pp. 38-60
Journal of Experiential Education • 2006, Volume 29, No. 1
Reducing Academic Achievement
Gaps: The Role of Community
Service and Service-Learning
Peter C. Scales, Eugene C. Roehlkepartain, Maryheth Neal,
fames C. Kielsmeier, and Peter L. Benson
Three large and diverse data sets were used to study the relations among
6th-12th grade students' community service and service-learning experiences, academic success, and socioeconomic status (SES). Principals in
high-poverty, urban, and majority nonwhite schools were more likely to
judge service-learning's impact on student attendance, engagement, and
academic achievement as very positive. Students with higher levels of
service/service-learning reported higher grades, attendance, and other
academic success outcomes. Low-SES students with service/servicelearning scored better on most academic success variables than their lowSES peers with less or no service or service-learning. Service-learning may
be especially attractive to principals oflow-SES schools, in part because it
may be related to higher achievement generally and to smaller achievement gaps between higher- and lower-income students.
Key Words: Community Service, Service-Learning, Academic
Achievement, Achievement Gaps
Feter C. Scales is a Senior Fellow, Office of the Fresident, at the Search Institute.
E-mail: [email protected]
Eugene C. Roehlkepartain is the Director of Family and Congregation Initiatives,
at the Search Institute. E-mail: [email protected]
Marybeth Neal is the Director of Research for the National Youth Leadership
Council. E-mail: [email protected]
James C. Kielsmeier in the Fresident of the National Youth Leadership Council.
E-mail: [email protected]
Feter L. Benson is Fresident and Chief Executive Officer of the Search Institute.
E-mail: [email protected]
This research was supported by a generous grant to the National Youth Leadership Council
from the State Farm Foundation. The opinions expressed are those of the authors.
2006, Volume 29, No. 1
39
romoting student achievement and greater equity in achievement of
different student groups are educational priorities that are difficult to
accomplish (Borman, Hewes, Overman, & Brown, 2002; Eamon,
2002; Johnson, 1997; Louis, Toole, & Hargreaves, 1999). Over the past 20
years, policymakers and practitioners have tried to raise levels of achievement and lessen traditional achievement gaps through several broad school
reform approaches. Most recently, the curriculum standards movement
and its associated yardstick, the standardized test, has become the dominant force organizing American elementary and secondary education
(Olson, 2000). States have scrambled to raise standards and require students to pass statewide subject matter tests. States are also issuing "report
cards" on how schools are performing and have toughened graduation
requirements by focusing curriculum on state standards and by lengthening the school year. There has been considerable debate regarding whether
the goal of providing equitable, high-quality education that meets comprehensive developmental needs is being sacrificed to the mandate to demonstrate adequacy on more narrowly conceived standardized achievement
tests (see, for example, Oakes, Quartz, Ryan, and Lipton, 2000). Some
scholars have argued that the testing focus has resulted in an overemphasis even in preschool children on structured learning versus play, contrary
to research on how best to promote early growth in positive learning
orientations and abilities (Hirsh-Pasek & Golinkoff, 2003). Indeed,
Duckenfield and Drew (2006) argue that the current focus on standardized
testing fails to measure "variables that matter because they enhance learning, variables that would indicate whether a student will succeed in school
and in life" (p. 33). Experiential education—particularly community service and service-learning—may help contribute to achieving academic outcomes while being developmentally responsive (Billig, 2004). The current
study more deeply explores whether service and service-learning could
play a role in improving achievement among low-income students.
P
Review of the Literature
Achievement and SES
Research repeatedly shows that low-SES is linked to a wide range of
indicators of child and adolescent well-being, including student academic achievement (Beauvais & Jensen, 2003). Research repeatedly shows that
low-SES is negatively linked to a wide range of indicators of child and
Journal of Experiential Education
adolescent well-being, including student academic achievement
(Beauvais & Jensen). According to a study by Evans (2004), lower income
children have less stable families, greater exposure to environmental toxins and violence, and more limited extra-familial social support networks.
Evans additionally finds that they are cognitively stimulated less than
higher income children, from reading less and being read to less, to experiencing less complex communications with parents involving more limited vocabulary. In the preschool years, even a small increase in family
annual income, for example, has been found to be significantly associated
with increased cognitive development (e.g., social skills, knowledge of letters and shapes, vocabulary)—a large contributor to school readiness—as
children grow from ages 15 months to 3 years (Dearing, McCartney, &
Taylor, 2001). In elementary, middle, and high school, the parents of poor
children are less involved in school activities, and the children themselves feel less connected to school (Evans). They also are twice as likely
to attend schools with less-qualified and less-experienced teachers, especially in schools where the majority of students are nonwhite (Mayer,
Mullens, Moore, & Ralph, 2000).
Youth from low-income settings are disproportionately youth of
color; about 40% of African American and Hispanic students are poor as
compared to less than 20% of White youth (U.S. Department of Education,
2000). Youth from low-income families are more likely to attend schools
with the least resources and the least qualified teachers (Clements &
Mclntyre, 2004; Mayer et al., 2000). For example, Louis and colleagues
(Louis et al., 1999) noted the positive effects of Head Start, but that the
minority graduates of Head Start programs still tend to enroll in less wellftmded and less supportive school settings than middle-class students.
Low-income youth are more often taught using memorization, drills, and
other basic instructional methods that are not conducive to engagement or
learning, and they suffer from generally lower expectations for their
achievement (Evans, 2004).
Recent trends show some narrowing of traditional educational
inequities across socioeconomic and ethnic lines, but disparities remain.
For example, a recent study of the San Diego Unified School District
showed that students in the poorest 20% of schools took until grade 10 to
reach the same average reading level as that of the most affluent 20% at
grade 5 (Betts, Zau, & Rice, 2003). Although dropout rates have stabilized
for all income groups since 1990, students from lower income families
(the lowest 20%) drop out of school at double the rate of middle-income
students and six times the rate of students in the upper 20% of income
(U.S. Department of Education, 2004).
2006, Volume 29, No. 1
41
However, other factors have long heen recognized as equally if not
more important than SES. For example, Wang's (1990) meta-analysis concluded that participation in extracurricular activities, and cooperative,
active student learning methods explained ahout as much variance in
achievement as SES. More recently, Wenglinsky (2002) reported classroom
practices such as the use of hands-on learning had eifects comparahle to
or greater than SES. Thus, a developmentally attentive school environment and a more experiential curriculum may help compensate for the
detrimental effects of low-SES.
