ArticlePDF Available

The development of competence-related and Motivational beliefs: An investigation of similarity and influence among friends

Authors:

Abstract

This research examined the degree to which children's achievement-related beliefs could be predicted from their friends' beliefs, both concurrently and over time. For 3 semesters, 4th-, 5th-, and 6th-grade students (N = 929) completed measures of their competence-related beliefs, motivational beliefs, and friendship choices. Concurrent analyses indicated that friends showed consistent, albeit modest, similarities with regard to their self-perceptions of competence, academic standards, importance of meeting standards, and preference for challenge. During the academic year, friends appeared influential with regard to children's ability attributions for success and the importance they placed on meeting academic standards. Over a grade-level transition, friends appeared influential with regard to children's ability attributions for failure. Overall, associations were stronger among reciprocated than among unilateral friends. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The Development of Competence-Related and Motivational Beliefs:
An Investigation of Similarity and Influence Among Friends
Ellen Rydell Altermatt and Eva M. Pomerantz
University of Illinois at Urbana–Champaign
This research examined the degree to which children’s achievement-related beliefs could be predicted
from their friends’ beliefs, both concurrently and over time. For 3 semesters, 4th-, 5th-, and 6th-grade
students (N 929) completed measures of their competence-related beliefs, motivational beliefs, and
friendship choices. Concurrent analyses indicated that friends showed consistent, albeit modest, simi-
larities with regard to their self-perceptions of competence, academic standards, importance of meeting
standards, and preference for challenge. During the academic year, friends appeared influential with
regard to children’s ability attributions for success and the importance they placed on meeting academic
standards. Over a grade-level transition, friends appeared influential with regard to children’s ability
attributions for failure. Overall, associations were stronger among reciprocated than among unilateral
friends.
Research suggests that only 25% of the variability in children’s
achievement outcomes can be accounted for by their scores on
tests of intelligence (Neisser et al., 1996). A likely source of some
of the additional variability is the specific achievement-related
beliefs that children bring to the learning context: How talented am
I in science? Am I a poor reader because I’m not trying hard
enough, or am I just not very smart? How well do I want to
perform in math and how hard will I strive to meet this standard?
Indeed, considerable empirical evidence exists to support the claim
that children’s beliefs about their intellectual competencies and
views about the importance of school success have a powerful
impact on their academic behaviors and outcomes (see Eccles,
Wigfield, & Schiefele, 1998, for a review).
In exploring the development of children’s achievement-related
beliefs, researchers have begun to focus on how these beliefs are
socialized by the significant others in children’s lives. Much of this
research has been directed toward parents and teachers as the key
socializers (e.g., Eccles Parsons, Kaczala, & Meece, 1982; Frome
& Eccles, 1998; Grolnick & Ryan, 1989; Skinner & Belmont,
1993), while neglecting the role that peers may serve in influenc-
ing children’s beliefs (see Eccles et al., 1998). This lack of atten-
tion is surprising given that the amount of time children spend with
peers increases dramatically once they begin school (e.g., Ellis,
Rogoff, & Cromer, 1981) and may account for more time than that
spent with parents, siblings, or other agents of socialization as
children approach adolescence (e.g., Larson & Richards, 1991;
Medrich, Rosen, Rubin, & Buckley, 1982). Peers have, moreover,
been recognized as an important source of influence in other
domains, including adolescent delinquency and drug use (e.g.,
Kandel, 1973; Vitaro, Tremblay, Kerr, Pagani, & Bukowski,
1997).
The purpose of the present study was to address this gap in the
achievement socialization literature. Our central goal was to ex-
amine the degree to which children and their friends hold concor-
dant achievement-related beliefs and the extent to which these
similarities can be explained by friends’ apparent influence on
children’s beliefs over time. We examined these relations for both
reciprocated and unilateral friendship dyads over the course of
three semesters. Our focus was on two classes of psychological
constructs that have been identified as important predictors of
children’s academic behaviors and school success (see Eccles et
al., 1998).
First, we examined similarity and influence among friends with
regard to children’s competence-related beliefs—that is, beliefs
related to the question, Can I do this task? Within this category, we
were particularly concerned with children’s evaluations of their
intellectual abilities, both generally (e.g., Am I good at math?) and
in the context of personal success and failure (e.g., Did I fail my
math test because I’m not smart?). Prior research has shown that
children who view themselves as intellectually competent (e.g.,
Frome & Eccles, 1998; Harter, 1982; Phillips, 1984; Pokay &
Blumenfeld, 1990) and who attribute their successes, but not their
failures, to ability (e.g., Ames, 1978; Berndt & Miller, 1990;
Diener & Dweck, 1978, 1980; Stipek & Gralinski, 1991) tend to
perform well academically and to remain engaged in academic
tasks, even in the context of failure. When examining children’s
attributional styles, we chose to focus on children’s ability attri-
butions, given that these attributions have received considerable
Ellen Rydell Altermatt and Eva M. Pomerantz, Department of Psychol-
ogy, University of Illinois at Urbana–Champaign.
This research was supported by National Science Foundation Grant
BCS-9809292 and National Institute of Mental Health and Office for
Research on Women’s Health Grant R01 MH5 7505-02. This article is
based in part on a dissertation submitted by Ellen Rydell Altermatt to the
University of Illinois. Portions of the manuscript were presented at the
2001 meeting of the Society for Research in Child Development in Min-
neapolis, Minnesota.
We are grateful to the children, parents, and staff of the Mattoon and
Charleston school districts for their help with this research. We also thank
Missa Murry Eaton, Gwen Kenney, and Jill Saxon for their assistance in
collecting and managing the data and Larry Hubert, Jasna Jovanovic, Gary
Ladd, and Allison Ryan for their helpful feedback.
Correspondence concerning this article should be addressed to Ellen
Rydell Altermatt, who is now at the College of Education, Michigan State
University, 513D Erickson Hall, East Lansing, Michigan 48824. E-mail:
ealterma@msu.edu
Journal of Educational Psychology Copyright 2003 by the American Psychological Association, Inc.
2003, Vol. 95, No. 1, 111–123 0022-0663/03/$12.00 DOI: 10.1037/0022-0663.95.1.111
111
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
attention in a variety of models of achievement motivation. More-
over, research based on these models provides clear and fairly
consistent evidence that ability attributions for success are adap-
tive, whereas ability attributions for failure are maladaptive (e.g.,
Dweck & Leggett, 1988; Eccles et al., 1998; Weiner, 1985).
Although research has also focused on the implications for chil-
dren of attributing their successes and failures to other factors
(especially effort), the consequences of making these alternative
attributions are not entirely clear (see Berndt & Miller, 1990;
Eccles et al., 1998; Eccles Parsons et al., 1982; Stipek & Gralinski,
1991).
Although competence-related beliefs are thought to play a fun-
damental role in motivating students’ academic behavior (e.g.,
Frome & Eccles, 1998; Harter, 1982; Phillips, 1984), researchers
have suggested that they are not sufficient to ensure school suc-
cess: Even if a child feels confident in her ability to succeed at a
given academic task, she may decide that she does not want to
pursue it (Eccles et al., 1998; Harter, 1981). In keeping with the
notion that children’s decisions about whether to become engaged
in academic tasks are important predictors of success in the aca-
demic context, our second aim was to investigate friends’ similar-
ity and influence with regard to children’s motivational beliefs
that is, beliefs related to the question, Do I want to do this task and
why? Within this category, we focused on the academic standards
that children set for themselves, their beliefs regarding the impor-
tance of meeting these standards, and their self-reported prefer-
ences for challenging academic work. Like positive competence-
related beliefs, positive motivational beliefs have been associated
with desirable achievement outcomes. Specifically, children who
set and strive to meet challenging academic standards and who
view school as important tend to implement effective learning
strategies and to set high expectations for future success (e.g.,
Eccles Parsons, Adler, & Meece, 1984; Harter, 1981; Meece,
Wigfield, & Eccles, 1990; Phillips, 1984; Pintrich & de Groot,
1990; Pokay & Blumenfeld, 1990; Ryan, 2001).
Similarity Among Friends
Interest in the degree to which friends are similar has yielded a
plethora of research. This research has generally focused on de-
mographic variables such as children’s gender, ethnicity, and
religion (e.g., Hamm, 2000; Kandel, 1978b; Tuma & Hallinan,
1979) and on behavioral indices, including children’s engagement
in aggressive and prosocial behaviors (e.g., Haselager, Hartup, van
Lieshout, & Riksen-Walraven, 1998; Kupersmidt, DeRosier, &
Patterson, 1995) and adolescents’ use of alcohol, drugs, and cig-
arettes (e.g., Fisher & Bauman, 1988; Hamm, 2000; Kandel,
1978a). With regard to these constructs, friends generally appear
more similar than nonfriends.
Although there is considerable evidence to suggest that similar-
ities in attitudes, ideals, and values are associated with interper-
sonal attraction among adults (e.g., Byrne & Nelson, 1965), the
degree to which children’s friendships are characterized by con-
cordances on these types of abstract, psychological characteristics
has received relatively little attention. Still, evidence is beginning
to emerge that children’s friendships are based, in part, on psy-
chological similarity. In one recent study, Haselager et al. (1998)
reported small to moderate associations between friends on mea-
sures of shyness and depression. In another, Kurdek and Krile
(1982) reported moderate to strong correlations between recipro-
cated friends on measures of interpersonal understanding and
self-perceptions of social competence.
