Does School Connectedness Mediate the

Relationship between Teacher Support and

Depressive Symptoms?
Hilary D. Joyce

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

Strong teacher support is associated with and predictive of fewer depressive symptoms
among youths. However, school connectedness may play a role in explaining this relation-
ship. Using the National Longitudinal Study of Adolescent to Adult Health, this study
examined the mediating role of school connectedness on the relationship between teacher
support and depressive symptoms in adolescents. The findings showed that school connect-
edness partially mediated the relationship between depressive symptoms and both feeling
cared about by teachers and getting along with teachers. The study findings offer important
implications for school social workers who may play an important role in strengthening
student–teacher relationships and promoting strategies that enhance student connection to
schools.
KEY WORDS: depressive symptoms; school connectedness; student–teacher relationships; teacher
support

B ronfenbrenner’s ecological perspective em-

phasizes the powerful influence the social
environment exerts on youth development
and provides a framework to understand the system-
level factors (for example, school, family) that affect
and empirical literature points to a number of factors
that affect youth vulnerability to depression, including
individual factors such as temperament and cognitions
(Dallaire et al., 2008; Gilbert, 2004) and environmen-
tal factors including relationships at school (Gilbert,
the emotional health of youths, including depression 2004). Using the ecological perspective as a lens, this
(Bronfenbrenner, 1979; Hess & Shultz, 2008). A article focuses on the relationship between the school
recent analysis of national trends in the prevalence of environment and depression in adolescents and offers
depression in adolescents found that the rates of implications for social work practice in schools.
depression increased from 8.7 percent in 2005 to
11.3 percent in 2014 (Mojtabai, Olfson, & Han, CONCEPTUALIZATION OF ADOLESCENT
2016). The upward trend seems to be continuing. In DEPRESSION
2016, 12.8 percent of youths ages 12 through 17 had Bronfenbrenner’s ecological perspective asserts that
a major depressive episode (MDE) during the past individual systems (for example, biological and psy-
year and 9 percent experienced related impairment chological) act in combination with environmental
such as significant problems completing school work systems (for example, family, school) to influence
(Substance Abuse and Mental Health Services Admin- youth development (Bronfenbrenner & Morris,
istration [SAMHSA], 2017). The criteria for an MDE 2006; Hess & Shultz, 2008). From this perspective,
include a period of two weeks or longer in the past 12 the individual is nested in multiple environments
months when an individual experiences a depressed including the micro-, meso-, exo-, and macrosystems.
mood or loss of interest or pleasure in daily activities The microsystem refers to an individual’s immediate
and some additional symptoms, for instance difficulty social environment including student–teacher rela-
with sleep (SAMHSA, 2017). Understanding the tionships and exerts the most influence on develop-
system-level factors that contribute to an adolescent’s ment (Hess & Shultz, 2008). The meso, exo, and
vulnerability to depression is critical to developing tar- macrosystems are more distal and consist of the inter-
geted interventions to prevent its occurrence. Consis- action between microsystems (for example, parents–
tent with the ecological perspective, the theoretical school), indirect environmental influences, and social

doi: 10.1093/cs/cdy024 © 2018 National Association of Social Workers 7

and cultural values (Hess & Shultz, 2008). Pertinent include small school size, effective classroom manage-
to this article, the relationships that youths have with ment strategies; moderate, less harsh school discipline
teachers and with their school represent important, policies; and opportunities for students to provide
proximal microsystems that influence youth develop- meaningful input into school policies (McNeely,
ment and outcomes including emotional health and Nonnemaker, & Blum, 2002; Whitlock, 2006).
well-being. A recent systematic review of the effects Teacher support is defined as the “extent to which
of the school environment on the emotional health teachers are supportive, responsive, and committed

