Content uploaded by Carol Goodenow
Author content
All content in this area was uploaded by Carol Goodenow on Mar 17, 2015
Content may be subject to copyright.
The
Relationship
of
Scl~ool
Belonging
and
Friends'
Values
to
Academic
Motivation
Among
Urban
Adolescent
Students
CAROL
COODENO\V
Tufts
University
KATHLEEN
E
CRADY
blrruchusctls
institute
or Behavioral Medicine
ABSTRACT.
Students' subjective sense of school belunging recently has heen identi-
fied
ns
a porentiall~ important infiuence
on
acadenric nrotivatioa, eilgagrmcet, aud
participation, especially anrung students Prom groups ut
risk
of
rcl~uul drupuut. Stu-
den&' friends also influence their academic motivatiui-r, so~lletin~es negatively. 111
tlris
study. the relationship among early adolescent students' serrse of sclruul Ireluagiag,
perreptions of their friends' academic values, and acadrn~ic motivation was
investi-
gated
among
301
African-American, \Vhite/Anglo. and lfispanic students
in
two
ctr-
ban junior high schools. School belonging was sigr~ificantly associated wit11 sever;~l
motiration-related measures-expectancy of success, valuing sclruulwork, gceer111
rthwl motir-ation, and self-reported eKort. Students' beliefs about their friencls' ar:t-
demic values were more ~eakly related lo these o~btcon~es. The correl;rtiurrs Irettveci~
school belonging and the motivation-related nleasures ren~air~ed positive and stotisti-
ally
significant
eben
after the enects of friends' acaden~ic values were p:rrtiallccl
oul. Schwl belonging uas more highly associated with expectancy fur sllrccss alllo11g
Hispanic
students than among African-American students, and
sarong
girls
11r:tit
among
bajs.
THERE
IS
A
GROWING
CONSENSUS
that academic motivation is not
n
purely individual, intrapsychic state; rather, it grows out of a complex weh
of
social
and
personal relationships. As Weiner
(
1990) stated, "Sct~ool niotivation
cannot
be
understood apart from the social fabric in which
it
is embedded" (p.
621). Students' associalions with cultural and ethnic groups, their families, ant1
lhcir friends lespecially in adolescence) are fundar~~ental aspects of tl~is soci;rl
fabric. Another potentially important element of the social context is students'
sense
of
belonging in the school or classroom, that is, the extent to which they feel
personally accepted, respected, included, and supported by others-especially
teachers and other adults in the school social environn~ent.
The influence of friends and peer groups on students' academic achievement
and attitudes toward school has been widely noted (e.g.. Brown, 1990; Coleman.
1961; Steinberg, Dornbusch,
&
Brown. 1992). with early adolescence marking
the highest point of
conformity
to
peers (Bemdt, 19791. Because peer influence
can be negative as well as positive. it may pose special dilemmas for students
whose personal friendship groups hold antiacademic norms (Phelan. Davidson.
&
Cao, 1991). a situation often more common among ethnic minority
than
majority
students (Steinberg et al.,
1992).
The African-American high school sludenls
studied by Fordham (1988). for example, described overt striving for academic
achievement as aclit~g
hire,
a
sign of ethnic disloyalty. Whether peer influence
is tilore likely to have
a
negative educational impact on African-American or His-
panic adolescents than
on
WhitelAnglo students from similar socioeconomic
bi~ckgrounds, however. has not been determined.