Developmental Approaches to Promoting School Success
A complementary strategy to the standards-based approach to promoting school success might be called developmental attentiveness
(Benson, Scales, & Mannes, 2004), or human development as a core
achievement strategy. The developmental attentiveness approach links
school reform with the developmental needs of children and adolescents
and the broader community environment. The central assumptions of this
approach are: (a) restructuring the school experience to provide a better
"fit" with the developmental needs of children and adolescents will lead
to greater achievement for all (Eccles et al., 1993); and (b) all elements of
the young person's environment (family, school, peers, and community)
play both independent and intertwined roles in contributing to positive
development (Benson, Leffert, Scales, & Blyth, 1998; Greenberg et al.,
2003). Research has suggested, for example, that students with higher levels of 40 "developmental assets," including caring school climate, community service, and school engagement, have higher levels of thriving
indicators such as school grades and lower levels of risk hehaviors such as
violence or suhstance ahuse, with the linkages hetween assets and positive
outcomes comparahle across race/ethnicity, and between urban and
nonurban students (Benson et al., 1999; Scales et al., 2005; Sesma &
Roehlkepartain, 2003).
Some of the elements of the developmental attentiveness approach
include: promoting closer relationships among students and teachers;
"authentic" instruction that engages students and connects suhject matter
to real-world problems, such as through service-learning programs;
extended learning time, including plentiful co-curricular programs; and
partnerships with community organizations (Carnegie Council on
Adolescent Development, 1989; Council of Great City Schools, 1999;
Education Trust, 1999; Lipsitz, Jackson, & Austin, 1997). Studies have
found that these practices are related to student engagement, effort, and
competency heliefs (Chen & Stevenson, 1995; Patrick, Hicks, & Ryan,
1997) as well as to dropout rates and grades (Fellner et al., 1997; Lee &
42
Journal of Experiential Education
Smith, 1993; Lee, Smith, & Croninger, 1995; Mahoney & Cairns, 1997;
Newmann, Lopez, & Bryk, 1998).
Such elements have long been staples of experiential education in
general. As Sheckley and Keeton (1997) noted. Experiential Learning
Theory posits that concrete experiences help students grasp information
when students can reflect on those experiences and experiment actively
with the concepts they are learning. Experiential education can provide
greater depth of information processing, and thus a greater potential
impact on learning, than less active methods. An important part of the
process of experiential learning is the element of "surprise" in the experience, that is, the degree to which it does not easily and immediately confirm an individual's prior expectations. In Sheckley and Keeton's (1997)
model, the "expectations, values, images, and models of meaning developed from prior experience that learners hold in semantic memory," (p.
50), and the cultural values, norms, and beliefs they have assimilated,
interact with the concrete experience to confirm or disconfirm prior
expectations, values, etc. Without strong opportunities for reflection on
the experience, learners will tend to assimilate the experience into their
existing models of meaning. But with the structured opportunity to reflect,
describe, discuss, and construct meaning from fhe experience, learners
have the potential to develop more complex understandings and more
comprehensive intellectual functioning.
Academic Effects of Community Service and Service-Learning
Service-learning has tremendous potential in the lives of marginalized youth—those who typically do not participate in community activities [sometimes described as vulnerable, disadvantaged, or at-risk youth).
There is evidence emphasizing testing may drive up school dropout rates,
in part because these approaches do not engage disengaged students but
push them further from schooling (Massachusetts Department of
Education, 2005; Nichols, Glass, and Berliner, 2005). According to the
research of the National Dropout Prevention Center at Clemson University
in Clemson, South Carolina, Duckenfield and Drew (2006) conclude that
over the past 15 years, the best research-based dropout prevention strategies include school/community collaboration, family engagement, early
literacy development, and service-learning. An evaluation of the National
Service-Learning Initiative and the Generator Schools Project concluded
that students who are most disengaged from school when they entered a
service-learning program were most likely to experience positive change
(Blyth, Saito, & Berkas, 1997). Students who were most at risk or more disengaged from school when they got involved in service-learning saw a
number of positive changes during the time of their involvement. By the
end, they were more likely to:
2006, Volume 29, No. 1
43
• believe they were contributing to the community;
• be less bored than in traditional classrooms;
• be engaged in academic tasks and general learning; and
• be more accepting of diversity.
Billig's (2006) review of research on how students learn suggests
three major characteristics of effective learning environments: Engaging
schools build students' competence, in part, by giving them more control
over their learning than less engaging schools; they promote academic values and goals; and they promote a sense of belonging. In keeping with
such findings, and the theoretical reasoning described above, Newmann,
Wehlage, and Lamborn (1992) note that student engagement increases
when curriculum involves students in the construction of knowledge,
ownership of the cognitive work, and authentic connection to the "real
world" and community. Service-learning is a primary example of experientially engaging students in such "shared inquiry," meaningful decisionmaking, and integration of classwork and community life (Zeldin, 2004),
all of which support disadvantaged students in both their academic and
commimity involvements. For example, Billig's (2004) extensive review of
the service-learning research concluded that students participating in
service-learning generally do better than others on school engagement,
attitudes toward school, attendance, communication with parents about
school, test scores, grade point average, and problem-solving skills.
Most of this research is cross-sectional, and so cannot inform causality. But Scales and Roehlkepartain (2004a) reported on a longitudinal
study that more validly suggests the contribution that service might make
to higher grades. They found that consistent or "emerging" volunteers
(those who started volunteering in grades 7 to 9 rather than in grades 6 to
8) had significantly higher grade point averages (GPAs) in high school than
those who never volunteered or "faded" in their volunteering (served in
grades 6 to 8 but not afterward). Students who had greater "connection to
community" in middle school, including participating in community
service, youth programs, and religious community, were three times more
likely than others to have a B-H or higher average three years later in high
school (Scales, Benson, Roehlkepartain, Sesma, & van Dulmen, in press).
Hanson, Austin, and Lee-Bayha (2003) studied 7th, 9th, and 11th
grade students in nearly 1,700 California schools. They reported that student "opportunities for meaningful participation" in schools, homes, communities, and among peers were strongly related to standardized test
scores. Certainly, the causal connection can work in both directions, with
well-performing students enjoying closer relationships with adults and
more opportunities for leadership, responsibility, and contribution.
44
Journal of Experiential Education
Nevertheless, community service and service-learning clearly provide
opportunities for students to experience meaningful participation in various life contexts. If meaningful participation has any role in contributing
to school success, then service experiences are important hecause they are
a key source of such opportunities.
Community service and service-learning may especially help to
increase the engagement and motivation of students who are disadvantaged. For example, in the national evaluation of Learn and Serve America
programs, Brandeis University researchers found that service-learning's
academic and civic impact was greater for lower income, minority, and
more at-risk youth (Center for Human Resources, 1999). In addition, there
is evidence that hoth smaller class sizes and more "meaning-centered"
teaching and learning approaches appear to have greater educational
impact among minority and low-income students than among more
advantaged students (Clements & Mchityre, 2004; Kruger & Whitmore,
2001). Similarly, in a program to build five developmental assets, including service, among students in a highly violent California community,
Araque (2002) reported that, compared to controls, students in assetbuilding schools significantly increased their standardized test scores,
school engagement, honding to school, and other assets.