Similar evidence for friends’ psychological similarity has not
yet been established in the academic domain. Although a host of
studies have documented moderate correlations between friends on
measures of achievement outcomes such as report card grades and
standardized test scores (e.g., Berndt & Keefe, 1995; Kandel,
1978b; see Ide, Parkerson, Haertel, & Walberg, 1981, for a review)
and at least one has examined concordances between close friends
on a behavioral measure of classroom involvement and disruption
(Berndt & Keefe, 1995), very little research has examined simi-
larities among close friends with regard to achievement-related
beliefs. The exceptions to this general rule are studies that have
reported moderate correlations between friends on educational and
occupational aspirations (e.g., Haller & Butterworth, 1960; Halli-
nan & Williams, 1990; Hanks & Eckland, 1976; Kandel, 1978a;
see Ide et al., 1981, for a review). Although important, research
that has examined concordances between friends with regard to
aspirations is limited in its focus on a single achievement belief.
Moreover, this research has almost exclusively relied on high
school samples.
Influence Among Friends
Although establishing the degree to which friendships are char-
acterized by psychological similarity is important in its own right,
the practice of examining similarities among friends has been
undertaken, in large part, to support claims that peers are influen-
tial agents of socialization. This practice has, however, received
considerable criticism, as concurrent measures of similarity fail to
distinguish between the processes of selection (i.e., individuals
who are similar in certain attributes purposefully select each other
as friends) and those of influence (i.e., individuals are socialized by
their friends, thereby becoming more similar to them; see Ennett &
Bauman, 1994). Concretely, friends may be similar not because
they have influenced each other but because their friendships were
formed on the basis of attraction due to similarity in the first place
(Cohen, 1983; Epstein, 1983; Hallinan, 1983; Kandel, 1978a). In
an attempt to sort out the relative weight of the processes of
selection and influence in explaining similarity among friends,
most recent work on peer socialization has turned to short-term
longitudinal analyses. In general, this methodology has relied on
regression analyses to predict changes in children’s attitudes and
behavior over time from the initial characteristics of children’s
friends (Kindermann, McCollam, & Gibson, 1996).
To date, only a handful of studies have examined peer influence
in the academic domain. Results are promising, however, and
suggest that friends not only perform in similar ways academically
but that changes in children’s academic outcomes (e.g., report card
grades, standardized test scores) and educational aspirations can be
predicted from those of friends and peer group members (e.g.,
Epstein, 1983; Kandel, 1978a). Recent research has extended these
findings to behavioral indices of achievement motivation, specif-
ically children’s classroom involvement and disruption. In a study
of fourth- and fifth-grade students, Kindermann (1993) reported
that changes in a child’s classroom involvement (e.g., diligent
attention to academic tasks) at the end of the school year could be
predicted from the average classroom involvement of the members
of the clique to which the student belonged at the beginning of the
school year. Similar results are reported by Berndt and Keefe
112
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
(1995) for seventh- and eighth-grade students. Specifically, chil-
dren who perceived their group of friends to be involved (e.g.,
frequently participating in class discussions) and nondisruptive
(e.g., infrequently misbehaving in class) at the beginning of the
year became more involved and less disruptive themselves over
the course of the academic year.
Whether friends play a similar role in influencing changes in
children’s achievement-related beliefs remains unclear. To date,
research has examined only a few achievement-related beliefs and
has yielded a complex set of results. In one study, Ryan (2001)
predicted seventh graders’ motivational beliefs in the spring from
those of their selected peer group (i.e., their clique) in the fall.
Changes in children’s liking and enjoyment of school were posi-
tively related to peer group members’ beliefs earlier in the aca-
demic year. Peer group beliefs were not, however, found to sig-
nificantly predict changes in the extent to which children viewed
school as useful (i.e., as having utility value) and appeared only
marginally influential with regard to children’s expectations for
academic success. In another study, Berndt, Hawkins, and Jiao
(1999) found that although adolescents’ evaluations of their cog-
nitive competence in the spring could be predicted from friends’
perceptions in the fall, the relation was negative: Children who
selected friends who felt confident in their academic capabilities
early in the school year came to feel relatively less competent by
year’s end.
Although these studies offer insight into friends’ influence, the
number of beliefs so far investigated is quite limited. Moreover,
findings have been both inconsistent (i.e., influence has been
demonstrated for some variables but not for others) and potentially
paradoxical (i.e., with some initial evidence that students may
become less similar to their friends over time). Together, these
results suggest that a good deal remains to be learned about
friends’ influence on children’s achievement-related beliefs.
Children’s Friendships
An important issue of discussion among peer socialization re-
searchers is how children’s friendships are best measured and
whether children’s friends or the larger peer group are likely to
exert the most influence (see Rubin, Bukowski, & Parker, 1998).
In reviewing this literature, Sage and Kindermann (1999) sug-
gested that although the peer group has been shown to influence
academic behaviors that are readily observable by classmates (e.g.,
children’s engagement in classroom activities), the intimate nature
of children’s friendships renders them especially powerful social-
ization contexts for children’s private achievement beliefs. Given
that many of the achievement-related beliefs examined in the
present study are both private (e.g., tapping children’s personal
perceptions of their abilities and standards for academic success)
and sensitive (e.g., assessing children’s reactions to school fail-
ures), we chose to investigate similarity and influence among close
friends. We were especially interested in examining shared beliefs
between children and their very best friends, paying particular
attention to whether these friendships were reciprocated (i.e., nom-
inations were mutual) or unilateral (i.e., nominations were nonre-
ciprocated). This approach allowed us to explore the possibility
that reciprocated friendships would yield stronger indices of sim-
ilarity and influence because of the heightened levels of intimacy
and trust that have been found to characterize this relationship
(Hartup & Stevens, 1997; Newcomb & Bagwell, 1995). Stronger
associations among reciprocated than among unilateral friends
would also offer some insight into the mechanisms of friends’
influence, suggesting that the heightened levels of intimate discus-
sion, information exchange, and advice-giving that characterize
these relationships may play a prominent role.
The Role of Temporal Context
An important task confronting achievement socialization re-
searchers is to document not only whether parents, teachers, and
peers influence children’s school attitudes and outcomes but to
explore how changing social contexts impact the socialization
process (see Eccles, Midgley, & Adler, 1984; Higgins & Parsons,
1983, for reviews). To date, most peer socialization research has
focused on friend and peer group influences during the course of
a single academic year. This approach is logistically simpler than
examining influence as children move from one grade level to the
next. In addition, it might be expected to yield stronger indices of
influence, as children’s friendships are less likely to be disrupted
by changing classroom assignments (Neckerman, 1996) and the
mechanisms of influence (e.g., modeling, discussion, shared expe-
riences) remain uninterrupted by an intervening summer break. In
the only other study to examine friends’ influence across a grade-
level transition, Berndt et al. (1999) provided support for the
relative strength of friends’ influence during the academic year as
compared with the strength over the course of a transition. Spe-
cifically, although changes in children’s perceptions of social and
cognitive competence at the end of the school year were predicted
by friends’ perceptions at the beginning of the year, neither set of
findings was replicated over a grade-level transition (i.e., from the
spring of students’ sixth-grade year to the fall of their seventh-
grade year). We expected that similar findings would emerge in the
present study. At the same time, the opposite prediction also seems
plausible. Specifically, because children’s beliefs are likely to be
less stable over the course of a grade-level transition than during
the course of a single academic year, there may be greater oppor-
tunity for friends to play a role in influencing the direction of the
change during the transition time period.
Overview of the Present Research
Recent research suggests a growing interest in examining peers
as socializers of children’s achievement attitudes, behaviors, and
outcomes. Despite new evidence that friends are both similar and
influential with regard to children’s academic behaviors and be-
liefs, there are still significant gaps in our knowledge. The present
research examined friends’ similarity and influence in the aca-
demic domain among a large sample (N 929) of fourth-, fifth-,
and sixth-grade students and, in so doing, extended prior research
in three key ways.
First, the present study investigated both similarity and influ-
ence among close friends with regard to children’s academic
performance and a number of important competence-related be-
liefs (i.e., ability attributions for success and failure) and motiva-
tional beliefs (i.e., level of standards, importance of meeting stan-
dards, and preference for challenge) that have not been previously
examined. Although friends’ influence on children’s self-
perceptions of competence has been investigated in a prior study
(see Berndt et al., 1999), the small sample size utilized and
unanticipated findings (i.e., indicating potential dissimilarity over
113
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
time) suggest that replication of these results is needed. We were
especially interested in examining competence-related and moti-
vational beliefs given their prominence in a variety of theories of
achievement motivation (e.g., Deci & Ryan, 1985; Weiner, 1985;
Zimmerman, 1989; see Eccles et al., 1998, for a review) and, also,
given that previous research provides clear evidence of their im-
port in influencing children’s achievement outcomes (e.g., Frome
& Eccles, 1998; Harter, 1981; Pokay & Blumenfeld, 1990; Stipek
& Gralinski, 1991). Second, the present research examined
whether indices of similarity and influence differ for reciprocated
and unilateral friendship dyads. We anticipated that associations
would be stronger among reciprocated than unilateral dyads, in
part because of the heightened levels of intimacy and trust that
characterize this relationship (Newcomb & Bagwell, 1995; Saxon,
1996). Third, the present study investigated associations between
children’s academic performance and beliefs and friends’ aca-
demic performance and beliefs at different points during the aca-
demic year. This approach is important in determining whether
indices of similarity and influence are stable and replicable across
time and in determining whether changes in children’s character-
istics are better predicted by friends’ characteristics during the
course of a single academic year or over a grade-level transition.
