attachment and loss: Attachment (Vol 1., 2nd ed.). New York: Basic Books.
Academically at-risk students
24
Cataldi, E. F., Laird, J., KewalRamani, A., & Chapman, C. (2009). High school dropout and
completion rates in the United States: 2007 compendium report (No. NCES 2009-064): U.S.
Department of Education, National Center for Education Statistics.
Crisp, G., & Cruz, I. (2009). Mentoring college students: a critical review of the literature
between 1990 and 2007. Research in Higher Education, 50, 525-545.
Converse, N. & Lignugaris/Kraft, B. (2009). Evaluation of a school-based mentoring program
for at-risk middle school youth. Remedial and Special Education, 30, 33-46.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum Press.
DuBois, D. L., Holloway, B. E., Valentine, J.C., & Cooper, H. (2002). Effectiveness of
mentoring programs for youth, a meta-analytic review. American Journal of Community
Psychology, 30, 157-197.
DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and
adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69-92.
Eby, L. T., Allen, T. D., Evans, S. C., Ng, T., & DuBois, D. L. (2008). Does mentoring matter?
A multidisciplinary meta-analysis comparing mentored and non-mentored individuals. Journal of
Vocational Behavior, 72, 254-267.
Erikson, L. D., McDonald, S., & Elder Jr., G. H. (2009). Informal mentors and education:
complementary or compensatory resources? Sociology of Education, 82, 344-367.
Gormley, B. (2008). An application of attachment theory: Mentoring relationship dynamics and
ethical concerns. Mentoring & Tutoring: Partnership in Learning, 16, 45-62.
Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in
youth mentoring relationships. American Journal of Community Psychology, 30, 199-219.
Academically at-risk students
25
Herrera, C., Baldwin Grossman, J., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z.
(2007). Making a difference in school: The Big Brothers Big Sisters school-based mentoring
impact study. Philadelphia: Public/Private Ventures
Holt, L. J., Bry, B. H., & Johnson, V. L. (2008). Enhancing school engagement in at-risk, urban
minority adolescents through a school-based, adult mentoring intervention. Child & Family
Behavior Therapy, 30, 297-318.
Hughes, J. N., Cavell, T. A., Meehan, B., Zhang, D., & Collie, C. (2005). Adverse school
context moderates the outcomes of two selective intervention programs for aggressive children.
Journal of Consulting and Clinical Psychology, 73, 731-736.
Karcher, M. J., Davidson, A. J., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in the program:
The role of teenage peer mentors’ attitudes in shaping their mentees’ outcomes. Applied
Developmental Science, 14, 212-227.
Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on
connectedness and academic achievement. The School Community Journal, 12, 35-50.
Karcher, M. J., Nakkula, M. J., & Harris, J. (2005). Developmental mentoring match
characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality.
The Journal of Primary Prevention, 26, 93-110.
Keller, T. E. (2005). A systemic model of the youth mentoring intervention. The Journal of
Primary Prevention, 26, 169-188.
Kendall, D.L. (2007). Does choice matter? The impact of allowing proteges to select their own
mentors (Doctoral dissertation), University of Central Florida, United States - Florida.
Academically at-risk students
26
Larose, S., Bernier, S., & Soucy, N. (2005). Attachment as a moderator of the effects of security
in mentoring on subsequent perceptions of mentoring and relationship quality with college teachers.
Journal of Social and Personal Relationships, 22, 399-415.
Larose, S., Cyrenne, D., Garceau, O., Harvey, M., Guay, F., Godin, F., Tarabulsy, G. M., &
Deschênes, C. (2011). Academic mentoring and dropout prevention for students in math, science
and technology. Mentoring & Tutoring: Partnership in Learning, 19, 419-439.
Larose, S., Cyrenne, D., Garceau, O., Brodeur, P. & Tarabulsy, G. (2010). The structure of
effective academic mentoring in late adolescence. New Directions for Youth Development, 126,
123-140.
Larose, S., Cyrenne, D., Garceau, O., Harvey, M., Guay, F., & Deschênes, C. (2009). Personal
and social support factors involved in students’ decision to participate in formal academic
mentoring. Journal of Vocational Behavior, 74, 108-116.
Larose, S., Monaghan, D., Chaloux, N., & Tarabulsy, G. (2010). Working alliance as a
moderator of the impact of mentoring relationships among academically at-risk students. Journal of
Applied Social Psychology, 40, 2656-2686.
Larose, S., & Tarabulsy, G. (2005). Academically at-risk students. In D. L. DuBois & M. J.
Karcher (Eds), Handbook of youth mentoring. (pp. 440-453). Thousand Oaks, CA: Sage
Publications.
Larose, S., Tarabulsy, G., Harvey, M., Guay, F., Deschênes, C., Cyrenne, D., & Garceau, O. (in
press). Impact of a college student academic mentoring program on perceived parental and teacher
educational involvement. Journal of Applied Social Psychology.
Academically at-risk students
27
Morrow-Howell, N., Jonson-Reid, M., McCrary, S., Lee, Y., & Spitznagel, E. (2009).