Despite the apparent academic benefits of service or service-learning,
these kinds of experiential education are uncommon. Only ahout half of
adolescents volunteer at least an hour a week (Benson et al., 1999) and
only ahout 30% of U.S. puhlic schools offer service-learning, with even
fewer high-poverty schools doing so (Kielsmeier, et al., 2004).
Research Questions
We examined a numher of questions that were suggested hy the
research reviewed here.
1. Perceived academic impacts of service-learning. How positively
do principals who lead urban, high-poverty, majority nonwhite
schools rate academic achievement, school engagement, and
attendance as positive impacts of service-learning, compared to
other principals?
2. Relation of service to achievement gaps. How do student poverty
status and participation in community service or service-learning
affect school grades, attendance, various developmental assets in
the Commitment to Learning asset category (e.g., engagement,
honding to school), and students' prosocial orientation? That is, is
there evidence that low-SES students who participate in community service or service-learning do better on those outcome variables than their low-SES peers who do not participate in service
or service-learning?
2006, Volume 29, No. 1
45
3. Effect of longer exposure to service-learning programs. Although
length of exposure to fhe service-learning experience is just one of
the elements that can influence the quality of a program, a brief
experience appears to have only limited potential for academic
impact, compared to experiences that last at least 10-15 hours (see
review in Scales, Blyth et al., 2000). Is there evidence that gaps in
academic outcomes between high- and low-SES students are smaller when low-SES students have longer exposure to service-learning?
Particularly in an era of high accountability for student achievement,
data showing such positive links would be useful in providing a resultsbased rationale for schools to use these forms of experiential education.
Method
Samples
National sample of U.S. principals. A total of 2,002 schools were
selected from the 2001-2002 Common Core of Data public school universe
file, stratified by instructional level, urbanicity, average class size, and
minority status. Survey responses were received from 1,799 schools (91%
response rate). Forty-seven percent of participating principals were from
elementary schools, 26% middle schools, and 28% high schools (more
details on the study are found in Scales & Roehlkepartain, 2004b, and
Kielsmeier et al., 2004).
Aggregate sample of U.S middle and high school students. A large,
diverse, but not nationally representative sample of more than 217,000
6th-12th grade students was aggregated from more than 300 U.S. communities that administered the Search Institute Profiles of Student Life:
Attitudes and Behavior survey (PSL-AB) in the 1999-2000 school year. The
PSL-AB measures 40 developmental assets, 10 risk behavior patterns, 5
developmental deficits, and 8 thriving behaviors (further details on the survey and the sample are found in Leffert et al., 1998; Scales, Benson, Leffert,
& Blyth, 2000; and Sesma & Roehlkepartain, 2003). The sample was
weighted by race/ethnicity and urbanicity proportions of the 2000 Census.
Sample of middle and high school students in Colorado Springs. A
racially and economically diverse sample of 5,136 6th-12th grade students from Colorado Springs, Colorado, took both the PSL-AB survey and
Search Institute's Youth Supplement Survey in February 1999. The Youth
Supplement Survey measures selected developmental assets and thriving
behaviors in greater depth and with more internal consistency reliability
than the PSL-AB (further details are found in Scales, Leffert, & Vraa, 2003).
Measures
Perceived impact of service-learning. Principals were asked whether
46
Journal of Experiential Education
they thought the impact of service-learning on 10 different academic,
social development, school climate, and community relations outcomes
was "very positive," "somewhat positive," or whether service-learning
had "little or no positive impact." For these analyses, we calculated the
mean "positive impact" score for each of the individual outcomes.
Service to others. Service to others is measured hy a single item asking students how much in an average week they do formal volunteering
without pay to help others ("such as helping out at a hospital, day care
center, food shelf, youth program..."). Response choices were 0, 1, 2, 3-5,
6-10, or 11 or more hours per average week. Students reporting at least 1
hour per week were counted as contrihuting service. This simple hinary
cutoff has heen found to significantly differentiate students with higher
and lower levels of risk hehaviors such as suhstance use, and thriving indicators such as school grades (Benson, Scales, Leffert, & Roehlkepartain,
1999; Scales & Roehlkepartain, 2004a)
Duration of service-learning. The Colorado Springs sample was asked
about how long they had participated in "community service or servicelearning" in the last school year. Response choices were not at all, up to a
few hours, a few days, a few weeks, a few months, and most of the year.
Because the majority of students reported no service-learning (see Results),
and hecause at least 10—15 hours of exposure appears to he needed for various effects to be observed (Scales, Blyth, et al., 2000), students with a few
weeks or more of service-learning were considered to have relatively
"high" levels of exposure, and students with anj^hing less but greater than
none were considered to have "some" exposure to service-learning.
School poverty level. In analyses using the national principals'
study, three school-level poverty groups were created by dividing the
sample of schools into thirds on the basis of the proportion of students eligible for the free or reduced-price federal lunch program. The top third
were considered "high-poverty schools." This division resulted in schools
with 56% or more students being eligible for free or reduced-price lunches being high-poverty schools; those with 27% to 55% being mediumpoverty schools; and those with 0% to 26% of students eligible being lowpoverty schools.
Student SES. A composite SES proxy was created from student selfreports on two items, level of mother's education and whether the student
lives in a single-parent family, common proxies for SES (Theokas et al.,
2005). Students living in a single-parent family and whose mother completed only high school or less were considered low-SES. About 11% of
the large aggregate 1999-2000 school year sample, or 21,883 students, met
the definition of being in a low-SES family.
School grades. Students reported whether they got mostly As, about
half As and Bs, mostly Bs, about half Bs and half Cs, mostly Cs, about half
2006, Volume 29, No. 1
47
Cs and half Ds, mostly Ds, or mostly below Ds. Self-report grades have a
high {approximately 0.75) correlation with actual grades (Leffert et al.,
1998). Students who got mostly As were considered to have high grades.
In previous uses of this criterion, students' level of service experience has
been found to differentiate those with mostly As from those with lower
averages (Benson et al., 1999; Scales & Roehlkepartain, 2004a).
Attendance. Students were asked how many days of school they
skipped or "ditched" during the last month. Response choices were 0, 1,
2, 3, 4-5, 6-10, or 11 or more days. Because relatively few students (less
than 20%) acknowledge missing any days of school (Benson et al., 1999),
a high cutoff was set and students had to report skipping no days of school
to be considered to have high attendance.