Method
Participants
The data used in the present study were collected as part of the Univer-
sity of Illinois Self-Evaluation Project, a research program aimed at ex-
amining the development of children’s achievement-related beliefs (see
Pomerantz & Saxon, 2001; Pomerantz, Saxon, & Oishi, 2000). Participants
were 929 children (463 boys, 466 girls) in the fourth (n 270; mean
age 9.25 years), fifth (n 449; mean age 10.25 years), and sixth (n
210; mean age 11.25 years) grades.
1
The majority of participants were
European American (95%), but the sample also included African Ameri-
cans (4%) and other minorities (1%). Approximately 20% of the sample
participated in free- or reduced-lunch programs. All students attended one
of two school districts in the Midwest. Within these school districts, nine
schools representing 58 classrooms participated. Letters describing the
study were sent home to parents. Parents were asked to contact the school
or investigators if they did not want their children to participate. Only 4%
of parents did not permit their children to participate.
Procedure
Children participated in three waves of data collection approximately 6
months apart. The attrition rate was 11% (primarily because of children
moving out of the school district), yielding the sample described previ-
ously. The first wave took place during the spring (Wave 1), the second
wave took place during the following fall (Wave 2), and the third wave
took place during the following spring (Wave 3). Hence, by the second
wave, the fourth graders (fourth-grade cohort) were in fifth grade, the fifth
graders (fifth-grade cohort) were in sixth grade, and the sixth graders
(sixth-grade cohort) were in seventh grade. Children remained in these
same grades during the third wave. In addition to the grade-level transition
that all children experienced after the first wave of data collection, children
in the sixth-grade cohort made a school transition (i.e., from elementary to
middle school) at this time. Seventy-one percent of fourth graders also
made a school transition at this time.
2
At each wave, children took part in
two 45-min classroom sessions during which questionnaires were admin-
istered. A trained research assistant read each item to children who marked
their responses on their own.
Measures
Table 1 provides an overview of the measures, including the number of
items, the potential range, the internal reliability, and the stability over the
three waves. Means and standard deviations for each of the measures are
presented in Table 2. Bivariate correlations between the measures are
presented in Table 3.
Academic Performance
Children’s report card grades were obtained in six academic subject
areas (English, math, reading, science, social studies, and spelling).
3
Letter
grades were converted to numerical values, ranging from 0 (F) to 12 (A).
The mean of the grades in the six subjects across the two academic quarters
overlapping with each wave was used as an index of academic performance
for each wave.
Achievement-Related Beliefs
Competence-Related Beliefs
Self-perceptions of competence. Self-perceptions of competence were
assessed following Wigfield, Eccles, Mac Iver, Reuman, and Midgley
(1991). Children rated how skilled they were in each of the six subjects for
which they received grades (e.g., “How good at reading are you?”) and
their relative position in their class (e.g., “If you were to rank all of the
students in your class from the worst to the best in reading, where would
you put yourself?”). The mean of the two items across the six subjects was
used, with higher numbers indicating more positive perceptions.
Ability attributions for success and failure. Following past research
(e.g., Eccles Parsons et al., 1984; Eccles Parsons, Meece, Adler, & Kac-
zala, 1982; Pomerantz & Saxon, 2001), children were asked to imagine that
they experienced an academic success (i.e., did well in school) or failure
(i.e., did poorly in school). Children were provided with four possible
causes for their imagined performance: ability (e.g., “You are smart”),
effort (e.g., “You tried very hard”), luck (e.g., “You were lucky”), and task
(e.g., “The work was easy”). Children indicated the first, second, and third
best reason for their performance. The ranks were reverse scored so that
higher numbers indicated that children were more likely to choose an
ability attribution for their performance relative to other attributions.
1
Only students who completed questionnaires at all three waves were
included in the analyses. Three additional children were excluded because,
although they completed most of the questionnaire measures at each of the
three waves, they failed to complete the friendship assessment measures at
any of the three waves.
2
To determine whether indices of similarity and influence differed for
children who experienced a school transition from those who experienced
only a grade-level transition, “school transition” was entered as a dummy
variable in our concurrent and longitudinal analyses, and interactions
between school transition and friends’ characteristics in predicting chil-
dren’s characteristics were examined. Cohort (and its interaction with
friends’ characteristics) was entered as a control variable in these analyses
to ensure that any differences were not due to children’s grade level. Few
significant interactions emerged. Moreover, the direction of the effects was
inconsistent. Future work will be important in examining the effects of
school transitions versus grade-level transitions on friends’ similarity and
influence.
3
At Waves 2 and 3, children in the sixth-grade cohort received grades
in four academic subjects only (English, math, science, and social studies).
For these children, report card grades and self-perceptions of competence
are based on 8 rather than 12 items. In addition, level and importance of
standards are based on 4 rather than 6 items.
114
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
Motivational Beliefs
Level of standards. A scale measuring the level of children’s standards
for academic performance was developed for this and related studies (see
Nicholls, 1975; Phillips, 1984, for similar, single-item measures). Specif-
ically, children were asked to indicate the grade they deemed to be
personally acceptable in each of the subjects for which they received
grades (e.g., “What is the lowest grade that would be OK for you to get in
reading?”). Letter grades were converted to numerical values, ranging
from 0 (D) to 9 (A). The mean of children’s standards across the six
subject areas was computed, with higher numbers indicating a higher level
of standards. In addition to being face valid, this measure is internally
reliable (
s .87; see Table 1), stable over time (rs .62, ps .01; see
Table 1), and is related in predictable ways to children’s academic perfor-
mance and achievement-related beliefs (see Table 3).
Importance of meeting standards. A scale measuring the importance
children assigned to meeting academic standards was developed for the
purpose of this and related studies. Specifically, after children were asked
to indicate the grade they deemed to be personally acceptable in each of the
six subjects for which they received grades, they were asked to indicate
how important it was to obtain that grade (e.g., “How important is it to you
to get a grade at least this high in reading?”). The mean of children’s
responses across the six subject areas was used, with higher numbers
indicating greater importance. In addition to being face valid, this measure
is internally reliable (
s .85; see Table 1), reasonably stable over time
(rs .51, ps .01; see Table 1), and has been shown to be predictive of
children’s level of depression, anxiety, and self-esteem (Pomerantz et al.,
2000; see also Table 3).
Preference for challenge. Preference for difficult over easy academic
work was assessed with five of the six items of the Preference for Chal-
lenge Subscale of Harter’s (1981) Intrinsic Orientation in the Classroom
Scale. Children were presented with descriptions of two types of children
differing in the type of academic work they prefer (e.g., “Some kids would
rather just learn what they have to in school, but other kids would rather
learn about as much as they can”). Children decided which they were more
like and indicated if the given statement was really or sort of true for them.
The mean of the five items was used, with higher numbers indicating
greater preference for challenge.
Friendship Assessment
To gather information on children’s friendships, we gave children a class
roster and asked them to circle the names of their three “best friends.”
4
Children were instructed that they could circle fewer than three names but
not more. From this list, children were asked to select their single, very best
friend. The percentage of children selecting a very best friend varied from
74% to 92% across the three waves of data collection. Two subsets of
children who selected a very best friend were identified. A child was
deemed to have a reciprocated very best friendship if his or her nominated
very best friend also selected the child as a very best friend. Otherwise, the
child’s self-reported very best friendship was determined to be unilateral
(i.e., nonreciprocated). The percentage of children whose very best friend-
ship was reciprocated varied from 38% to 42% across the three waves of
data collection. Children’s scores on each of the academic outcome and
achievement-related belief measures were matched to those of their single,
very best friend. It is important to note that these analyses were also
conducted by matching children with their very best friend using a less
stringent criterion for reciprocation (i.e., a child’s selected very best friend
nominated the child as one of his or her three best friends) and by matching
children with their three best friends. The results obtained from these
analyses were, in general, similar to or weaker than those reported here.
Because the percentage of children selecting a very best friend differed
from wave to wave, the final sample for all analyses using friendship data
ranged from 678 (285 reciprocated, 393 unilateral) to 851 (359 recipro-
cated, 492 unilateral) across the three waves of data collection. The sample
for each analysis varies somewhat because children or friends sometimes
failed to complete the majority of items for one of the measures included
in the analysis.
4
At Waves 2 and 3, children in the sixth-grade cohort met with different
teachers and classmates for each subject area. For these students, friendship
nominations were limited to children’s English classrooms. One concern
with this procedure is that similarity and influence may be attenuated
among seventh-grade students because they are forced to select friends
from a subset of students that may not include their closest friends. This
does not appear to be the case, however. Specifically, estimates of simi-
larity and influence generally decreased rather than increased when
seventh-grade students were excluded from the analyses. A second possible
consequence of this procedure is that it may produce heightened estimates
of similarity and influence among older students because of the greater
homogeneity in student performance and achievement-related beliefs
among seventh graders because of tracking. Again, however, this does not
appear to be the case. Specifically, at Waves 2 and 3, standard deviations
for academic performance and achievement beliefs did not differ signifi-
cantly in seventh-grade classrooms from those obtained in fifth- and
sixth-grade classrooms.