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

of adolescents supports school connectedness and to students’ well-being” (Wang, 2009, p. 242).
teacher support as important predictors of future Teacher support and school connectedness are dis-
emotional health (Kidger, Araya, Donovan, & Gunnell, tinct concepts (McNeely & Falci, 2004); however,
2012). Furthermore, school connectedness and teacher they are also linked. For instance, teacher support
support may inhibit the initiation of high-risk behaviors leads to increases in students’ reports of liking school
such as suicide attempts, substance use, violent (Roeser, Eccles, & Sameroff, 1998), feelings of belong-
behavior, and truancy (Catalano, Haggerty, ing (Roeser, Midgley, & Urdan, 1996), and interest and
Oesterle, Fleming, & Hawkins, 2004; Coker & enjoyment in school (Wentzel, 1998). A recent meta-
Borders, 2001; Fleming, Merry, Robinson, Denny, analysis examining correlates of school belonging
& Watson, 2007; LaRusso, Romer, & Selman, found that teacher support, along with positive
2008; McNeely & Falci, 2004; Roeser, Eccles, personal characteristics (for example, optimism,
& Sameroff, 2000). self-esteem), is one of the strongest predictors of
The school environment is particularly important school belonging (Allen, Kern, Vella-Brodrick,
for the emotional well-being of high-risk youths (for Hattie, & Waters, 2018). A growing body of
example, youths in and out of home care or those research points to positive student–teacher relation-
who have been maltreated), who are often identified ships and strong perceptions of school connected-
in state and federal education laws (that is, Elementary ness as significantly associated with fewer reported
and Secondary Education Act of 1965 [P.L. 89-10] depressive symptoms ( Joyce & Early, 2014; Kidger
and subsequent reauthorizations) as most in need of et al., 2012; Shochet, Dadds, Ham, & Montague,
intervention. Given the powerful influence of the 2006; Shochet, Homel, Cockshaw, & Montgomery,
school environment, youths (particularly those who 2008). Furthermore, positive perceptions of teacher
are most at risk) may benefit from school-based support and school connectedness may even predict
interventions that promote protective factors to fewer future depressive symptoms in adolescents
prevent mental health conditions such as depres- (Bond et al., 2007; Murberg & Bru, 2009; Reddy,
sion (Markowitz, 2017). School social workers Rhodes, & Mulhall, 2003; Shochet et al., 2006;
are well positioned to play a role in reducing the Shochet & Smith, 2014; Wang, 2009; Way, Reddy,
risk for depression by intervening in ways to facil- & Rhodes, 2007). School connectedness appears to be
itate and enhance the school-based relationships particularly influential as the positive impact on
that affect the emotional well-being of adoles- depressive symptoms seems to persist into early
cents. School social workers may engage in this adulthood (Markowitz, 2017). Moreover, among
work through one-on-one contact with students vulnerable youths, school connectedness is associated
in tier 3 of the multitiered systems of support with a number of positive outcomes such as fewer re-
(MTSS) and through primary prevention strate- ported depressive symptoms, less suicidal ideation,
gies in tiers 1 and 2 to address environmental fac- and higher levels of self-esteem (Foster et al., 2017).
tors that affect emotional well-being (Kelly et al., Given the relationship between school connected-
2010). The available research points to teacher ness and teacher support, school connectedness may
support and school connectedness as important mediate or account for the effects of teacher support
targets for school-based interventions. on depressive symptoms. For example, LaRusso et al.
School connectedness refers to the extent to which (2008) found that prospective teacher support was
students “feel personally accepted, respected, included, positively related to social belonging, which was
and supported by others—especially teachers and inversely related to depressive symptoms. In addition,
other adults in the school social environment” prospective and cross-sectional analyses by Shochet
(Goodenow & Grady, 1993, pp. 60–61). School and Smith (2014) showed that school connectedness
factors related to increased school connectedness partially explained the relationship between the

8 Children & Schools Volume 41, Number 1 January 2019

classroom environment (using the classroom along with teachers and depressive symptoms? To
environment scale, which includes teacher sup- answer these questions, the study performed a sec-
port items) and depressive symptoms in seventh- ondary data analysis of waves 1 and 2 of the In-
and eighth-grade students. However, each variable Home Adolescent Questionnaire data collected as
still accounted for significant unique variance in part of the National Longitudinal Study of Adoles-
depressive symptoms, suggesting partial mediation. In cent to Adult Health (Add Health). Add Health is a
summary, the available research demonstrates that nationally representative, school-based probability