Commitment to Learning assets. The PSL-AB meastires five assets in
the category of Commitment to Learning. Three items measure achievement motivation (alpha reliability of 0.64), and four items measiu-e school
engagement (alpha reliability of 0.63). Bonding to school, homework
(average hours per school night), and reading for pleasure (hours per
week) each are meastired with one item. To "have" these assets, students
needed to average a response of Agree or Strongly Agree to the component
questions for achievement motivation, school engagement, and bonding to
school, and to say they averaged an hour per school night of homework,
and three hours per week of reading for fun. In a longitudinal study, students' level of these measures of the Commitment to Learning assets in
middle school were found to have a significant positive relationship with
their grade point average three years later in high school (Roehlkepartain,
Benson, & Sesma, 2003). Three of the measures—achievement motivation,
school engagement, and bonding to school—have been found across various Search Institute studies to have significant bivariate and multivariate
relations with at least two of five academic outcomes meastires: grades;
attendance; sense of belonging; motivation; and academic self-efficacy
(Scales & Roberts, 2005). All five assets have been found to be related to
varied academic outcomes in a considerable body of other researchers'
studies (reviewed in Scales & Leffert, 2004).
To explore the research questions, a series of ANOVAs were run,
with Tukey post-hoc tests performed when overall F tests were significant
at p < .05. Student groups were compared on grades, attendance, and the
five Commitment to Learning assets according to the SES, school poverty,
and service/service-learning expostne groups described above. Of particular interest were groups formed by crossing high and low levels of SES
with high and low levels of service exposure.
Journal of Experiential Education
Results
Perceived Academic Impacts of Service-Learning
Table 1 shows that principals of urban, high-poverty, or majority
nonwhite schools are significantly more likely than other principals to
judge service-learning's impact on attendance, school engagement, and
academic achievement to be "very positive." Moreover, principals of
schools that have all three characteristics—urban, high poverty, and
majority nonwhite student population—also are more likely than all other
principals to consider service-learning to have a very positive impact on
attendance, school engagement, and academic achievement. In this
national sample, there were no majority white schools that were also lowSES. However, within low-SES schools, principals of majority AfricanAmerican schools were significantly more positive than principals of lowSES and majority Hispanic schools about service-learning's impact on academic achievement (mean 1.5 versus 1.7, F{2, 21597) = 175.04, p < .0001,
with lower means indicating more positive perceptions).
Relation of Service to Achievement Gaps
As expected, low-SES students appear slightly less likely than other
students to serve; 47% of low-SES students contribute an hour or more per
week of community service, versus 52% for other students. This difference also is reflected in a lower mean number of hours of service for lowSES students [F[l, 212658) = 15.87, p < .0001). Likewise as expected (not
shown—available from authors), low-SES students and those not doing
community service had significantly less experience of each of the five
Commitment to Learning assets (achievement motivation, school engagement, bonding to school, homework, and reading for pleasure); they also
skipped more school days, and had lower grades than either high-SES students or students who did community service.
Of considerably more interest, however, is the question, does contributing community service seem to be related to lessened achievement
gaps between low-SES and high-SES students? Table 2 shows that the
answer appears to be yes. We divided the sample into four groups: lowSES students with service; low-SES students without service; high-SES
students with service; and high-SES students without service. We then
compared the four groups on the means of the five Commitment to
Learning assets, attendance, and grades.
Table 2 shows that on each of the five Commitment to Learning
assets high-SES students with service had the highest means. However,
low-SES students with service had more of these assets than low-SES students without service. They also had higher or equivalent levels ofthe academic assets than did high-SES students without service. In other words.
2006, Volume 29, No. 1
49
CO
^1
rH
q
CD
r-l in
t>. rH i n
^ CO i n
in
CM C
O
O5
in 05
II
I^
1
in
in
CO
r-l
r-ifi:
i
O5 CO
in
i
Q
o
CO
CO CO
"io
(N
*
CD
CD CO
rH
II
II
CO"
O5
CO
o"
*
rH
CO
in
CO
rH
CO
II
II
0?
"o
in
in
CO
CM
in
in CD
o
rH
CM
CO
in
CO
rH'
r^ GT
OCJ
o
CO
CM
CD
r—~
CM
CO
CO
in
rH
Ln
*
CJI
rH
CO
CD
CD
II
O5
*
in
•K
O5
^^
CO
LO
II
v*
CD 00 rH_
rH rH' k .
'^
CSl
in
in
CM
"^JH
.-
LO
CO
in
O5
CO
CO
CO
rH tC"
CM
I
in
in
CO
CM
.,
in
CD r H
CM
*
*
CD CO
^
CD
in
in CD
o
r^te:
I
o
CM
CD
CM
CO
u
I
C/5
CD
CO
CO
00
O
CO CM
rH
^
^
CO CM
CO CO
O
in
O
O5
CO rfl
CO CO
o
o
q
-a
*
O5
CD
CD
CO
CD
rH
<N
'^
"
O5 ^
VI
CM
CO
C^]
*
*
CD CM
C^ O5
CM CM
t^ 05
CD
O5 rH
CJ
S
I
o
CO
I
g
o
CO
CD
O
CD"
.rf
rl
--H
s 1o
cd cd
Itt
03
CJ
Cd
O
O
Cd
DO
. r -
03
c
iffere]
>,
II
I—I
II
CJ
CO
CO
in
_03
o
choc
tur
chal lenge
I
03
ca
03
03
O
DC
nn .sPmi
*
50
Journal of Experiential Education
03
lO
03
CO
rH
CO
CD
CO
03
U
C/3 ?
CJ
331"
hJ
+
I
CO
CO
T3
CD
CO
XI
XI
-a
CO
O3
Csl
CO
CSI
s
O)
CD
in
'> ui
—" "O
O
3 o
S "
ti
CO
CD
? '3ou
CO
lO
CO
o
o
CO
IO
CD
CD
3 B
CO
IO
03
U
o
CO
03
Cd
O>
tv
lO
rH
".^
o
U
CM
IO
in
CD
CM
CD
o
IO
IO
in
CM
03
O
Xi
VI
II
^
S
03
W
03
SI
03 >-.
O
O
XI
Xi
tx
CO
Xi
CO
rH
O
I
I
CJ
O
rH
fx
tv
CJ
^
CO
CO
Cvl
CD
03
I
CO
CJ
9)
cd
K +
CO
lO
O
lO
CO
CM
rH
cQ
CO
CO
t
CO
CO
lO
CO
CD
O
.,
CO 0 3
UJ
rj
0)
03
M
o
u
^I
^
3
o
e 2 o
CO
O 03
o ^
I
.2 s:
^
^ 3 CO
CO S
to
03
9)
si
2 d S
•3
O
CO
CSI
in
'I
•^
2 -^03
a^'^
03
C« 4)
03
II
> _o
Is
-^
•"
tH XI
O fl
CO _
S
g 8 "^
o
CO J 3
§ J -a
03
U
«
03
a
a
tH
ft"! c/-a
o
3
O
u
O 03
o
o
I
uo
o
o
q
q
vi
VI
Q,
n.
co"
a en
o q
tv.