A
inore recent approach to understanding social influences on schml moliva-
tion, especially alllong disadvantaged students. has focused on students' subjec-
tive sense
of
scl~ool belonging. Finn (1989). for example, posited an identifica-
1io1i-1~u~ic.il~atiot1 model to account for school withdrawal among at-risk
stutlents. Basically. the model suggests that unless students identify with the
sclrool to at least a minimal extent; feel that they belong as pan of the school; and
bclicve thc~nselves to be welcon~ed, respected, and valued
by
others there. they
III;~~
begin tlre gradual disengage~~~ent process of which officially dropping out
is
oc~ly the li~~al step. I~~vestigatii~g over
5.000
eighth-grade students from dem*
graphic groups statistic;illy at risk of school failure (urban minority. low socioeco-
~roinic stntus, or non-IJuglis11 home language). Finn
(
1992) found that some di-
~~rcnsions of school belonging-particularly students' perceptions of teacher
support-predicted severill measures of school participation and engagement.
llsing priinarily ell~nograpliic n~e~hods to examine schools that had been suc-
cessful with at-risk students. Wehlage (1989) and Wehlage. Rutter. Smith. Lesko.
i111tl
Fernantlez (1989) used the term scllool nte~r~bership to refer to
a
construct
a1111ost idelltical to Fin~is belo~~gi~~g. The psychological sense of school member-
slrip. according to Wehlage,
is
not simply ~echnical enrollment but rather students'
perceptions that others in the school, especially adults, are
for
them and that they
co~~nt in the school.
This sense of n~eabership heavily influences students' commitment to school-
ieg i111d acceptance of educational values. Other ethnographic accounts (Farrell,
1990; Fine, 1991; Kratner-Schlosser, 1992; see also Kagan. 1990) also have
re-
vealed the ways in which students perceive (or unfortunately ohen fail to per-
ceive) tlre school as a pcrsonally supportive community.
Several rjuestions regarding scl~ool belonging call for more thorough empirical
research. Do
most students. especially urban adolescents and ethnic minority stu-
dents, feel
a
sense of schcml belonging? To what extent is belonging associared
with academic motivation and engagement? Is it
ttrorr
highly asswiatetl with Ino-
.
lir~tion for some subgroups of students than for others? Finally, is a solid sense
of membership in, and SuppMl from. the schrml community in general (especially
including 1e;lchet-s and other adults in the school) s~~fliciet~t to or~tweigl~ tl~c i~~llu-
enre
of
a stu&nt's friends or inmediate peer group in affecting motivation?
In the present study we investigated the Llypotlrcsis th;rl st~~tletits' sclise
ol'
Iae-
longing
in
school would be significantly associated with tneasures of sclrt)ol
no ti-
vation. expectations of academic success. valuing of academic work, and pcrsis-
tencc in difficult tasks. \Ve furlher hypo~lrrsizetl
111111
tlresc effects
ell'
school
belonging would be significant even after controlling for the inlluence of friends'
academic values. Although we made no direclional hypotheses regartling elhitic
or
gender diffefences in school belonging and motivation, or in the strength of
association between the two, we also investigated these differences.
Parliripants
and
Proctdrrres
We conducted the study in a middle-sized city in the Northeast. The city is
mostly working-class and has large Hispanic and African-American populations;
the average per capita income in the city places it in the bottom quartile of the
state.
AI
the time that this research project was conducted, studelits were assigned
lo
elementary and junior high schools on the basis
of
residence, resulting itr a
pattern
of
de facto segregation.
We conlacled two of the six junior high schools (Grades F9) in the city,
ir~itl
their staffs agreed to participate in the study. Procedures and participants dif'lkrcd
slightly by school. In School
I.
a randomly chosen half ofthe stude~lt body partic-
ipated
in
the study by filling out a questionnaire adrninistered in their hooreroon\s.
The other students
completed
a different questionnaire
i~t
this tinre.
A
total of 198
students completed the survey; 10-1 were boys and 87 were girls
(7
did not indicitle
gender) Eighty-nine students identified tIrenisel\,es as Africi~tr-An~ericatr or
Black.
32
as
Hispanic, 66 as White,
2
as Asian. and
9
indicated no etlr~rie ider~tili-
cation. Studen1 ages ranged from 12 to 16. averaging 13.80 years of age
(SLj
=
1.0).
Pariicipants included 74 seventh-grade students, 79 eighth-grade studc~~ts.
and
45
ninth-grade students.