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

both teacher support and school connectedness con- study of U.S. adolescents in grades seven through
tribute to the current report and prediction of depres- 12. The In-Home Adolescent Questionnaire gath-
sive symptoms. However, whether or not school ered information on topics such as general health,
connectedness mediates the link between teacher academics and education, and depression and
support and depressive symptoms needs further was administered from September 1994 through
examination. Understanding the contributions of December 1995 (wave 1) and April 1996 through
both teacher support and school connectedness to August 1996 (wave 2) (Harris, 2005). The question-
depressive symptoms is important to developing naires were administered using computer-assisted
and informing targeted interventions that pro- personal interviews and audio computer-assisted
mote positive youth mental health and reduce self-interviews (Harris, 2005). During development
high-risk behaviors and emotional distress. of the in-home adolescent interview, the survey
instrument and component parts were extensively
PRESENT STUDY pilot tested (Udry, 2001). Although the data are
To help inform school social work interventions and over 20 years old, studies using data collected from
build on prior research, the present study examined youths today support a strong relationship between
the extent to which school connectedness mediated school connectedness and depressive symptoms
the relationship between teacher support and depres- (Foster et al., 2017; Shochet et al., 2006) and demon-
sive symptoms in adolescents. The conceptual model strate the potential role that school connectedness
is illustrated in Figure 1. Based on past research find- plays as a mediator between the classroom environ-
ings, it was expected that school connectedness ment and depressive symptoms (Shochet & Smith,
would partially mediate the relationship between 2014). The current study was approved by the
teacher support and depressive symptoms. institutional review board of Auburn University,
Auburn, Alabama.
METHOD
Sample
The present study addressed two main questions:
A total of 14,738 participants completed two waves
(1) Does school connectedness mediate the relation-
of the In-Home Adolescent Questionnaire. The
ship between feeling cared about by teachers and
sample of youths who participated in both waves
depressive symptoms? and (2) Does school connect-
and had completed responses on all study variables
edness mediate the relationship between getting
included 13,120 participants. The study sample char-
acteristics are provided in Table 1. The majority of
Figure 1: Conceptual Model: How the sample (63 percent) were older adolescents (10th
School Connectedness Mediates the through 12th grades) with a slightly higher percent-
Relationship between Teacher Support age of female participants (52 percent). Fifty-two per-
and Depressive Symptoms cent of the participants were white, 22 percent black,
16 percent Hispanic, 7 percent Asian, and 3 percent
Covariates Native American or other.
School
connectedness Measures
(wave 2)
a
b School Connectedness. School connectedness was
measured with three items, “I feel close to people
c
at this school,” “I feel like I am part of this school,”
Teacher support (c’) Depression and “I am happy to be at this school.” The three
(wave 1) (wave 2)
items were scored on a five-point scale ranging

Joyce / Does School Connectedness Mediate the Relationship between Teacher Support and Depressive Symptoms? 9
 Table 1: Sample Characteristics
using the 20-item CES-D (McDowell & Newell,
(N = 13,120) 1996). Scores ranged from 0 to 57, with higher scores
indicating greater depression. The items had a good
Characteristic n (%)
internal consistency reliability score of α = .87.
Gender
Male 6,324 (48)
Teacher Support. Teacher support was measured
Female 6,766 (52) with two separate items. The first item asked, “Since
Grades school started this year, how often have you had trou-

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

Seventh–ninth 4,444 (34) ble getting along with your teachers?” Response cate-
10th–12th 8,269 (63) gories ranged from 0 = never to 4 = every day.
Race and ethnicity Responses to this item were reverse coded so that a
White 6,804 (52) higher score indicated a stronger relationship. The sec-
Black 2,911 (22)
ond item about teachers asked, “How much do you
Hispanic 2,135 (16)
feel that your teachers care about you?” The response
Asian 911 (7)
Native American or other 359 (3) categories ranged from 1 = not at all to 5 = very
much. A higher score indicated a stronger relationship.
Covariates. Gender, race, and prior depressive
from 1 = strongly agree to 5 = strongly disagree. symptoms were included as covariates given that
The items were reverse coded so that higher scores they are often significantly related to youth reports
reflected higher levels of school connectedness. of depressive symptoms. Gender was measured as a
The items were summed to create a composite score dichotomous variable on the basis of respondent
and averaged to create an average composite score. self-report (0 = male). Race and ethnicity were mea-
These items, developed by Bollen and Hoyle (1991), sured as non-Hispanic white, Hispanic, black or
captured the social belonging dimension of school con- African American, Asian or Pacific Islander, American
nectedness. In the study sample, the items had low but Indian or Native American, and other. Race was later
acceptable internal consistency reliability of α = .78. dummy coded into two categories: racial minority
Depressive Symptoms. Depressive symptoms and racial majority (1 = racial majority). Prior depres-
were captured with 19 items from the Center for sive symptoms were identified using the 19 items
Epidemiologic Studies Depression Scale (CES-D) from the CES-D collected at wave 1 (α = .85).
(Radloff, 1977). The CES-D is a short self-report
scale used to assess depressive symptoms. The items in
DATA ANALYSIS
the scale are symptoms associated with depression (for
All analyses for this study used Stata 13.1 software due
example, “I was bothered by things that usually don’t
to its capacity to manage large, complex survey data
bother me”). Adolescents’ responses were based on
sets. The purpose of the Sobel–Goodman test is to
how frequently in a seven-day period they experi-
test whether a mediator carries the influence of an
enced a particular symptom. Response categories
independent variable to a dependent variable (Baron
ranged from 0 = rarely or none of the time; less than
& Kenny, 2004; Preacher & Hayes, 2004). Therefore,
one day to 3 = most or all of the time; five to seven
two separate Sobel–Goodman tests of mediation
days. Four positively worded items (for example,
were performed to answer the research questions.
“You felt you were just as good as other people”)
The first Sobel–Goodman test examined whether
were reverse scored so a higher number represented
school connectedness carried the influence of the
greater depressive symptoms. The Add Health study
impact of feeling cared about by teachers on depres-
used an investigator-adapted CES-D, with 19 items,
sive symptoms; the second test determined whether
rather than the standard 20 items (Crocket, Randall,
school connectedness carried the influence of the
Shen, Russell, & Driscoll, 2005). To be consistent
relationship between getting along with teachers
with the CES-D for youths, the Add Health version
and depressive symptoms.
did not ask about trouble sleeping or crying spells but
added a question on whether the individual thinks life
is not worth living (Perreira, Deeb-Sossa, Harris, & RESULTS
Bollen, 2005). The score was calculated by summing With the covariates included, stronger perceptions of
the individual item responses, dividing by 19, and feeling cared about by teachers at wave 1 was signifi-
multiplying by 20 to ensure comparability with studies cantly associated with fewer depressive symptoms at