CO
CO
'•5 «>
CO (D
> rH
•-O Cvl
ijl
.^
CO
c ^
S II
is
03 CD
icM
- C - ]
Sco" o(
HC/3
CO
u [
C/D;
o
o
q
VI
o
o
q
o
o
q
VI
o
o
q
VI
o..
VI
n.
CO
q
rH
ccj
, CD
CO
I Cvl
lO
I CSI
VI
a,
o"
q
CD"
CO
CD
o
o
q
I
'3
o lo
%^ ^
g
§ XI( 03
03
CO
CO
11
'!
2
I i
lO
CSI
IE
lO
CO
-H
Sco"
ES
II
03
'>
cq
^ ^
^^
O IO
..
"^ O3 .^
DOrH
S 1^
*- 53 o
T 3 CSI
^
o
t n CS)
QJ
"S co"
m co" ' ^ co"
SET
Bfe' Str
DiCl <C1
OCl
g &
?
03
CO
>
|1
C/D - ^
II +
03
^
'>
^
sag
C/D
DO ^
00
2006, Volume 29, No. 1
51
service of only one hour per week among lower income students was
related to a significantly smaller gap in achievement-related developmental assets between higher and lower income students.
Similarly, both groups of low-SES students skipped more days of
school and had lower grades than either set of high-SES students. But Table
2 shows that the low-SES students who contributed community service
reported significantly fewer missed school days and significantly higher
grades than other low-SES students who did not participate in service.
For example, only 8% of low-SES students without service reported
getting "mostly As," whereas 11% of low-SES students who did service
had high grades, a considerable 38% difference within low-SES students
by whether or not they served. Thus, service seems to have a positive relation to a smaller school success gap between students from lower and
higher income backgrounds.
Effects of Longer Exposure to Service-Learning Programs
In the Colorado Springs sample, only 18% of students had at least a
few weeks of service-learning, compared with 21% who had a few hours
to a few days. The majority of the sample—62% of the students—reported
no service-learning at all. Table 3 shows that students who reported a "few
weeks" or more of service-learning had more positive results on all the
Commitment to Learning assets than all other students. Students with a
few weeks of service-learning also had better attendance and grades than
students with no service-learning, but not significantly better than students with a few hours of service-learning. However, the grades trend was
as hypothesized, with only 25% of students without service-learning
reporting "mostly As," compared to 35% for students with a few hours or
days of service-learning, and 42% for students with a few weeks or more
of service-learning.
Finally, we examined levels of the academic success indicators by
those iom SES by exposure to service-learning groups. Results for the
Colorado Springs sample were somewhat more mixed than for the large
aggregate Search Institute sample discussed above, but relatively small
sample sizes (-30 or fewer students) in a number of analysis cells likely
influenced some of these results. As expected. Table 4 shows that high-SES
students, especially those with high levels of service-learning, were generally higher in their academic success outcomes than low-SES students. But
as for the previous results, the gap was smaller between high-SES students
and low-SES students who had longer exposure to service-learning.
Service-learning appeared to be associated with smaller gaps for
attendance, achievement motivation, school engagement, reading for
pleasure, and especially for bonding to school, but not for homework or
52
)ournal of Experiential Education
cfl
I
o
CO
CM
in
CO
CM
o
O
S^
rH
co"
^
to
c/i
>,
CO
• ' ^
U
rH
00
O
CJ
CM
CM
rH
i!
X:
CM
CO
o
CO
(_,
CD
53 CO
-.1:
U
CD
rH
rH
•aa
Q>
CO _g
u
^
3
CO
O)
So
CO
CM
IN
CO
CO
CO
IV
in
tv
tv
iv
CM
in
CO
CO
CO
q^
CQ
o
rH
o
o
o
0) CO
CJ ( «
X
O
<^
CO
-§
°- g
vi 2;
B S u
o
CO
CO
*
O
CO
in
CO
CO
CO
tv
CO
c^a
• ^
CO
CO
II
2
^
^
'—!
cn
in
•5
o
o^
CO
1=1 a
II
* j
a
C
CO
C30
a
£ '3
11
-§1 * I
T3
-O
CO
^
_CJ
CD
u
O
CO
CO
CO
CO
c^^
o
CM
CO
cd
CO
rH
O
rH
CO
CO
o
o
o
o
o
o
o
o
°
VI
VI
VI
VI
n.
Q.,
q
«
CO
O CO
G CO
53 CO
S CO
Is
IN
00.—>
C CO
(D CN
x"
q
<^ °^
rH
CO
3 05^
w •*
CO
II
- 2 CM
C^ „
00 f^
in
CO
CO
>
O
O
VI
CT)
-d
CO
T3 m
CO
CO
CO
C^l
a B
£P
SI
o f^
o
I
CD
a3
OH
O
DO
^
o
q
•Si;
VI
n.
•S
.3
in
CM
CO
q:
C3
in
II
TO
0
cu
u
II
o
m
«
IV
in
in
II
g cs
So
.S
ScJ '§cJ
o in
q
cd
O
c-g
o
o
q
CO
01)
r^
idai
500
0
497
a,
;s
CO
CO
ing
B
U3
fO
C-)
CO
.13
3
C
a-s
8 S
"3 ^
09
O
O
CO
rH
to o>
2;
T3
CO CM"
"c3
5
ES
o
l§ l
oi b
^
ft _ d
£ 5
DO
2006, Volume 29, No. 1
53
self-reported grades. The trend for grades was as hypothesized, however,
with 22% of low-SES/high service-learning students having high grades
versus only 11% of low-SES/low service-learning students.
Using the "bonding to school" outcome as an example, low-SES students in Colorado Springs who also had deeper exposure to service-learning had the second-highest proportion (63%) of feeling connected to
school, bested only by high-SES students with equally lengthy servicelearning exposure (71%).
Discussion
Our results suggest that service-learning may be an especially valued strategy for student engagement and achievement for principals in
schools that are urban, or majority nonwhite, or high poverty. Moreover,
principals' positive beliefs about service-learning are supported by the
positive association between students' involvement in community service
and their school success. In particular, our results offer suggestive evidence that community service appears to be related to a smaller achievement gap between students from lower and higher income backgrounds.