In
School 2. only seventh-grade s~udents participated. As with the first schtx~l,
half of all seventh graders conrpleted the questionniiirr; the other Iralf corr~pleted
an
alicrna~i\~e questionnaire. The 103 participants ilrcludcd
54
boys :rnd
43
girls
(6
no answer).
\r
ith a11 average age of 13.
I I
ycors
(Sl)
=
.89).
'file
st~tdc~rt populir-
lion was largely Hispanic
(77
students, 16
of
who111 co~npleted tlre cluestion~liiirc
in Spanish). with
7
African-American students. 16 WL~itelAnglo students, ai~d
I
Asian-American. Two students did not indicate ethnicity.
The attendance rate for relevant students on the testing date, midweek in hla):
wits approximately 73%. School personnel indicated that this was not unusually
low, especially late iq the school year.
Scltool
belorlgittg.
School belonging was assessed by the Psychological Sense of
Scllool Me~nbersliip Scale (PSSM), an 18-item scale developed for use with
ear-
ly- and a~idadolescent stude~rls (Goodenow. 1993b3. The
PSShl
includes items
that involve trot only perceived liking, personal acceptance. and inclusion (e.g..
"Most teachers at this school are interes~ed in me,"
"
I feel like a real pan of this
school") but also respect and encouragement for participation
(e.g..
"People hem
notice
wires
I'm good at something," "Other students in this school take my opin-
ions seriously").
lterns on the PSSM and on the measures described below contained
a
5-point
1,ikert-type format, with choices ranging from ~ot
ur
all rnte
(1)
to roncplerely
rrrce
(5);
tlrese were then averaged to produce
a
scale score. Internal consistency
reliability (Cronbach's alpha) for this sanlple was
.80.
fi-ieltr1.r' trdrre.~.
To
assess
the
influence on
motivation
of students' personal
friendship network, we asked students to rate their agreement with the statement
"My friends think that it is iarportant to do well in school."
Alc~tirnriorz. Students were administered two scales based on
an
expectancy-value
tlteory of ~rrotivation (Atkinson. 1964; Eccles el al., 1983). According lo this the-
ory.
no ti vat ion
to
engage in achievenrent-related behavior is a join1 func~ion of
the expectation that such behavior will be successful and
he
value attached to
tlrat success. One of the scales contains five items measuring students' expectancy
of
success in scllcn~lwork; the other scale contains six iterns about the intrinsic
virl~e, ii~terest, and iliiportance that students attribute lo academic schoolwork.
Borh were sllortened versions of scales used by Pintrich and De Groot
(
1990) in
studying motivatio~~ in junior high school students. In this study, internal consis-
tency 1.eliabili1y (Cronbach's alpha) was .72 for the expectancy scale and .81 for
tlrc valtre scale.
To
supplctnent the expectancy and value scales that focus specifically on
aca-
demic scl~oolwork. we also administered a four-item general school motivation
Irteasure. 'l'llis scale. a shortened version of Ford and Tisak's
(
1982) School hloti-
vi~tion Scale, tneasures the more global set of beliefs and feelings that being in
sclrool is sirtisfying, worllr\vlrile, and important, rather than boring or irrelevant.
lnt~rtliil cc)~rsislclicy rcliebility of this scale for the sa~nplc
was
.61.
EIfi)rt/,rwrsiste,rc.e.
Two itenls on the questionnaire asked studenrs to rate the ex-
tent to which they put forth effort and persisted despite diffculties in schoolwork;
"I
work lrard lo gel good grades, even if
I
don't like
a
class," "When schoolwork
is hard.
1
just
do
the easy pans or
I
give LI~." Altlrot~gh these ilerns (lo tiot capture
the
full
range
of
school panicipation and engagement displayed
hy
coni~iii~~etl and
successful
students. they do target behaviors most likely to decline if studc~rts
btcomc discouraged or alienated from the schooling process.
Results
Descripti~r Statistics
There were
no
significant differences between the two schools
or
among grade
levels
on
students' mean scores on any of the scales. Therefore, we comlji~led
sludcn~s across schools and grade levels for all subsequent analyses.