Table 1
Measure Characteristics
Variable
Number
of items
Potential
range
Internal reliability Stability over time**
Spring
(W1)
Fall
(W2)
Spring
(W3)
W1 to
W2
W2 to
W3
W1 to
W3
Academic performance
Report card grades 12 0–12 .95 .95 .95 .82 .92 .78
Competence-related beliefs
Self-perceptions of competence 12 1–7 .88 .89 .90 .70 .76 .63
Ability attributions for success 1 1–4 .43 .54 .39
Ability attributions for failure 1 1–4 .37 .43 .25
Motivational beliefs
Level of standards 60–9 .87 .92 .91 .70 .71 .62
Importance of meeting standards 6 1–7 .85 .89 .91 .57 .65 .51
Preference for challenge 51–4 .79 .79 .77 .57 .62 .47
Note. W wave.
** p .01.
115
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
Results
Two sets of central analyses were conducted. In the first set, we
examined concurrent relations between children’s academic per-
formance, competence-related beliefs, and motivational beliefs
and those of their very best friends. In the second set, we examined
these relations longitudinally to determine whether children’s per-
formance and beliefs were predicted by friends’ performance and
beliefs over time. In describing the results of both sets of analyses,
we highlight instances in which the findings varied by friendship
type. In addition to these central analyses, two sets of supplemen-
tary analyses were conducted to explore alternative explanations
for the findings obtained in our central analyses.
Central Analyses
Examining Similarities Among Friends
A series of hierarchical regression analyses was conducted to
determine whether children and their very best friends performed
similarly in school and held concordant achievement-related be-
liefs. Each academic outcome or achievement-related belief vari-
able (e.g., children’s standards at Wave 1) was entered as the
dependent variable in a separate analysis. At Step 1, friends’
characteristics at the same wave (e.g., friends’ standards at Wave
1) were entered. If, at this step, the analyses reveal that friends
characteristics are a significant predictor of children’s character-
istics, it suggests that friends are similar with regard to their
academic outcomes or achievement-related beliefs. The standard-
ized beta coefficients from this step of our analyses are presented
in Table 4.
To determine whether relations between children’s characteris-
tics and friends’ characteristics differ by friendship type, a dummy
variable representing friendship type (i.e., reciprocated versus uni-
lateral) was entered at the second step in each analysis. At the third
step, a term that reflected the interaction of friendship type and
friends’ characteristics was entered. If significant, the interaction
term suggests that the degree of friends’ similarity differs by
friendship type. In such cases, correlation coefficients were calcu-
lated and are reported separately for reciprocated and unilateral
friends.
Academic Performance
As demonstrated in prior research, children and their very best
friends were similar with regard to their report card grades. Asso-
ciations were significant at all three waves of data collection,
s
.27 to .34, ts(671–729) 7.30, ps .01. A significant friendship
type interaction emerged at the fall (Wave 2) time period,
.10,
Table 2
Means and Standard Deviations
Variable
Spring (W1) Fall (W2) Spring (W3)
MSDMSDMSD
Academic performance
Report card grades 7.85 2.31 7.90 2.52 7.79 2.58
Competence-related beliefs
Self-perceptions of competence 5.06 1.00 5.14 1.03 5.17 1.06
Ability attributions for success 2.88 1.09 3.07 1.08 3.22 1.08
Ability attributions for failure 1.44 0.84 1.48 0.85 1.53 0.89
Motivational beliefs
Level of standards 4.74 1.72 4.51 1.88 4.43 1.94
Importance of meeting standards 5.79 1.18 5.75 1.19 5.61 1.33
Preference for challenge 2.62 0.76 2.61 0.72 2.58 0.69
Note. W wave.
Table 3
Correlations Between Academic Performance, Competence-Related Beliefs,
and Motivational Beliefs
Variable** 1234567
Academic performance
1. Report card grades
Competence-related beliefs
2. Self-perceptions of competence .60
3. Ability attributions for success .43 .53
4. Ability attributions for failure .23 .31 .29
Motivational beliefs
5. Level of standards .48 .57 .34 .16
6. Importance of meeting standards .20 .41 .20 .14 .29
7. Preference for challenge .30 .43 .34 .16 .33 .20
Note. Values presented are correlations and p values averaged across the three waves of data collection.
** p .01.
116
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
t(727) 2.26, p .05. Follow-up analyses indicated that recip-
rocated friends were more similar with regard to their report card
grades (r .46, p .01) than were unilateral friends (r .27, p
.01).
Competence-Related Beliefs
Among the competence-related beliefs, very best friends were
most similar with regard to their self-perceptions of competence.
Associations were significant across all three waves of data col-
lection,
s .18 to .22, ts(593–823) 4.50, ps .01. At the
spring (Wave 1) semester, a significant friendship type interaction
emerged,
.14, t(821) 3.02, p .01. Follow-up analyses
indicated stronger associations for reciprocated (r .34, p .01)
than for unilateral (r .13, p .01) friends. A marginal effect in
the same direction emerged during the fall (Wave 2) semester,
.08, t(654) 1.68, p .10. This interaction again indicated a
tendency for reciprocated friends to be more similar with regard to
their self-perceptions of competence (r .27, p .01) than
unilateral friends (r .13, p .01).
Friends appeared only modestly similar in the degree to which
they attributed their successes and failures to their intellectual
competencies. For success attributions, a significant association
emerged only at the spring (Wave 1) semester,
.20,
t(829) 5.89, p .01. For failure attributions, associations were
significant but weak at both the fall (Wave 2) and spring (Wave 3)
semesters,
s .12 and .10, ts(666 and 636) 2.60, ps .01.
Only one significant friendship type difference emerged,
.13,
t(827) 2.94, p .01. Specifically, reciprocated friends were
significantly more similar (r .32, p .01) than unilateral friends
(r .12, p .01) with regard to attributions for success to ability
during the spring (Wave 1) semester.
Motivational Beliefs
Very best friends showed small to moderate similarities with
regard to their motivational beliefs. Here, the strongest effects
across the three waves of data collection emerged for the level of
standards construct,
s .16 to .18, ts(595–825) 4.30, ps
.01. A significant friendship type interaction emerged at the fall
(Wave 2) semester,
.10, t(652) 2.06, p .05, indicating that
reciprocated friends (r .28, p .01) were more similar with
regard to the level of their academic standards than were unilateral
friends (r .12, p .01).
Concordances between children and their friends in the impor-
tance they assigned to meeting their academic standards were
significant but weak at both the Wave 1 and Wave 3 spring
semesters,
s .07 and .11, ts(825 and 592) 1.95, ps .05.
These effects were, however, qualified by significant friendship
type interactions,
s .12, ts(823 and 590) 2.50, ps .01. In
both cases, reciprocated friends appeared similar (rs .16 and .25,
ps .01), whereas unilateral friends did not (rs –.01, ns). A
marginal friendship type interaction also emerged at the fall (Wave
2) semester,
.09, t(649) 1.81, p .10. Again, there was a
tendency for reciprocated (r .14, p .05) but not unilateral
friends (r –.01, ns) to appear similar.
Friends also appeared somewhat similar across the three waves
in their preference for academic challenge,
s .13 to .14,
ts(602–816) 3.20, ps .01. Qualifying these effects, however,
were two significant friendship type interactions,
s .11, ts(814
and 600) 2.40, ps .01. Specifically, at both the spring (Wave
1) and fall (Wave 2) semesters, reciprocated friends appeared
similar in their preference for challenge ratings (rs .23 and .25,
ps .01), whereas unilateral friends did not (rs .07 and .05, ns).
Summary
Similarity coefficients were, in all cases, positive in valence and,
in general, small to moderate in size. Consistent with prior work on
friends’ psychological similarity in the social domain (e.g., Kurdek
& Krile, 1982), there was some tendency for stronger associations
to appear among reciprocated than among unilateral very best
friend dyads. All significant differences were, in fact, in this
direction.
Table 4
Predicting Children’s Academic Performance and Achievement-Related Beliefs From Their
Friends’ Performance and Beliefs at the Same Wave
Variable Spring (W1) Fall (W2) Spring (W3)
Academic performance
Report card grades .27** .34**
,a
.34**
Competence-related beliefs
Self-perceptions of competence .22**
,a
.18**
,b
.18**
Ability attributions for success .20**
,a
.06† .07†
Ability attributions for failure .03 .12** .10**
Motivational beliefs
Level of standards .16** .18**
,a
.18**
Importance of meeting standards .07*
,a
.05
b
.11**
,a
Preference for challenge .14**
,a
.13**
,a
.13**
Note. Values represent the standardized regression coefficients from hierarchical regression analyses predict-
ing children’s characteristics from friends’ characteristics at the same wave (W).
a
Denotes a significant (p .05) interaction between friends’ characteristics and friendship type in predicting
children’s characteristics.
b
Denotes a marginal (p .10) interaction between friends’ characteristics and
friendship type in predicting children’s characteristics. Where significant or marginal interactions emerge,
reciprocated friends appear more similar than unilateral friends.
p .10. * p .05. ** p .01.
117
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
Examining Friends’ Influence
Our second set of analyses was longitudinal and was designed to
examine the degree to which friends appear to influence children’s
achievement-related beliefs. Specifically, a series of hierarchical
multiple regression analyses was conducted to predict changes in
students’ academic performance and achievement-related beliefs
from the performance and beliefs of children’s friends at an earlier
wave. Relations were assessed over three time periods: spring
(Wave 1) to fall (Wave 2), fall (Wave 2) to spring (Wave 3), and
spring (Wave 1) to spring (Wave 3). Although each of the first two
time periods represents an approximately 6-month lag, the fall
(Wave 2) to spring (Wave 3) interval examines influence as it
occurs during a single academic year, whereas the spring (Wave 1)
to fall (Wave 2) interval examines influence as it occurs across a
grade-level transition (e.g., fourth graders became fifth graders).