Moreover, experiencing service-learning for at least a few weeks appears
to be related to a smaller gap in most academic outcomes between low and
high-SES students. Although we cannot establish causality with this
study, these results suggest a link between service or service-learning and
academic success and are consistent with: (a) the theory of experiential
learning; (b) a growing amount of largely cross-sectional work; and (c)
results from longitudinal research (e.g.. Scales & Roehlkepartain, 2004a)
that provide stronger direct evidence of causality. These results would
appear to lend support to the call recently issued by the National
Commission on Service-Learning to integrate service-learning throughout
the K-12 curriculum and make it a "significant part of district reform agendas" (Glenn & Edelman, 2005, p. 40).
Of course, even high-quality service-learning is just one of many elements of the school community necessary to have a widespread and deep
impact on student achievement (National Research Council and Institute of
Medicine, 2004). But principals should evaluate higher quality servicelearning even more positively, since research has shown that higher quality
of service-learning has been associated with higher levels of academic
achievement and other positive impacts (Billig, 2004; Scales, Blyth,
Berkas, & Kielsmeier, 2000). Such elements as integrating service with the
curriculum, meeting genuine community needs, and providing opportunities for students to think critically, reflect on their service, and have a
voice in decision-making are among the "essential elements" of servicelearning described in a consensus statement from 13 major organizations
Journal of Experiential Education
54
o
O g
in
X)
CM
CO
in
00
-d
ear nin
a;
CO
o
0
0
ys.
e35
en
o
CM
CM
05
in
CM
CM
IN
^^
i
u
05
CO
6 CO
CM
u
las
o
^ en q
Q o S
0
_00 0
ce.
CO p.
•
a
3 E
CO
CD
CD
eo
CA
W
sc s
-d
CO
in
0
O S
B
o
o
to o
CJ
Z
eo
II
o
CO
(A
o
o
JD
CO
O
O5
CO
CO
CO
O
CO
0
a
en
CO
VI
ft
0
to
6 XI
g
o
O
a:
u
"Oo5 5
a
05
in
C/3
CO
CD
CD
CM
^ E
I
o u
CO
CD
t
rH
cO
CD
O
rH
u
^
a
JD
CO
mo
i.b'
-d
u
0
CM
CO
IN
rH
eo
CO
'd
0
e
in
in
IN
5
CO
CM
in
CO
o o
o
CO
O5
6
X
q
q
VI
VI
VI
_
VI
Q,
o
C/3
C/3
•*
-^co
CO II
ooA
a ino
0
_
o
t^
•*
^e
en II
O .—.
*H rH
^2
d
CO
CM
CO
IN'
CO
II
O
O
o
o
q
q
VI
o
o
q
VI
VI
0
Cl.
»-
o"
a>
in
^
„
3 C>O*"
en
CO
2 CD
-H
CD
in
rH
^11
II
"I o? .Sin" l ' ^
5 § a'B -d 2 «
0 ^
.3 ^
6 co" "S CO
C
1
o
-2
CO S
03 crT " ^
ized
that'
ngsu
o
o
q
s
en
XI
"d
C3
o
en
0
iS
.0
0
11
eans w ere
siti ve peri ormanc
Means thdi
•vice-le ing =
03
o
o
O
O
1
0
ft
o
o
0
'S
ft
0
ei)
^00 o
'eo CO
iabl
rscri
CO
O5
CO
CO
CO
II
op
CJ
CO
1X)
in
O
ft
are
rH
_O
a
_o X I
CO
I
6
dar
ce.
fieant
-d
u
u
0
o P
Q
Be
CO
CO
'o a •ti
u
CO
CO
CD
CM
CM
CO
XI
•3
o
CO
CO
CD
rH
u
0
en"
jviati
wer I
iffere
earni
•
CO
'o
indi tes
V)
rH
CM
o
o
exc
•5)CA)
ft
.05.
one
4^
1
CO
>
so
•4...
ft
o o
II
ej
I
0
en
p^
x"
co"
00
X
2006, Volume 29, No. 1
55
funded by the Corporation for National Service (National Youth
Leadership Council, 2004).
Our data do not directly suggest the reasons for the positive relations
between service/service-leaming and both academic achievement and
principals' perceptions of academic impact. Principals in more resourcerich schools appear not to give the same academic importance to servicelearning or see as much academic impact, perhaps because they have so
many resources other than service-learning to deploy. It may also be that
principals of high-poverty schools are attracted to service-learning
because they see it as a creative way to identify, connect with, and leverage community resources. Principals of low-SES schools may also be seeing first-hand the connection between service and service-learning, and
engagement and achievement, evident in our student self-reports.
Community service and service-learning may be related to academic success because they provide young people with two key resources: A
feeling of usefulness and being valued, and a way of tangibly demonstrating to students the utility in the "real world" of what they learn in school.
Indeed, Search Institute has conceptualized "service to others" as a developmental asset in the category of "empowerment" (Benson et al., 1998).
Because a sense of empowerment and playing useful roles in society may
be in shorter supply for low-SES students, service and service-learning
may have a greater influence among socioeconomically disadvantaged
students on the constellation of factors that affect academic success.
Moreover, service-learning may provide students with multiple sources of
both instructive feedback and high expectations. That is, with this kind of
experiential education, more individuals function in the role of "teacher"
to a given student (service-recipients, adult volunteers, as well as peers
and the official school teacher[s]), and the relationship between these
"teachers" and the student are generally positive, characterized by both
caring and high expectations. In support of such reasoning, Follman and
Muldoon (1997) reported that students' school attendance increased on
days when they had service-learning, with the interpretation that on those
days the students realized that to skip school would be to disappoint the
service recipients with whom they feel connected.
Limitations and Conclusions
All of the results in this study are correlational and not longitudinal;
thus, cause and effect relations among the variables cannot be established.
For example, perhaps regardless of SES, students who already are more
academically motivated are more likely to participate in service-learning,
more so than service-learning contributing to academic motivation.
Moreover, perhaps service-learning benefits low-SES students, but benefits
Journal of Experiential Education
at least as much or even more the high-SES students who typically already
are more academically advantaged. Ceci and Papierno (2005) have recently discussed the ethical, political, psychological, and economic dilemmas
produced when policy or program measures intended to close "have" and
"have not" gaps actually maintain or increase them, even though in
ahsolute terms benefiting the "have-nots." In addition, all our data are selfreports, either from principals or students. Common method variance thus
may have inflated the relations reported here. We also have no direct measure of the quality of community service or service-learning, merely measures of their frequency. Variations in the quality of the service-learning
programs in which students participated likely affected some results.
Additional studies, especially with longitudinal designs and with
deeper measures of service-learning quality, are needed to confirm and
more fully explain the reasons for these findings. Although only suggestive, these results, which were obtained across three different data sets,
rather consistently point to the promising possibility that community
service and service-learning programs may contribute to both of the key
achievement goals of American elementary and secondary education:
higher achievement and equity of achievement across student groups.