10 Children & Schools Volume 41, Number 1 January 2019

 Figure 2: How School Connectedness Figure 3: How School Connectedness
Mediates the Relationship between Mediates the Relationship between
Feeling Cared About by Teachers and Getting Along with Teachers and
Depressive Symptoms Depressive Symptoms

Covariates Covariates
School

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

connectedness School
(wave 2) connectedness
.10 (SE = .00)*** –1.57 (SE = .07)*** .04 (SE = .00)*** –1.61 (SE = .07)***
(wave 2)
–.47 (SE = .05)***
–.36 (SE = .06)***
Feeling that teachers Depression Getting along with Depression
care about you (–.30 [SE = .05]***) (–.29 [SE = .05]***)
(wave 2) teachers (wave 2)
(wave 1) (wave 1)
***p < .001. ***p < .001.

wave 2: B = −.47 (SE = .05; p < .001). Feeling cared DISCUSSION

about by teachers at wave 1 was also significantly This study examined the extent to which school con-
associated with higher perceptions of school connect- nectedness mediated the relationship between teacher
edness at wave 2: B = .10 (SE = .00; p < .001). When support and depressive symptoms in adolescents. The
the mediator, school connectedness, was included in study findings reinforce the critical impact that the
the model the impact of feeling cared about by teach- school environment has on the emotional health of
ers at wave 1 on depressive symptoms at wave 2 was adolescents (Kidger et al., 2012; Roeser et al., 2000).
still significant: B = –.30 (SE = .05; p < .001); how- Consistent with prior research, both teacher support
ever, it was reduced suggesting partial mediation (see and school connectedness were associated with fewer
Figure 2). The mediation effect of school connected- depressive symptoms (Foster et al., 2017; Joyce &
ness was statistically significant with approximately 35 Early, 2014; Shochet et al., 2006). The current study
percent of the total effect of feeling cared about by suggests one way that schools may reduce depressive
teachers on depressive symptoms being mediated: symptoms is by fostering strong teacher–student rela-
tionships. Specifically, getting along with teachers
Indirect efffect .167337 and feeling cared about by teachers at wave 1 signifi-
= = .35 cantly predicted fewer depressive symptoms at wave 2,
Total effect .4725591
which is consistent with other prospective study find-
Reports of getting along with teachers at wave 1 ings (Reddy et al., 2003; Wang, 2009; Way et al.,
also corresponded with fewer depressive symptoms at 2007). This finding is supported by other research
wave 2: B = −.36 (SE = .06; p < .001) with the asserting the importance of teacher–student relation-
covariates included. Getting along with teachers at ships, particularly for youths experiencing adversity
wave 1 was also significantly associated with higher (McLaughlin & Clarke, 2010; Resnick et al., 1997).
perceptions of school connectedness at wave 2: Also, as with past research (for example, LaRusso
B = .04 (SE = .00; p < .001). When the mediator, et al., 2008), the current study findings suggest that
school connectedness, was included in the model, both getting along with teachers and feeling cared
the impact of getting along with teachers at wave 1 about by teachers influence depressive symptoms, in
on depressive symptoms at wave 2 was still signifi- part due to their impact on perceptions of school
cant: B = −.29 (SE = .05; p < .001). However, it connectedness. However, getting along with teachers
was reduced, indicating partial mediation (see Figure 3). and feeling cared about by teachers continued to
The mediation effect of school connectedness was have a unique impact on depressive symptoms. The
statistically significant with approximately 18 percent evidence of partial mediation found in this study is
of the total effect of feeling cared about by teachers similar to the findings of Shochet and Smith (2014).
on depression being mediated:
Limitations
Indirect efffect .067585 The study findings should be interpreted with con-
= = .18
Total effect .36365 sideration of a few limitations. For instance, the
school connectedness measurement only included