The spring (Wave 1) to spring (Wave 3) time period allowed us to
investigate influence as it occurs over a relatively longer, 1-year
time period that also included a grade-level transition.
Following Berndt and Keefe (1995), each academic outcome or
achievement-related belief variable (e.g., children’s standards at
Wave 2) was entered as the dependent variable in a separate
hierarchical regression analysis. At Step 1, children’s characteris-
tics at the earlier wave (e.g., children’s standards at Wave 1) were
entered. This step is important, given the stability of the measures
over time (see Table 1), and ensures that significant findings
reflect a relation between children’s and friends’ characteristics
over time, rather than a simple, concurrent association. At Step 2,
friends’ characteristics at the earlier wave (e.g., friends’ grades at
Wave 1) were entered. If, at this step, the analyses reveal that
friends’ characteristics at the earlier wave are a significant predic-
tor of children’s characteristics at the later wave, it suggests that
friends may play a role in influencing changes in children’s aca-
demic outcomes or achievement beliefs over time. Table 5 presents
the standardized beta coefficients from this second step of our
analyses.
To test for friendship type differences in friends’ influence, we
entered a dummy variable representing friendship type at the third
step in each analysis. At the fourth step, we entered a term that
reflected the interaction of friendship type and friends’ character-
istics. If significant, the interaction term suggests that friends’
influence varies by friendship type. To clarify these interactions,
we ran separate regression analyses for reciprocated and unilateral
friends (i.e., predicting children’s characteristics from friends’
characteristics, controlling for children’s earlier characteristics).
The results of these analyses are reported in the text.
Academic Performance
Consistent with prior work, friends appeared influential with
regard to children’s academic performance. At all three waves,
changes in children’s report card grades were predicted by the
previous academic performance of their very best friend,
s .04
to .06, ts(694–728) 2.10, ps .05. At the fall (Wave 2) to
spring (Wave 3) time period, this relation was qualified by a
significant interaction between friends’ academic performance and
friendship type,
.04, t(726) 1.95, p .05. To clarify this
interaction, we conducted regression analyses separately for recip-
rocated and unilateral friends. These analyses revealed that, during
the course of the academic year, changes in children’s academic
performance were significantly predicted by the prior academic
performance of reciprocated,
.10, t(275) 3.57, p .01, but
not unilateral,
.02, t(449) 0.88, ns, friends.
Competence-Related Beliefs
During the course of the academic year, very best friends
appeared influential with regard to children’s ability attributions
for success. Specifically, from fall (Wave 2) to spring (Wave 3),
changes in children’s attributions for success to ability were pos-
itively related to the attributional styles of friends at the earlier
wave,
.13, t(649) 3.91, p .01. Friends also appeared
Table 5
Predicting Children’s Academic Performance and Achievement-Related Beliefs From
Their Friends’ Performance and Beliefs at an Earlier Wave
Variable
Within the academic year Across a grade-level transition
Fall (W2) to
Spring (W3)
Spring (W1) to
Fall (W2)
Spring (W1) to
Spring (W3)
Academic performance
Report card grades .04**
,a
.05* .06**
Competence-related beliefs
Self-perceptions of competence .04† .02 .03
Ability attributions for success .13** .02 .00
Ability attributions for failure .03 .08** .08*
,a
Motivational beliefs
Level of standards .02 .01 .01
Importance of meeting standards .06* .01 .03
Preference for challenge .01 .05† .06†
Note. Values represent the standardized regression coefficients from hierarchical regression analyses predict-
ing children’s characteristics at the later wave from friends’ characteristics at the earlier wave (controlling for
children’s characteristics at the earlier wave). W wave.
a
Denotes a significant (p .05) interaction between friends’ characteristics and friendship type in predicting
children’s characteristics such that reciprocated friends appear influential, whereas unilateral friends do not.
p .10. * p .05. ** p .01.
118
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
influential with regard to children’s ability attributions for failure.
Here, however, significant associations emerged only across the
two time intervals during which children experienced a grade-level
transition. Specifically, during both the spring (Wave 1) to fall
(Wave 2) and spring (Wave 1) to spring (Wave 3) time intervals,
changes in children’s failure attributions to ability were predicted
by the attributional style of their very best friend,
s .08, ts(767
and 780) 2.30, ps .05. During the spring (Wave 1) to spring
(Wave 3) time period, the association between friends’ attributions
and children’s attributions was qualified by a significant friendship
type interaction,
.09, t(778) 1.94, p .05. Analyses
conducted separately by friendship type revealed that changes in
children’s tendency to attribute failure to ability were predictable
from the attributional styles of reciprocated,
.17,
t(328) 3.16, p .01, but not unilateral,
.02, t(449) 0.41,
ns, friends.
More modest evidence was found for friends’ influence on
children’s self-perceptions of competence. Specifically, during the
course of a single academic year, changes in children’s compe-
tence perceptions were marginally predicted by the competence
perceptions of their very best friends,
.04, t(597) 1.61, p
.10.
Motivational Beliefs
Some modest evidence also emerged for friends’ influence on
children’s motivational beliefs. During the course of the academic
year, changes in children’s ratings of the importance of meeting
academic standards were significantly predicted by the earlier
ratings given by their very best friends,
.06, t(592) 1.99,
p .05. In addition, during both the spring (Wave 1) to fall (Wave
2) and spring (Wave 1) to spring (Wave 3) transitions, friends
appeared marginally influential with regard to children’s prefer-
ence for challenge ratings,
s .05, ts(724 and 749) 1.60, ps
.10. Friends did not appear to influence the level of academic
standards that children set,
s –.01, ts(596–762) 1, ns. In
addition, no significant friendship type interactions emerged for
any of the motivational variables.
Summary
Across the three waves of data collection, friends appeared most
influential with regard to children’s report card grades. During the
course of a single academic year, friends also appeared influential
with regard to the degree to which children made ability attribu-
tions for their successes and the extent to which they deemed it
important to meet academic standards. Over the course of both
grade-level transition time periods, friends appeared influential
with regard to children’s ability attributions for failure. More
modest evidence was found for friends’ influence on children’s
self-perceptions of competence and preference for challenge. In
both cases, findings reached only marginal levels of significance.
Only two significant friendship type differences emerged. How-
ever, both were in the predicted direction, with reciprocated
friends appearing more influential than unilateral friends.
Supplementary Analyses
Two sets of supplementary analyses were conducted to examine
alternative explanations for the findings reported thus far. In the
first set, we predicted children’s beliefs from those of randomly
selected, nonfriend classmates. These analyses help to ensure that
findings reflect interpersonal selection rather than classroom se-
lection effects. In the second set, we revisited our longitudinal
analyses to determine whether evidence for friends’ influence may
be stronger among stable (as compared with unstable) friendship
dyads.
Classroom Selection Effects as an Explanation for
Similarity and Influence
Because children were limited to selecting friends from their
classrooms, there is some possibility that our findings could rep-
resent tracking or classroom-selection effects rather than
interpersonal-selection effects. That is, children could appear to be
similar to their friends or influenced by their friends simply be-
cause they are placed into classrooms where even nonfriend class-
mates tend to be of similar levels of ability and hold similar
school-related attitudes. This is, in fact, a potential problem with
the broader peer socialization literature, given that even when
children are permitted to select friends from outside of their own
classrooms, they are more likely than not to select classmates
(Parker & Asher, 1993).
To test this possibility, we matched each child with a randomly
selected nonfriend from that child’s classroom. Across the three
waves of data collection, only two significant concurrent effects
emerged (r .10 for level of standards at spring [Wave 1] and r
–.08 for ability attributions for success at fall [Wave 2]). This is no
more than would be expected by chance. Moreover, one of the
correlations was negative (i.e., suggesting dissimilarity). Similar
results emerged from our longitudinal analyses with randomly
selected nonfriends. Across the three time periods, only one sig-
nificant effect emerged (
–.06 for ability attributions for
success from fall [Wave 2] to spring [Wave 3]). Again, this is no
more than would be expected by chance. Moreover, the beta
coefficient was negative (i.e., indicating that children become less
similar to nonfriends over time).
Influence in Stable Versus Unstable Friendship Dyads
Researchers have often suggested that the stronger and longer
lasting the relationship between children and their friends, the
more likely friends are to exert an influence on children’s beliefs
and behaviors (e.g., Berndt et al., 1999; Epstein, 1983; but see
Berndt & Keefe, 1995; Ryan, 2001; Urberg, Degirmencioglu, &
Pilgrim, 1997). Given this, a possible reason for the rather weak
findings regarding friends’ influence is that evidence of influence
is attenuated when children’s friendships are not stable from wave
to wave. In the present study, very best friendships were generally
unstable over the Wave 1 to Wave 2 and Wave 1 to Wave 3
grade-level transition time periods, with only 10% of students
retaining a very best friend (i.e., only 10% of students who
nominated a very best friend at Wave 1 selected this same student
as one of their three best friends at Wave 2 or Wave 3). In contrast,
77% of students’ very best friendships were stable over the Wave 2
to Wave 3 time period, when no grade-level transition took place.