Evaluation of Experience Corps, Student Reading Outcomes, (CSD Research Report 09-01). St.
Louis, MO: Washington University, Center for Social Development.
Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Pavinelli, N. (2002). Mentoring
relationships for youth: Investigation of a process-oriented model. Journal of Community
Psychology, 30, 367-388.
Pour la Solidarité, (2009). L’Union européenne s’intéresse-t-elle au décrochage scolaire? État
des lieux et perspectives en Europe. Série Affaires sociales.
Randolph, K. A., & Johnson, J. L. (2008). School-based mentoring programs: A review of the
research. Children and Schools, 30, 177-185.
Rhodes, J. E. (2005). A model of youth mentoring. In. D. L. DuBois and M. J. Karcher (Eds).
Handbook of youth mentoring (pp. 30-43). Thousand Oaks, CA: Sage.
Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring
movement. Social Policy Report, 20, 3–20.
Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: Pathways through
which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71,
1662-1671.
Rhodes, J. E., Reddy, R., & Grossman, J. B. (2005). The protective influence of mentoring on
adolescents’ substance use: Direct and indirect pathways. Applied Developmental Science, 9, 31-47.
Ritter, G., Denny, G., Albin, G., Barnett, J., & Blankenship, V. (2006). The effectiveness of
volunteer tutoring programs: A systematic review. The Campbell Collaboration Reviews of
Intervention and Policy Evaluations (C2-RIPE). Philadelphia, Pennsylvania: Campbell
Collaboration.
Academically at-risk students
28
Rodríguez-Planas, N. (2009). Longer-term Impacts of Mentoring, Educational Services, and
Incentives to Learn. Retrieved from http://www.cemfi.es/research/seminars/pewpa.asp?lang=en
Sanchez, B., Esparza, P., & Colon, Y. (2008). Natural mentoring under the microscope: An
investigation of mentoring relationships and Latino adolescents’ academic performance. Journal of
Community Psychology, 36, 468-482.
Sinclair, M.F., Christenson, S.L., & Thurlow, M.L. (2005). Promoting school completion of
urban secondary youth with emotional or behavioural disabilities, Exceptional Children, 71, 465-
482.
Sipe, C. L. (2002). Mentoring programs for adolescents: A research summary. Journal of
Adolescent Health, 31, 251-260.
Tyler, J. H., & Lofstrom, M. (2009). Finishing high school: Alternative pathways and dropout
recovery. The Future of Children, 19, 77-103.
Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials
of school-based mentoring: Making sense of mixed findings. Social Policy Report from SRCD,
24(3), 1-26.
Whiting, M. A., & Mallory, J. E. (2007). A longitudinal study to determine the effects of
mentoring on middle school youngsters by nursing and other college students. Journal of Child and
Adolescent Psychiatric Nursing, 20, 197-208.
Wood, S. & Mayo-Wilson, E. (2012). School-based mentoring for adolescents: A systematic
review and meta-analysis Research on Social Work Practice. Online article found at:
http://rsw.sagepub.com/content/early/2012/01/03/1049731511430836.
Academically at-risk students
29
Table 1
Key Reflective Actions for Practitioners Working with Academically At-Risk Students
________________________________________________________________________
1. Promote formal mentoring among AARS with fewer personal resources (e.g., negative
help-seeking attitudes, low school motivation) by targeting them in recruitment efforts (both
seeking them out and promoting their participation through general invitation messages).
2. Quickly identify the factors involved in a youth’s lack of academic success (e.g., personal,
contextual) and determine the nature and level of risk, by using diagnostic profiles.
3. Attract mentors who are teachers or who have experience in the helping professions, such
as is currently practiced in the Achievement Mentoring Program (Holt, Bry, & Johnson, 2008).
4. Offer mentors comprehensive initial training that aims to achieve the following goals: a)
identify the protection and risk factors of school dropout; b) enhance abilities for empathy,
authenticity, and collaboration; c) learn how to practice efficient academic problem solving; d)
understand the potential effects of the student’s academic and interpersonal profile on
expectations, emotions and behaviors throughout the mentoring process.
5. Encourage mentors to intervene with parents and teachers (e.g., serving as an advocate) by
informing them about program objectives, rules and policies, mentor profile and role, and
program expectations with regard to parent and teacher collaboration.
6. Allow AARS to select their mentors or match them based on shared academic or
vocational interests.
7. Offer mentors support and supervision during the intervention and foster their perception
of competence. Providing mentors early feedback on their mentee's view of the match may
serve to bolster mentor efficacy.
8. Adjust the quantity of mentoring based on the student’s level of risk, and be clear about the
expected duration of the relationships (e.g., one year, one semester, longer).
9. Establish mechanisms to support mentoring relationships over multiple years, ideally
across key normative school transitions (e.g., middle school to high school). Provide mentors
and AARS with support to fulfill this goal.
10. Create academic cultures that value natural mentoring between adults and AARS through
strategies such as social marketing, focus groups, and social network connections.
________________________________________________________________________