References
Araque, J. C. (2002). The A.P.I, study: Academics and prevention inseparable.
Costa Mesa, CA: Orange County Department of Education.
Beauvais, C, & Jenson, J. (2003). The well-being of children: Are there "neighbourhood" effects? Ottawa, Ontario: Canadian Policy Research Networks.
Benson, P. L., Leffert, N., Scales, P. C, & Blyth, D. A. (1998). Beyond the "village"
rhetoric: Creating healthy communities for children and adolescents. Applied
Developmental Science, 2, 138-159.
Benson, P L., Scales, PC, Leffert, N., & Roehlkepartain, E.C. (1999). ^/ragiVe
foundation: The state of developmental assets among American youth.
Minneapolis, MN: Search Institute.
Benson, P. L., Scales, P. C, & Mannes, M. (2004). Developmentally attentive communities. In C. B. Fisher and R. M. Lerner (Eds.), Encyclopedia of applied
developmental science (pp. 357-359). Thousand Oaks, CA: Sage
Publications.
Betts, J. R., Zau, A. C, & Rice, L. A. (2003). Determinants of student achievement:
New evidence from San Diego. San Francisco: Puhlic Policy Institute of
California.
Billig, S. H. (2004). Heads, hearts, hands: The research on K-12 service-learning.
In Growing to greatness 2004: The state of service-learning project (pp.
12-25). St. Paul, MN: National Youth Leadership Council.
2006, Volume 29, No. 1
57
Billig, S. H. (2006). Service-learning as effective instruction; Lessons from
research on teaching and learning. In Growing to greatness 2006: The state of
service-ieamingproject (pp. 25-32). St. Paul: National Youth Leadership
Council.
Blyth, D. A., Saito, R., & Berkas, T. (1997). A quantitative study ofthe impact
of service-learning programs. In A. S. Waterman (Ed.), Service-ieaming:
applications from ttie research (pp. 39-56). Mahwah, NJ: Lawrence Erlbaum.
Borman, G. D., Hewes, G. M., Overman, L. T., & Brown, S. (2002).
Comprehensive schooi reform and student achievement: A meta-anaiysis.
Baltimore: Center for Research on the Education of Students Placed At Risk,
Johns Hopkins University.
Carnegie Council on Adolescent Development. (1989). Turning points: Preparing
American youth for the 21st century. Washington DC: Carnegie Council on
Adolescent Development.
Ceci, S. J., & Papierno, P. B. (2005). The rhetoric and reality of gap closing: When
the "have-nots" gain hut the "haves" gain even more. American Psychoiogist,
60, 149-160.
Center for Human Resources. (1999). Summary report: National evaluation of
Learn and Serve America. Waltham, MA: Brandeis University, Center for
Human Resources.
Chen, C, and Stevenson, H. W. (1995). Motivation and mathematics achievement. A comparative study of Asian-American, Caucasian-American, and
East Asian high school students. Child Development, 66,1215-1234.
Clements, S. K., & Mclntyre, E. (2004). Academic achievement gaps: The problem in national and state context. Lexington, KY: Pritchard Committee for
Academic Excellence.
Council of Great City Schools. (1999). Closing the achievement gaps in iu"han
schools: A siKvey of academic progress and promising practices in the great
city schools. Washington, DC: Council of Great City Schools.
Dearing, E., McCartney, K., & Taylor, B. A. (2001). Change in family income-toneeds matters more for children with less. Child Deveiopment 72,
1779-1793.
Duckenfield, M., & Drew, S. (2006). Measure what matters and no child will be
left behind. In Crowing to greatness 2006: The state of service learning project (pp. 33-39). St. Paul: National Youth Leadership Council.
Eamon, M. K. (2002). Effects of poverty on mathematics and reading achievement
of yovmg adolescents. Journal of Early Adolescence, 22, 49-74.
Eccles, J. S., Midgley, C, Wigfield, A., Miller Buchanan, C, Reuman, D.,
Flanagan, C, & Maclver, D. (1993). Development during adolescence: The
impact of stage-environment fit on young adolescents' experiences in schools
and in families. American Psychologist, 48, 90-101.
Education Trust. (1999). Dispelling the myth: High poverty schools exceeding
expectations. Washington, DC: Education Trust.
58
Journal of Experiential Education
Evans, G. W. (2004). The environment of childhood poverty. American
Psychologist, 59, 77-92.
Fellner, R. D., Jackson, A. W., Kasak, D., Mulhall, P., Brand, S., & Flowers, N.
(1997). The impact of school reform for the middle years. Phi Delta Kappan,
78, 528-532, 541-550.
Follman, J., & Muldoon, K. (1997). Florida Leam and Serve 1995-1996: What
were the outcomes? NASSP Bulletin (National Association of Secondary
School Principals), 81. 29-36.
Glemi, J., & Edelman, M. W. (2005). Educating each generation for democracy.
Education Week, 24(23), 40.
Greenberg, M. T., Weissberg, R. P., O'Brien, M. U., Zins, J. E., Fredericks, L.,
Resnik, H., & Elias, M. J. (2003). Enhancing school-based prevention and
youth development through coordinated social, emotional, and academic
learning. American Psychologist, 58, 466-474.
Hanson, T. L., Austin, G., & Lee-Bayha, J. (2003). Student health risks, resilience,
and academic performance. Los Alamitos, GA: WestEd.
Hirsh-Pasek, K., & Golinkoff, R. M. (2003). Einstein never used flash cards: How
cur children really learn—and why they need to play more and memorize
less. New York: Rodale.
Johnson, D. H. (Ed.). (1997). Minorities and girls in school: Effects on achievement and performance. Thousand Oaks, GA: Sage Publications.
Kielsmeier, J. G., Scales, P.G., Roehlkepartain, E. G., & Neal, M. (2004).
Preliminary findings: Gommunity service and service-learning in public
schools. In Growing to greatness 2004: The state of service-learning project
(pp. 6-11). St. Paul, MN: National Youth Leadership Goimcil.
Krueger, A. B., & Whitmore, D. M. (2001). Would smaller classes help close the
black-white achievement gap? Princeton University Working Paper #451.
Princeton, NJ: Princeton University. (Available from
www.irs.princeton.edu/puhs/pdfs.451.pdf).
Lee, V. E., & Smith, J. B. (1993). Effects of school restructuring on the achievement and engagement of middle-grade students. Sociology of Education, 66,
164-187.
Lee, V. E., Smith, J. B., & Groninger, R.D. (1995). Another look at high school
restructuring. Issues in Restructuring Schools. Issue Report No. 9. Madison:
Wisconsin Genter for Educational Research.