Joyce / Does School Connectedness Mediate the Relationship between Teacher Support and Depressive Symptoms? 11
three items, and the teacher support measures were strengths focus, empathy, and inclusion and equity,
single item indicators. In addition, the findings may align with the values, theoretical orientation, and
have been stronger and provided more evidence for skills base of the social work profession. With train-
causality if the outcome variable (that is, depressive ing, school social workers may facilitate RAP-T
symptoms) was measured at a third time point. programs. Furthermore, enhanced teacher training
Finally, the Add Health data were collected over 20 or professional development on the key relationship
years ago, so the passage of time could affect the gen- features associated with positive youth outcomes

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

eralizability of the findings to youths today. may be beneficial (Mihalas, Morse, Allsopp, &
Mchatton, 2009; Shochet & Wurfl, 2006). Ele-
Implications for Practice and Future ments of such relationships include adults who
Research express care, challenge growth, and provide support
Based on the findings from this study, implications (Search Institute, 2016). Other important relationship
may be drawn for social work practice in schools and features include active involvement, open communi-
for future research. School social workers serve at cation, responsiveness, clear rules, positive feedback,
multiple levels of practice, intervening with individ- and warmth (Murray & Greenberg, 2001; Murray &
ual students, small groups, teachers, parents, and the Pianta, 2007). Of note, Kelly and colleagues (2016)
overall school environment (Kelly et al., 2010; Kelly surveyed school social workers and found that 28.6
et al., 2016). The findings from this study support the percent reported providing in-service support for
importance of intervening at multiple levels to sup- teachers, so nearly a third of school social workers
port emotional well-being and prevent depression. are working in this capacity; however, at this time,
School social workers may enhance student–teacher many school social workers may not be engaged in
relationships by working one-on-one with students this work.
at the tier 3 or targeted intervention level of MTSS Teachers and student–teacher relationships should
and by working with teachers at the tier 2 or small not be the sole focus, however. Developing and sus-
group intervention MTSS level. For example, with taining schoolwide initiatives (tier 1 of MTSS) to
individuals, school social workers may address depres- promote school connectedness is also critical. For
sive symptoms through cognitive–behavioral therapy instance, school social workers may promote and
(for example, improving negative cognitions) and implement social and emotional curricula to help stu-
interpersonal therapy (for example, teaching social dents develop the positive individual characteristics
skills), two common evidence-based treatments for (for example, optimism) that appear to be strongly
depression (Shochet et al., 2006). These therapeutic associated with school belonging (Allen et al., 2018).
techniques may be useful to help adolescents with To prepare social workers, school social work train-
depression to build positive social skills that may ing should emphasize the skills and knowledge
improve teacher relationships and enhance school needed to target the whole school environment
connectedness (Shochet et al., 2006). (Kelly et al., 2016). Students may also experience
School social workers may also facilitate professional school connection through school-based activities
development and training for teachers at the tier 1 or and opportunities (Gowing & Jackson, 2016), parent
universal support level of MTSS, focusing on the involvement (Raferty, Grolnick, & Flamm, 2012),
important role they play in building school connect- and opportunities to provide meaningful input on
edness. For example, teachers may learn how to school policies (Whitlock, 2006).
enhance students’ connection to school through pro- The study findings also point to opportunities for
grams such as the Resourceful Adolescent Program future research. This study may be replicated using
for Teachers (RAP-T), which provides teachers with stronger measures of teacher support and school
strategies to create a warm and inclusive environment connectedness. Furthermore, it may be valuable to
(Shochet & Wurfl, 2006). The RAP-T program helps test these relationships among vulnerable groups of
teachers develop strategies to promote four critical ele- youths such as sexual minority youths who are at
ments of school connectedness: warm relationships risk for lower school connectedness, poorer-
(for example empathy, respect), inclusion, a focus quality student–teacher relationships ( Joyce, 2015),
on strengths, and equity and fairness (Shochet & and higher incidence of depression (Hatzenbuehler,
Smith, 2014). The program elements, specifically McLaughlin, & Nolen-Hoeksema, 2008).