To test the hypothesis that indices of influence would be strongest
among stable friendship dyads, a set of regression analyses was
conducted. Specifically, as in the analyses examining differences
between reciprocated and unilateral friends, friendship stability
119
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
was entered as a dummy variable in our longitudinal analyses, and
interactions between stability and friends’ characteristics were
examined. These analyses were conducted only during the fall
(Wave 2) to spring (Wave 3) time period, as the small number of
stable friendships over the spring (Wave 1) to fall (Wave 2) and
spring (Wave 1) to spring (Wave 3) time periods (ns 80 and 56,
respectively) precluded meaningful interpretation of the results.
The results of these analyses suggest that friendship stability
may, indeed, foster greater influence. Specifically, significant sta-
bility interactions emerged for report card grades,
.09,
t(726) 3.77, p .01, self-perceptions of competence,
.15,
t(595) 2.90, p .01, and ability attributions for success,
.13, t(649) 2.16, p .05. Follow-up analyses revealed that, in
every case, friends’ characteristics were predictive of a change in
children’s characteristics in stable friendship dyads,
s .08,
ts(435–507) 2.68, ps .01, but not in unstable friendship dyads,
s .02, ts(159–218) 1.50, ns. The relation for the self-
perceptions of competence measure,
.08, t(435) 2.68, p
.01, is particularly important, given that the association reached
only marginal levels of significance when analyses were collapsed
across stable and unstable friendship dyads. It is important to note
that although these findings provide some support for the notion
that influence may be stronger among stable friends, the results
should be interpreted cautiously. Because so many children re-
mained in stable friendships from Wave 2 to Wave 3 (77%) and so
few friendships remained stable from Wave 1 to Wave 2 or from
Wave 1 to Wave 3 (10%), we would expect consistently stronger
influence coefficients at Wave 2 to 3. This was not, however, the
case.
Discussion
A host of studies concerned with achievement motivation sug-
gest that parents and teachers play a key role in influencing
children’s achievement attitudes and academic outcomes (e.g.,
Frome & Eccles, 1998; Skinner & Belmont, 1993; see Eccles et al.,
1998, for a review). The present study extends this body of work
by examining friendships as an additional, and potentially impor-
tant, context for the development of children’s competence-related
and motivational beliefs.
Our results confirm that children’s friendships are based, in part,
on psychological similarity in the academic domain. In addition to
corroborating prior research indicating that friends perform simi-
larly in school, we found modest, yet consistent, concordances
between friends (and, particularly, reciprocated friends) with re-
gard to self-perceptions of competence and each of the motiva-
tional constructs examined. Although concordances between
friends were somewhat less strong and less consistent for chil-
dren’s attributions than for the other achievement-related beliefs,
there is little evidence that children base their friendship selections
more on motivational factors than on competence-related factors
or vice-versa. It appears that both how classmates evaluate them-
selves (e.g., “Can I do this task?”) and the choices they make
regarding the level of engagement they will dedicate to academic
tasks (e.g., “Do I want to do this task and why?”) are important to
children when establishing close relationships with peers.
Evidence for friends’ influence was, as in previous research,
generally modest. However, the associations that were revealed are
important. First, the present study replicates prior research show-
ing that changes in children’s academic performance are predict-
able from the academic performance of children’s friends. It is
important to note that this link held not only during the course of
a single academic year but also over the course of two grade-level
transition time periods. Second, the present study provides evi-
dence that changes in children’s self-perceptions of competence
are positively related to the competence perceptions of children’s
friends. Notably, this finding reached significance only among
stable very best friend dyads. Although this finding should be
replicated in future research, it suggests that the negative associ-
ations reported by Berndt et al. (1999) may have been due, in part,
to the instability of children’s friendships over time (i.e., only 55%
of children’s friendship were stable in Berndt et al.’s, 1999, sam-
ple). Third, the present research provides novel evidence for
friends’ influence on children’s reasoning about the causes of their
academic successes and failures and children’s ratings of the
importance of meeting academic standards. The findings for the
attributional indices are, we think, particularly important, given the
subtlety of these beliefs and their demonstrated usefulness in
predicting children’s achievement outcomes. The attributional
findings should be replicated in future research, particularly given
our single-item measure. It is important to keep in mind, however,
that the lower reliability typically associated with one-item mea-
sures should lead to underestimation of the effects of a variable.
Thus, the present approach likely represents a conservative test of
friends’ influence on children’s ability attributions (Kenny, 1979;
Pedhazur, 1982).
Friends’ apparent influence on children’s competence-related
and motivational beliefs varied across time periods. Specifically,
friends appeared influential with regard to children’s ability attri-
butions for success, self-perceptions of competence, and ratings of
the importance of meeting standards only during the course of a
single academic year. In contrast, evidence for friends’ influence
on children’s ability attributions for failure emerged only during
the course of the two grade-level transition time periods. The first
set of findings (i.e., indicating that friends exert an influence on
children’s beliefs during the academic year but not over the course
of a grade-level transition) is consistent with prior research (Berndt
et al., 1999) and with the notion that influence should be stronger
during this time period because children’s friendships may be
more stable and because the mechanisms of influence are uninter-
rupted by an intervening summer break. The reasons for the second
set of findings (i.e., indicating influence only over the course of a
grade-level transition) are not entirely clear. As noted previously,
one possibility is that, because children’s beliefs are at least
slightly less stable over the course of a transition (mean r .31)
than over the course of a single academic year (r .43), there is
greater room for friends to play a role in influencing the direction
of any change in children’s beliefs about the causes of their
failures. Another possibility is that this finding is related to the
achievement context in which children find themselves. Like the
transition from elementary school to middle school or from middle
school to high school, the transition from one grade level to the
next may pose a number of challenges for students (see Eccles et
al., 1984; Ruble & Seidman, 1996, for reviews). Children are
likely to be confronted with novel standards for performance and
to experience a temporary waning of social support as teachers and
peer groups change. One consequence is that children may be
particularly attentive to lessons learned from friends (particularly
about failure) as they approach this novel social environment.
Further research is needed to examine these and other alternative
120
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
possibilities. Future work will also be important in examining the
effects of school transitions versus grade-level transitions on
friends’ similarity and influence.
A key question to be addressed in future studies of friends’
influence on children’s competence-related and motivational be-
liefs is why friends show at least modest similarities with regard to
each of the beliefs examined in the present study, but strong,
consistent, and pervasive evidence for friends’ influence (particu-
larly with regard to children’s motivational beliefs) has not been
demonstrated. Perhaps the simplest answer to this conundrum is
that friends are not, in fact, very influential and that the similarities
we see are based largely on children’s tendency to select as friends
students who share concordant achievement beliefs. Although we
do not dispute that selection may play a major role in friends’
similarity and that future research should address this possibility,
we believe that other factors may have yielded these results.
First, the processes of influence may not be adequately captured
in the time period examined. One possibility is that friends’ influ-
ence occurs in much shorter time intervals than have previously
been investigated. Specifically, it may be that children modify
their beliefs to conform to those of friends very shortly after
forming a friendship (or even as the friendship is forming), before
assessments of students’ initial characteristics are typically made.
This possibility seems particularly plausible, given the early ado-
lescent samples examined in most studies of peer influence in the
academic domain. There is ample evidence that children are espe-
cially anxious to conform to the behaviors and beliefs of friends at
this time (e.g., Berndt, 1979; Costanzo & Shaw, 1966). Children at
this age also have the capabilities that are necessary to adequately
assess the beliefs of their friends, make the appropriate social
comparisons, and modify their beliefs accordingly (see Rholes,
Newman, & Ruble, 1990). Research that assesses friends’ influ-
ence very early in the school year just as friendship choices are
being made for the first time might prove especially enlightening
in examining this hypothesis. More research on developmental
differences in friends’ influence is also necessary. Although sev-
eral studies have examined age-related differences in children’s
susceptibility to friends’ influence (e.g., Epstein, 1983; Urberg,
Cheng, & Shyu, 1991), clear trends have not yet emerged.
A second possibility is that research has not yet adequately
considered the complexities of friends’ influence. The present
study, like most prior research, has adopted the theoretical frame-
work that influence should cause children to become more like
their friends over time (see Berndt et al., 1999). Alternative pos-
sibilities exist, however. Friends may, for example, play a role in
encouraging and supporting characteristics that make each mem-
ber of the dyad unique (e.g., “You’re really good at math, but I’m
a better basketball player”; see Tesser, Campbell, & Smith, 1984).
In such cases, friends’ beliefs may actually diverge rather than
converge. Friends may also play a role in influencing children not
to change (see Urberg, 1999). This possibility seems particularly
likely among friendship dyads where initial similarity is very high
and where achievement norms are clear (e.g., among highly tal-
ented or seriously challenged students).
One additional focus of future research will be to examine the
mechanisms of influence more directly. Evidence to date is lim-
ited, but experimental work suggests that children and their friends
may come to hold similar achievement beliefs through the pro-
cesses of modeling (e.g., Hall & Cairns, 1984; Sagotsky & Lepper,
1982; Schunk & Hanson, 1985; Schunk, Hanson, & Cox, 1987) or
through the adoption of norms that are discussed, promoted, and
reinforced by friends both within and outside of the classroom
setting (e.g., Berndt, Laychak, & Park, 1990; see also Kurdek,
Fine, & Sinclair, 1995; Steinberg, Dornbusch, & Brown, 1992;
Wentzel & Caldwell, 1997). Two key findings from the present
study suggest that communication about academic norms and
discussion of achievement beliefs among friends may play a key
role. First, a number of the beliefs examined (particularly chil-
dren’s competence-related beliefs) do not have clear behavioral
representations. In consequence, it is unlikely that children pick up
on these beliefs simply by watching and imitating the actions of
classmates. Instead, we expect that children pick up on these
beliefs either by observing or, more likely, by participating in
conversations in which these achievement beliefs are discussed.