As
shown in Table
I.
students, on average, scored well above rhe
3.0
scale
midpoint on the measures of expectancy, value. and effort. Students agreed that
TABLE
I
Darripllrc
StaMcs
Or
School
Belonging,
Friends'
Values, and hlotlvatlon hleasures
African- Whitel
Tnlrt Alacriran llisprnic Al~glo
hlnle
Fmt;rlc
hleruurr
IN
=
301)
(11
=
96)
In
=
109)
(11
=
82)
(11
=
158)
(11
=
130)
rbc
4
Be~ucen-cthnir-group or be~wccn-gender means followed
by the
sane ssllpcrscrip~
were
sigisliti-
c~ly d~llcren~
at
p
C
05
or
bel~ec.
lliey personally valued schoolwork and put
forth
effort; however. they uerc far
lcss likely to assert t!rat their friends thought that doing well
in
school was im-
portant
(M
=
3.05).
When responding to questions not focused on academic work per se but on
reactions to the school as a whole. students were less positive. The average score
on the scale measuring school belonging was only slightly above the
3.0
scale
~nidpoillt. In fact.
4
1
%
(11
.=
124)
of
the students were more inclined to disagree
tluan agree that they felt they belonged and were supported by others
in
the school.
Similarly,
40%
(11
=
122) of the students were more negative lhan positive in their
responses to the general school molivation scale.
We also investigated ethnic and gender differences in scale means. Several gen-
der differences appeared: Girls were more likely than boys lo express a high sense
of school belonging and general school motiva~ion and to say that their friends
~lrougllt that doing well in school was important. We found only one significant
etlinic difference: Hispanic students were less likely than WhitelAnglo students
to assert that they tried hard to do well
in
school.
The major focus of the study was the effec~ of school belonging and friends'
values on acaderriic
motivation
atrd effort. As shown
in
Table
2,
school belonging
was significantly correlated with friends' values and with all four outcome meas-
ures. accoutlling for approximately one fifth of the variance in expectancy
(r'
=
.I9)
and general school ~nolivation
(r'
=
.2l) and nearly one third of the variance
in
value
(rL
=
.30).
The
correlation
with
efforJpersistr~ice. although signitican~.
was quite small. Significant correlations between friends' values and hrce
of
he
four niotivation-related outcomes also were obtained, but each
was
smaller
than
the corresponding conelation for school belonging.
lntercorrelallons
of
Measures
of
School Bclonelng, Friends'
\'slues,
htotiratlon, and ERort
School Friends' School
hlcasures
belonging valuer Eapeclancy
Valuc
motituion
Scltool
belonging
-
l:~icnds'
va111c.s
.44**
-
I'.xl)cctnncy .4?**
.-
7j.I
-
VII~IIL'
01-
SCII(H)~WO~~
.55**
.37**
.54'*
-
SCIIIMII
III~IIIV~IIIIII
.Jb**
?I**
.
-
.)Us* .38**
-
L;.ll~~~c/pcrsis~ci~cr
.I?*
.02 .dl** .3l**
.22**
*/I
c
.05.
**/I
<
.WI.
" "
L
--
.:2;22
;
27r35
q
y'
0
2
-
5=
*
*
-
-od
s
z22.5
g.52,
=
-
2
a
d
"
"
z
S3"
2
s
-
4
c
3
&a
5
.-
-
-
UJ,
&"
g
;,?z2
CJ
-
3
52
2
g
2=3
5
S3
.cs
5
>.+a
*
p4Zu2
Z
tE
g.5
2
Esc-
=
Q(
n
o
"325
5
C5y2o,
g*s23
2.5
hu
0
L
W';.?g
5
ga
-.-
~2
a',
:
u
3
.=
.S!
53,~;;
C
5
$
2
-2
am-
a
-
,:era
-a,
ES
zg
0
2.a
tf
u't
uxg
ZL-