12 Children & Schools Volume 41, Number 1 January 2019

CONCLUSION Elementary and Secondary Education Act of 1965, P.L. 89-
Depression among adolescents is a serious public 10, 79 Stat. 27 (April 9, 1965).
Fleming, T. M., Merry, S. N., Robinson, E. M., Denny, S.
health issue that can threaten the healthy develop- J., & Watson, P. D. (2007). Self-reported suicide at-
ment of youths. The school environment includes tempts and associated risk and protective factors
among secondary school students in New Zealand.
multiple systems (for example, student–teacher rela- Australian and New Zealand Journal of Psychiatry, 41,
tionships) that play a valuable role in supporting the 213–221.
Foster, C. E., Horwitz, A., Thomas, A., Opperman, K.,
healthy development of adolescents. Specifically,

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

Gipson, P., Burnside, A., et al. (2017). Connectedness
both teacher support and school connectedness are to family, school, peers, and community in socially
important to preventing depressive symptoms among vulnerable adolescents. Children and Youth Services
Review, 81, 321–331. doi:10.1016/j.childyouth.2017
adolescents. Positive student–teacher relationships may .08.011
promote school connectedness and subsequently lower Gilbert, M. C. (2004). Childhood depression: A risk factor
perspective. In M. Fraser (Ed.), Risk and resilience in
depressive symptoms. Social workers in schools have childhood: An ecological perspective (pp. 315–346). Wash-
the opportunity to play an important role in strength- ington, DC: NASW Press.
ening student–teacher relationships and promoting Goodenow, C., & Grady, K. E. (1993). The relationship of
school belonging and friends’ values to academic
strategies that enhance student connection to motivation and among urban adolescent students. Jour-
schools. CS nal of Experimental Education, 62(1), 60–71.
Gowing, A., & Jackson, A. C. (2016). Connecting to
school: Exploring student and staff understandings of
connectedness to school and the factors associated
REFERENCES with this process. Educational and Developmental Psychol-
Allen, K., Kern, M. L., Vella-Brodrick, D., Hattie, J., & ogist, 33(1), 54–69. doi:10.1017/edp.2016.10
Waters, L. (2018). What schools need to know about Harris, K. M. (2005, September). Design features of Add
fostering school belonging: A meta-analysis. Educa- Health. Chapel Hill: University of North Carolina at
tional Psychology Review, 30(1), 1–34. doi:10.1007/ Chapel Hill, Carolina Population Center.
s10648-016-9389-8 Hatzenbuehler, M. L., McLaughlin, K. A., & Nolen-
Baron, R. M., & Kenny, D. A. (2004). The moderator- Hoeksema, S. (2008). Emotional regulation and inter-
mediator variable distinction in social psychological nalizing symptoms in a longitudinal study of sexual
research: Conceptual, strategic, and statistical consid- minority and heterosexual adolescents. Journal of Child
erations. Journal of Personality and Social Psychology, 51, Psychology and Psychiatry, 49, 1270–1278.
1173–1182. Hess, S. A., & Shultz, J. M. (2008). Bronfenbrenner’s eco-
Bollen, K., & Hoyle, R. H. (1991). Perceived cohesion: A logical model. In K. L. Kraus (Ed.), Lenses: Applying
conceptual and empirical examination. Social Forces, lifespan development theories in counseling (pp. 52–82).
69, 479–504. Boston: Lahaska Press.
Bond, L., Butler, H., Thomas, L., Carlin, J., Glover, S., Joyce, H. D. (2015). School connectedness and student–
Bowes, G., & Patton, G. (2007). Social and school teacher relationships: A comparison of sexual minority
connectedness in early secondary school as predictors youths and their peers. Children & Schools, 37,
of late teenage substance use, mental health, and aca- 185–192. doi:10.1093/cs/cdv012
demic outcomes. Journal of Adolescent Health, 40, 357. Joyce, H. D., & Early, T. J. (2014). The impact of school
e9–357.e18. doi:10.1016/j.jadohealth.2006.10.013 connectedness and teacher support on depressive
Bronfenbrenner, U. (1979). The ecology of human develop- symptoms in adolescents: A multilevel analysis. Chil-
ment. Cambridge, MA: Harvard University Press. dren & Youth Services Review, 39, 101–107. doi:10
Bronfenbrenner, U., & Morris, P. A. (2006). The bioecolo- .1016/j.childyouth.2014.02.005
gical model of human development. In W. Damon & Kelly, M. S., Berzin, S. C., Frey, A., Alvarez, M., Shaffer,
R. M. Lerner (Eds.), Handbook of child psychology: Vol. G., & O’Brien, K. (2010). The state of school social
1. Theoretical models of human development (6th ed., work: Findings from the National School Social Work
pp. 793–828). New York: Wiley. Survey. School Mental Health, 2, 132–141. doi:10
Catalano, R. F., Haggerty, K. P., Oesterle, S., Fleming, .1007/s1231-00-9034-5
C. B., & Hawkins, J. D. (2004). The importance of Kelly, M. S., Frey, A., Thompson, A., Klemp, H., Alvarez,
bonding to school for healthy development: Findings M., & Berzin Cosner, S. (2016). Assessing the national
from the social development research group. Journal of school social work practice model: Findings from the
School Health, 74(7), 252–261. Second National School Social Work Survey. Social
Coker, J. K., & Borders, L. D. (2001). An analysis of envi- Work, 6, 17–28. doi:10.1093/sw/swv044
ronmental and social factors affecting adolescent prob- Kidger, J., Araya, R., Donovan, J., & Gunnell, D. (2012).
lem drinking. Journal of Counseling & Development, 79, The effect of the school environment on the emo-
200–208. tional health of adolescents: A systematic review. Pedi-
Crockett, L., Randall, B., Shen, Y.-L., Russell, S., & Dris- atrics, 129, 925–943.
coll, A. (2005). Measurement equivalence of the Cen- LaRusso, M. D., Romer, D., & Selman, R. L. (2008). Tea-
ter for Epidemiologic Studies Depression Scale for chers as builders of respectful school climates: Im-
Latino and Anglo adolescents: A national study. Journal plications for adolescent drug use norms and
of Consulting and Clinical Psychology, 73, 47–58. depressive symptoms in high school. Journal of Youth
Dallaire, D. H., Cole, D. A., Smith, T. M., Ciesla, J. A., and Adolescence, 37, 386–398. doi:10.1007/s10964-
LaGrange, B., Jacquez, F. M., et al. (2008). Predicting 007-9212-4
children’s depressive symptoms from community and Markowitz, A. (2017). Associations between school con-
individual risk factors. Journal of Youth Adolescence, 37, nections and depressive symptoms from adolescence
830–846. through early adulthood: Moderation by early