Second, we provide evidence that children are more likely to share
beliefs with and, perhaps, be influenced by the beliefs of recipro-
cated rather than unilateral best friends. This finding may result, in
part, from the greater stability of reciprocated friendships. For
example, in the present study, whereas 78% of children who had a
reciprocated friend at Wave 2 maintained this friendship at
Wave 3, only 49% of children with a unilateral friend did so. At
the same time, our finding that associations are stronger among
reciprocated friends than among unilateral friends provides at least
preliminary evidence that the high levels of intimate discussion,
information exchange, and advice-giving that characterize these
relationships (see Newcomb & Bagwell, 1995; Saxon, 1996) may
be key mechanisms of influence. This hypothesis is consistent with
findings reported by Berndt et al. (1990) who demonstrated that
children are most likely to change their achievement beliefs in
discussions with friends when these friendships are characterized
by high levels of harmony and low levels of conflict.
Despite this evidence, it is important to keep in mind that other
mechanisms of influence (including many indirect mechanisms)
likely play a role. In particular, given evidence that children and
their friends may spend 25% to 40% of even their nonclass time
together (see Larson & Richards, 1991), it seems quite plausible
that shared time, shared experiences, and shared interactions with
other agents of socialization (e.g., parents, mentors, siblings) are
contributors. This last possibility (i.e., that persons and situations
outside of the friendship dyad may contribute to friends becoming
more like one another) points out the important caveat that al-
though peer influence may be indicated in our longitudinal anal-
yses, alternatives are possible. Consideration of the interactive role
that parents, teachers, and peers likely play in contributing to
changes in children’s achievement-related beliefs will be espe-
cially valuable in future work (see Eccles et al., 1998; Steinberg et
al., 1992).
In the end, the findings from the present study suggest that
children and their friends are (at least modestly) psychologically
similar in the academic domain and that, like parents and teachers,
friends play a role in influencing children’s achievement-related
beliefs. At the same time, the results of this and other recent
studies of peer influence highlight the importance of developing
and empirically validating theoretical models that address the
complexities of friends’ influence on children’s school attitudes
and outcomes. A key feature of these models will be to address not
only the ways in which children’s beliefs change over time to
match those of their friends but the ways in which children and
their friends continually modify both their beliefs and their friend-
ship choices to meet the myriad goals (e.g., school success, peer
121
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
acceptance, identity formation) that children confront in the aca-
demic setting.
References
Ames, C. (1978). Children’s achievement attributions and self-reinforce-
ment: Effects of self-concept and competitive reward structure. Journal
of Educational Psychology, 70, 345–355.
Berndt, T. J. (1979). Developmental changes in conformity to peers and
parents. Developmental Psychology, 15, 606616.
Berndt, T. J., Hawkins, J. A., & Jiao, Z. (1999). Influences of friends and
friendships on adjustment to junior high school. Merrill-Palmer Quar-
terly, 45, 13–41.
Berndt, T. J., & Keefe, K. (1995). Friends’ influence on adolescents’
adjustment to school. Child Development, 66, 1312–1329.
Berndt, T. J., Laychak, A. E., & Park, K. (1990). Friends’ influence on
adolescents’ academic achievement motivation: An experimental study.
Journal of Educational Psychology, 82, 664670.
Berndt, T. J., & Miller, K. E. (1990). Expectancies, values, and achieve-
ment in junior high school. Journal of Educational Psychology, 82,
319–326.
Byrne, D., & Nelson, D. (1965). Attraction as a linear function of propor-
tion of positive reinforcements. Journal of Personality and Social Psy-
chology, 1, 659663.
Cohen, J. (1983). Commentary: The relationship between friendship selec-
tion and peer influence. In J. L. Epstein & N. Karweit (Eds.), Friends in
school (pp. 163–174). New York: Academic Press.
Costanzo, P. R., & Shaw, M. E. (1966). Conformity as a function of age
level. Child Development, 37, 967–975.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-
determination in human behavior. New York: Plenum Press.
Diener, C. I., & Dweck, C. S. (1978). An analysis of learned helplessness:
Continuous changes in performance, strategy, and achievement cogni-
tions following failure. Journal of Personality and Social Psychol-
ogy, 36, 451–462.
Diener, C. I., & Dweck, C. S. (1980). An analysis of learned helplessness:
II. The processing of success. Journal of Personality and Social Psy-
chology, 39, 940–952.
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to
motivation and personality. Psychological Review, 95, 256–273.
Eccles, J. S., Midgley, C., & Adler, T. (1984). Grade-related changes in the
school environment: Effects on achievement motivation. In J. G. Ni-
cholls (Ed.), Advances in motivation and achievement (pp. 283–331).
Greenwood, CT: JAI.
Eccles, J. S., Wigfield, A., & Schiefele, U. (1998). Motivation to succeed.
In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child
psychology: Vol. 3. Social, emotional, and personality development (5th
ed., pp. 1017–1095). New York: Wiley.
Eccles Parsons, J., Adler, T., & Meece, J. L. (1984). Sex differences in
achievement: A test of alternate theories. Journal of Personality and
Social Psychology, 46, 2643.
Eccles Parsons, J., Kaczala, C. M., & Meece, J. L. (1982). Socialization of
achievement attitudes and beliefs: Classroom influences. Child Devel-
opment, 53, 322–339.
Eccles Parsons, J., Meece, J. L., Adler, T. F., & Kaczala, C. M. (1982). Sex
differences in attributions and learned helplessness. Sex Roles, 8, 421–
432.
Ellis, S., Rogoff, B., & Cromer, C. C. (1981). Age segregation in children’s
social interactions. Developmental Psychology, 26, 399407.
Ennett, S. T., & Bauman, K. E. (1994). The contribution of influence and
selection to adolescent peer group homogeneity: The case of adolescent
cigarette smoking. Journal of Personality and Social Psychology, 67,
653–663.
Epstein, J. L. (1983). The influence of friends on achievement and affective
outcomes. In J. L. Epstein & N. L. Karweit (Eds.), Friends in school (pp.
177–200). New York: Academic Press.
Fisher, L. A., & Bauman, K. E. (1988). Influence and selection in the
friend–adolescent relationship: Findings from studies of adolescent
smoking and drinking. Journal of Applied Social Psychology, 18, 289
314.
Frome, P. M., & Eccles, J. S. (1998). Parents’ influence on children’s
achievement-related perceptions. Journal of Personality and Social Psy-
chology, 74, 435–452.
Grolnick, W. S., & Ryan, R. M. (1989). Parent styles associated with
children’s self-regulation and competence in school. Journal of Educa-
tional Psychology, 81, 143–154.
Hall, W. M., & Cairns, R. B. (1984). Aggressive behavior in children: An
outcome of modeling or social reciprocity? Developmental Psychol-
ogy, 20, 739–745.
Haller, A. O., & Butterworth, C. E. (1960). Peer influence on levels of
occupational and educational aspiration. Social Forces, 38, 289–295.
Hallinan, M. T. (1983). Commentary: New Direction for research on peer
influence. In J. L. Epstein & N. L. Karweit (Eds.), Friends in school (pp.
219–231). New York: Academic Press.
Hallinan, M. T., & Williams, R. A. (1990). Students’ characteristics and
the peer-influence process. Sociology of Education, 63, 122–132.
Hamm, J. V. (2000). Do birds of a feather flock together? The variable
bases for African American, Asian American, and European American
adolescents’ selection of similar friends. Developmental Psychology, 36,
209–219.
Hanks, M. P., & Eckland, B. K. (1976). Athletics and social participation
in the educational attainment process. Sociology of Education, 49, 271–
294.
Harter, S. (1981). A new self-report scale of intrinsic versus extrinsic
motivation: Motivational and informational components. Developmental
Psychology, 17, 300–312.
Harter, S. (1982). The Perceived Competence Scale for Children. Child
Development, 53, 87–97.
Hartup, W. W., & Stevens, N. (1997). Friendships and adaptation in the life
course. Psychological Bulletin, 121, 355–370.
Haselager, G. J. T., Hartup, W. W., van Lieshout, C. F., & Riksen-
Walraven, J. M. A. (1998). Similarities between friends and nonfriends
in middle childhood. Child Development, 69, 1198–1208.
Higgins, E. T., & Parsons, J. E. (1983). Social cognition and the social life
of the child: Stages as subcultures. In E. T. Higgins, D. N. Ruble, &
W. W. Hartup (Eds.), Social cognition and social development (pp.
15–62). New York: Cambridge University Press.
Ide, J. K., Parkerson, J., Haertel, G. D., & Walberg, H. J. (1981). Peer
group influence on educational outcomes: A quantitative synthesis.
Journal of Educational Psychology, 73, 472–484.
Kandel, D. (1973). Adolescent marijuana use: Role of parents and peers.
Science, 181, 1067–1070.
Kandel, D. B. (1978a). Homophily, selection, and socialization in adoles-
cent friendships. American Journal of Sociology, 84, 427–436.
Kandel, D. B. (1978b). Similarity in real-life adolescent friendship pairs.
Journal of Personality and Social Psychology, 36, 306–312.
Kenny, D. A. (1979). Correlation and causality. New York: Wiley.
Kindermann, T. A. (1993). Natural peer groups as contexts for individual
development: The case of children’s motivation in school. Developmen-
tal Psychology, 29, 970–977.