Leffert, N., Benson, P. L., Scales, P. G., Sharma, A., Drake, D., & Blyth, D. A.
(1998). Developmental assets: Measurement and prediction of risk behaviors
among adolescents. Applied Developmental Science, 2, 209-230.
Lipsitz, J., Jackson, A. W., & Austin, L. M. (1997). What works in middle-grades
school reform. Phi Delta Kappan, 78, 517-519.
Louis, K. S., Toole, J., & Hargreaves, A. (1999). Rethinking school improvement.
In J. Murphy and K. S. Louis, Eds., Handbook of research in educational
administration (pp. 251-276). San Francisco: Jossey-Bass.
Mahoney, J. L., & Gairns R. R. (1997). Do extracurricular activities protect against
early school dropout? Developmental Psychology, 33, 241-253.
2006, Volume 29, No. 1
59
Massachusetts Department of Education. (2005). Dropouts in Massachusetts
Public Schools: 2003-2004. Maiden: Massachusetts Department of Education.
Mayer, D. P., Mullens, J. E., Moore, M. T., & Ralph, J. (2000). Monitoring School
Quality: An Indicators Report. Washington, DC: U.S. Department of
Education, National Center for Education Statistics.
National Research Council and Institute of Medicine. (2004). Engaging schools:
Fostering high school students' motivation to learn. Washington, DC: National
Academies Press.
National Youth Leadership Council. (2004). Essential elements of servicelearning. In Growing to greatness 2004: The state of service-learning project
(pp. 90-91). St. Paul, MN: National Youth Leadership Council.
Newmann, F. M., Lopez, G., & Bryk, A. S. (1998). The quality of intellectual work
in Chicago schools: A baseline report. Chicago: Consortium on Chicago
School Research.
Newmann, F. M., Wehlage, G. G., & Lamborn, S. D. (1992). The significance and
sources of student engagement. In F. M. Newmann, (Ed.), Student engagement and achievement in American secondary schools (pp. 11-39). New
York: Columbia University Teachers College.
Nichols, S. L., Glass, G. V. and Berliner, D. C. (2005). High-stakes testing and student achievement: Problems for the No Child Left Behind Act. Tempe, AZ:
Education Policy Studies Laboratory.
Oakes, J., Quartz, K. H., Ryan, S., & Lipton, M. (2000, Feb. 23). Civic virtue and
the reform mill. Education Week, 19(24), 68, 43.
Olson, L. (2000). Worries of a standards 'backlash' grow. Education Week 29(30),
1, 12-13.
Patrick, H., Hicks, L., & Ryan, A. M. (1997). Relations of perceived social efficacy
and social goal pursuit to self-efficacy for academic work. Journal of Early
Adolescence, 17, 109-128.
Roehlkepartain, E. C, Benson, R L., & Sesma, A. (2003). Signs of progress in putting children first: Developmental assets among youth in St. Louis Park,
1997-2001. Minneapolis, MN: Search Institute and Children First Initiative.
Scales, P. C, Benson, P. L., Leffert, N., & Blyth, D. A. (2000). Contribution of
developmental assets to the prediction of thriving among adolescents.
Applied Developmental Science, 4, 27-46.
Scales, P. C, Benson, P. L., Roehlkepartain, E. C, Sesma, A., Jr., & van Dulmen,
M. (in press). The role of developmental assets in predicting academic
achievement: A longitudinal study. Journal of Adolescence.
Scales, P C, Blyth, D. A., Berkas, T. H., & Kielsmeier, J. C. (2000). The effects of
service-learning on middle school students' social responsibility and academic success. Journal of Early Adolescence, 20, 322-358.
Scales, P C, Foster, K. C, Mannes, M., Horst, M. A., Pinto, K. C, & Rutherford,
A. (2005). School-business partnerships, developmental assets, and positive
outcomes among urban high school students: A mixed-methods study. Urban
Education, 40, 144-189.
Scales, P. C, & Leffert, N. (2004). Developmental assets: A synthesis of the scientific research on adolescent development (2nd ed.). Minneapolis: Search
Institute.
60
Journal of Experiential Education
Scales, P. C, Leffert, N., & Vraa, R. (2003). The relation of community developmental attentiveness to adolescent health. American fournal of Health
Behavior, 27 (Supp. 1), S22-S34.
Scales, P. C, & Roberts, C. (2005). Developmental assets and ocademic ochievement. Paper presented at Annual Health Communities. Healthy Youth
Conference, Search Institute, Dallas, TX, November 2005. Minneapolis, MN:
Search Institute.
Scales, P. C, & Roehlkepartain, E. C. (2004a). Service to others: A "gateway asset"
for school success and healthy development. In Growing to greatness 2004:
The state of service-learning project (pp. 26-32). St. Paul, MN: National
Youth Leadership Council.
Scales, P. C, & Roehlkepartain, E. C. (2004b). Community service and servicelearning in U.S. public schools, 2004: Findings from a national survey. St.
Paul, MN: National Youth Leadership Council; and Minneapolis, MN: Search
Institute.
Sesma, A., Jr., & Roehlkepartain, E. C. (2003). Unique strengths, shared strengths:
Developmental assets among youth of color. Search Institute Insights and
Evidence, 1{2), 1-13.
Sheckley, B. G., & Keeton, M. T. (1997). Service learning: A theoretical model. In
J. Schine (Ed.), Service homing: Ninety-sixth yearbook of the Notional
Society for the Study of Education (pp. 32-55). Chicago: University of
Chicago Press.
Theokas, C, Almerigi, J. B., Lerner, R. M., Dowling, E. M., Benson, P. L., Scales,
P. C, & von Eye, A. (2005). Conceptualizing and modeling individual and
ecological asset components of thriving in early adolescence. Journal of Early
Adolescence, 25, 113-143.
U. S. Department of Education, National Center for Education Statistics. (2000).
Condition of Education 2000. Washington, DC: U.S. Government Printing
Office (NCES Document 2000-602).
U. S. Department of Education, National Center for Education Statistics. (2004).
The condition of education 2004 (Event dropout rates by family income,
1972-2001). Washington, DC: National Center for Education Statistics.
Wang, M. C. (1990). Variables important in learning: A meta-review of reviews of
the research literature. Philadelphia: Temple University, Center for Research
in Human Development and Education (ERIC #405691).
Wenglinsky, H. (2002). How schools matter: The link between teacher classroom
practices and student academic performance. Education Policy Analysis
Archives, 10, no.l2, http://epaa.asu.edu/epaa/vl0nl2/ (accessed May 15,
2005).
Zeldin, S. (2004). Preventing youth violence through the promotion of community engagement and membership. Journal of Community Psychology, 32,
623-641.