Joyce / Does School Connectedness Mediate the Relationship between Teacher Support and Depressive Symptoms? 13
 adversity. Journal of Research on Adolescence, 27, experience in middle school. Developmental and Psycho-
298–311. doi:10.1111/jora.12275 pathology, 10, 321–352.
McDowell, I., & Newell, C. (1996). The Center for Epidemio- Roeser, R. W., Eccles, J. S., & Sameroff, A. J. (2000).
logic Studies Depression Scale in measuring health: A guide School as a context of early adolescents’ academic
to rating scales and questionnaires (2nd ed.). Oxford, UK: and social-emotional development: A summary of
Oxford University Press. research findings. Elementary School Journal, 100,
McLaughlin, C., & Clarke, B. (2010). Relational matters: A 443–471.
review of the impact of school experience on mental Roeser, R. W., Midgley, C., & Urdan, T. (1996). Percep-
health in early adolescence. Educational & Child Psy- tions of the psychological environment and early ado-

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

chology, 27(1), 91–103. lescents’ psychological and behavioral functioning in
McNeely, C. A., & Falci, C. (2004). School connectedness school: The mediating role of goals and belonging.
and the transition into and out of health-risk behavior Journal of Educational Psychology, 88, 408–422.
among adolescents: A comparison of social belonging Search Institute. (2016, July). The developmental relationships
and teacher support. Journal of School Health, 74, framework. Retrieved from http://www.search-
284–292. institute.org/downloadable/Developmental-
McNeely, C. A., Nonnemaker, J. M., & Blum, R. W. Relationships-Framework.pdf
(2002). Promoting school connectedness to school: Shochet, I. M., Dadds, M. R., Ham, D., & Montague, R.
Evidence from the National Longitudinal Study of (2006). School connectedness is an underemphasized
Adolescent Health. Journal of School Health, 72, parameter in adolescent mental health: Results of a
138–146. community prediction study. Journal of Clinical Child
Mihalas, S., Morse, W. C., Allsopp, D. H., & Mchatton, and Adolescent Psychology, 35, 170–179.
P. A. (2009). Cultivating caring relationships between Shochet, I. M., Homel, R., Cockshaw, W. D., &
teachers and secondary students with emotional and Montgomery, D. T. (2008). How do school connect-
behavioral disorders. Remedial and Special Education, edness and attachment to parents interrelate in predict-
30, 108–124. doi:10.1177/0741932508315950 ing adolescent depressive symptoms. Journal of Clinical
Mojtabai, R., Olfson, M., & Han, B. (2016). National Child & Adolescent Psychology, 37, 676–681. doi:10
trends in the prevalence and treatment of depression in .1080/15374410802148053
adolescents and youth adults. Pediatrics, 138(6), 1–10. Shochet, I. M., & Smith, C. L. (2014). A prospective study
doi:10.1542/peds.2016-1878 investigating the links among classroom environment,
Murberg, T. A., & Bru, E. (2009). The relationships school connectedness, and depressive symptoms in
between negative life events, perceived support in the adolescents. Psychology in the Schools, 51, 480–492.
school environment and depressive symptoms among Shochet, I. M., & Wurfl, A. (2006). Resourceful Adolescent
Norwegian senior high school students: A prospective Program for Teachers (RAP-T): A program for teachers to
study. Social Psychology of Education, 12, 361–370. promote school connectedness in teenagers. Brisbane, Aus-
Murray, C., & Greenberg, M. T. (2001). Relationships tralia: Queensland University of Technology, School
with teachers and bonds with school: Social emotional of Psychology and Counseling.
adjustment correlates for children with and without Substance Abuse and Mental Health Services Administra-
disabilities. Psychology in the Schools, 38, 25–41. tion. (2017). Key substance use and mental health indicators
Murray, C., & Pianta, R. C. (2007). The importance of in the United States: Results from the 2016 National Survey
teacher-student relationships for adolescents with high on Drug Use and Health (HHS Publication No. SMA
incidence disabilities. Theory into Practice, 46, 105–112. 17-5044, NSDUH Series H-52). Rockville, MD:
Perreira, K. M., Deeb-Sossa, N., Harris, K. M., & Bollen, Substance Abuse and Mental Health Services Admin-
K. (2005). What are we measuring? An evaluation of istration, Center for Behavioral Health Statistics and
the CES-D across race/ethnicity and immigrant gen- Quality. Retrieved from http://www.samhsa.gov/
eration. Social Forces, 83, 1567–1602. data/
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS proce- Udry, J. R. (2001). References, instruments, and questionnaires
dures for estimating indirect effects in simple media- consulted in the development of the Add Health in-home ado-
tion models. Behavior Research Methods, Instruments, & lescent interview. Retrieved from http://www.cpc.unc
Computers, 36, 717–731. .edu/projects/addhealth/documentation/guides/
Radloff, L. S. (1977). The CES-D Scale: A self-report refer.pdf
depression scale for research in the general population. Wang, M. T. (2009). School climate support for behavioral
Applied Psychological Measurement, 1, 385–401. and psychological adjustment: Testing the mediating
Raferty, J. N., Grolnick, W. S., & Flamm, E. S. (2012). effect of social competence. School Psychology Quarterly,
Families as facilitators of student engagement: Toward 24, 240–251.
a home-school partnership model. In S. L. Christen- Way, N., Reddy, R., & Rhodes, J. (2007). Students’ per-
son, A. L. Reschly, & C. Wylie (Eds.), Handbook of ceptions of school climate during the middle school
research on student engagement (pp. 343–364). New years: Associations with trajectories of psychological
York: Springer Science + Business Media. and behavioral adjustment. American Journal of Commu-
Reddy, R., Rhodes, J., & Mulhall, P. (2003). The influence nity Psychology, 40, 194–213.
of teacher support on student adjustment in the middle Wentzel, K. R. (1998). Social relationships and motiva-
school years: A latent growth curve study. Development tion in middle school: The role of parents, teachers,
and Psychopathology, 15(1), 119–138. doi:10.1017 and peers. Journal of Educational Psychology, 90,
.S0954579403000075 202–209.
Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, Whitlock, J. L. (2006). Youth perceptions of life at school:
K. E., Harris, K. M., Jones, J., et al. (1997). Protecting Contextual correlates of school connectedness in ado-
adolescents from harm: Findings from the National lescence. Applied Developmental Science, 10(1), 13–29.
Longitudinal Study on Adolescent Health. JAMA,
278, 823–832. Hilary D. Joyce, PhD, is assistant professor of social work and
Roeser, R. W., Eccles, J. S., & Sameroff, A. J. (1998). Aca-
demic and emotional functioning in early adolescence: BSW program director, Department of Sociology, Anthropology,
Longitudinal relations, patterns, and prediction by and Social Work, Auburn University, 7030-H Haley Center,

14 Children & Schools Volume 41, Number 1 January 2019

Auburn, AL 36849-5049; e-mail: [email protected]. This
research uses data from Add Health, a program project directed by
Kathleen Mullan Harris and designed by J. Richard Udry, Peter
S. Bearman, and Kathleen Mullan Harris at the University of
North Carolina at Chapel Hill, and funded by grant P01-
HD31921 from the Eunice Kennedy Shriver National Institute of
Child Health and Human Development, with cooperative funding

Downloaded from https://academic.oup.com/cs/article/41/1/7/5181599 by AIRLANGGA UNIVERSITY user on 26 November 2024

from 23 other federal agencies and foundations. Special acknowledg-
ment to Ronald R. Rindfuss and Barbara Entwisle for assistance in
the original design. Information on how to obtain the Add Health
data files is available on the Add Health Web site (http://www
.cpc.unc.edu/addhealth). No direct support was received from grant
P01-HD31921 for this analysis.
Original manuscript received June 22, 2017
Final revision received February 27, 2018
Editorial decision March 12, 2018
Accepted March 15, 2018
Advance Access Publication November 14, 2018

Joyce / Does School Connectedness Mediate the Relationship between Teacher Support and Depressive Symptoms? 15