Kindermann, T. A., McCollam, T. L., & Gibson, E. (1996). Peer networks
and students’ classroom engagement during childhood and adolescence.
In J. Juvonen & K. R. Wentzel (Eds.), Social motivation: Understanding
children’s school adjustment (pp. 279–312). New York: Cambridge
University Press.
Kupersmidt, J. B., DeRosier, M. E., & Patterson, C. P. (1995). Similarity
as the basis for children’s friendships: The roles of sociometric status,
aggressive and withdrawn behavior, academic achievement and demo-
graphic characteristics. Journal of Social and Personal Relation-
ships, 12, 439452.
Kurdek, L. A., Fine, M. A., & Sinclair, R. J. (1995). School adjustment in
122
ALTERMATT AND POMERANTZ
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
sixth graders: Parenting transitions, family climate, and peer norm
effects. Child Development, 66, 430445.
Kurdek, L. A., & Krile, D. (1982). A developmental analysis of the relation
between peer acceptance and both intrapersonal understanding and per-
ceived social self-competence. Child Development, 53, 1485–1491.
Larson, R., & Richards, M. H. (1991). Daily companionship in late
childhood and early adolescence: Changing developmental contexts.
Child Development, 62, 284–300.
Medrich, E. A., Rosen, J., Rubin, V., & Buckley, S. (1982). The serious
business of growing up. Berkeley: University of California Press.
Meece, J. L., Wigfield, A., & Eccles, J. S. (1990). Predictors of math
anxiety and its influence on young adolescents’ course enrollment in-
tentions and performance in mathematics. Journal of Educational Psy-
chology, 82, 60–70.
Neckerman, H. J. (1996). The stability of social groups in childhood and
adolescence: The role of the classroom social environment. Social De-
velopment, 5, 131–145.
Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., et al.
(1996). Intelligence: Knowns and unknowns. American Psycholo-
gist, 51, 77–101.
Newcomb, A. F., & Bagwell, C. L. (1995). Children’s friendship relations:
A meta-analytic review. Psychological Bulletin, 117, 306–347.
Nicholls, J. G. (1975). Causal attributions and other achievement-related
cognitions: Effects of task outcome, attainment value, and sex. Journal
of Personality and Social Psychology, 31, 379–389.
Parker, J. G., & Asher, S. R. (1993). Friendship and friendship quality in
middle childhood: Links with peer group acceptance and feelings of
loneliness and social dissatisfaction. Developmental Psychology, 29,
611–621.
Pedhazur, E. J. (1982). Multiple regression in behavioral research: Expla-
nation and prediction (2nd ed.). New York: Holt, Rinehart, & Winston.
Phillips, D. (1984). The illusion of incompetence among academically
competent children. Child Development, 55, 2000–2016.
Pintrich, P. R., & de Groot, E. V. (1990). Motivational and self-regulated
learning components of classroom academic performance. Journal of
Educational Psychology, 82, 33–40.
Pokay, P., & Blumenfeld, P. C. (1990). Predicting achievement early and
late in the semester: The role of motivation and use of learning strate-
gies. Journal of Educational Psychology, 82, 41–50.
Pomerantz, E. M., & Saxon, J. L. (2001). Children’s conceptions of ability
as stable and self-evaluative processes: A longitudinal examination.
Child Development, 72, 152–173.
Pomerantz, E. M., Saxon, J. L., & Oishi, S. (2000). The psychological
tradeoffs of making it personally important: Implication for anxiety and
depressive symptoms. Journal of Personality and Social Psychology, 79,
617–630.
Rholes, W. S., Newman, L. S., & Ruble, D. N. (1990). Understanding self
and other: Developmental and motivational aspects of perceiving per-
sons in terms of invariant dispositions. In E. T. Higgins & R. M.
Sorrentino (Eds.), Handbook of motivation and cognition: Foundations
of social behavior (Vol. 2, pp. 369407). New York: Guilford Press.
Rubin, K. H., Bukowski, W., & Parker, J. G. (1998). Peer interactions,
relationships and groups. In N. Eisenberg (Ed.), Handbook of child
psychology: Vol. 3. Social, emotional, and personality development (5th
ed., pp. 619–700). New York: Wiley.
Ruble, D. N., & Seidman, E. (1996). Social transitions: Windows into
social psychological processes. In E. T. Higgins & A. W. Kruglanski
(Eds.), Social psychology: Handbook of basic principles (pp. 830856).
New York: Guilford.
Ryan, A. M. (2001). The peer group as a context for the development of
young adolescent motivation and achievement. Child Development, 72,
1135–1150.
Sage, N. A., & Kindermann, T. A. (1999). Peer networks, behavior
contingencies, and children’s engagement in the classroom. Merrill-
Palmer Quarterly, 45, 143–171.
Sagotsky, G., & Lepper, M. R. (1982). Generalization of changes in
children’s preferences for easy or difficult goals induced through peer
modeling. Child Development, 53, 372–375.
Saxon, J. L. (1996). Distinguishing between the behavioral features and
the affective features of children’s friendships. Unpublished master’s
thesis, University of Illinois at Urbana–Champaign.
Schunk, D. H., & Hanson, A. R. (1985). Peer models: Influence on
children’s self-efficacy and achievement. Journal of Educational Psy-
chology, 77, 313–322.
Schunk, D. H., Hanson, A. R., & Cox, P. D. (1987). Peer-model attributes
and children’s achievement behaviors. Journal of Educational Psychol-
ogy, 79, 5461.
Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom:
Reciprocal effects of teacher behavior and student engagement across
the school year. Journal of Educational Psychology, 85, 571–581.
Steinberg, L., Dornbusch, S. M., & Brown, B. B. (1992). Ethnic differences
in adolescent achievement: An ecological perspective. American Psy-
chologist, 47, 723–729.
Stipek, D. J., & Gralinski, J. H. (1991). Gender differences in children’s
achievement-related beliefs and emotional responses to success and
failure in mathematics. Journal of Educational Psychology, 83, 361–
371.
Tesser, A., Campbell, J., & Smith, M. (1984). Friendship choice and
performance: Self-evaluation maintenance in children. Journal of Per-
sonality and Social Psychology, 46, 561–574.
Tuma, N. B., & Hallinan, M. T. (1979). The effects of sex, race, and
achievement on school children’s friendships. Social Forces, 57, 1265–
1285.
Urberg, K. A. (1999). Some thoughts on the investigation of peer influence.
Merrill-Palmer Quarterly, 45, 1–12.
Urberg, K. A., Cheng, C. H., & Shyu, S. J. (1991). Grade changes in peer
influence on adolescent cigarette smoking: A comparison of two mea-
sures. Addictive Behavior, 16, 21–28.
Urberg, K. A., Degirmencioglu, S. M., & Pilgrim, C. (1997). Close friend
and group influence on adolescent cigarette smoking and alcohol use.
Developmental Psychology, 33, 834844.
Vitaro, F., Tremblay, R. E., Kerr, M., Pagani, L., & Bukowski, W. M.
(1997). Disruptiveness, friends’ characteristics, and delinquency in early
adolescence: A test of two competing models of development. Child
Development, 68, 676689.
Weiner, B. (1985). An attributional theory of achievement motivation and
emotion. Psychological Review, 92, 548–573.
Wentzel, K. R., & Caldwell, K. (1997). Friendships, peer acceptance, and
group membership: Relations to academic achievement in middle
school. Child Development, 68, 1198–1209.
Wigfield, A., Eccles, J. S., Mac Iver, D., Reuman, D. A., & Midgley, C.
(1991). Transitions during early adolescence: Changes in children’s
domain-specific self-perceptions and general self-esteem across the tran-
sition to junior high school. Developmental Psychology, 27, 552–565.
Zimmerman, B. J. (1989). A social cognitive view of self-regulated learn-
ing. Journal of Educational Psychology, 81, 329–339.
Received February 28, 2001
Revision received December 14, 2001
Accepted December 17, 2001
123
SIMILARITY AND INFLUENCE
This
document
is
copyrighted
by
the
American
Psychological
Association
or
one
of
its
allied
publishers.
This
article
is
intended
solely
for
the
personal
use
of
the
individual
user
and
is
not
to
be
disseminated
broadly.
... Hardy et al. (2002) (Bukowski et al., 2009;Hartup, 1993;Wagner & Alisch, 2006;Youniss & Smollar, 1985). Freundschaften bilden sich typischerweise zwischen Schüler*innen, die einander im Hinblick auf Geschlecht, ethnische Herkunft oder Alter (Aboud et al., 2003;Graham et al., 1998;Maccoby, 1988;Titzmann et al., 2007), jedoch auch in Bezug auf schulbezogene Einstellungen, Verhaltensweisen und Leistungen, ähnlich sind (z.B. Altermatt & Pomerantz, 2003). Im Gegensatz dazu kann und sollte sich insbesondere aufwärts gerichtete Hilfe stärker am Kriterium der Kompetenz, weniger am Kriterium der Ähnlichkeit bezüglich sozialer Merkmale und insbesondere des Geschlechts orientieren (Altermatt & Pomerantz, 2003). ...
... Altermatt & Pomerantz, 2003). Im Gegensatz dazu kann und sollte sich insbesondere aufwärts gerichtete Hilfe stärker am Kriterium der Kompetenz, weniger am Kriterium der Ähnlichkeit bezüglich sozialer Merkmale und insbesondere des Geschlechts orientieren (Altermatt & Pomerantz, 2003). Dennoch zeigte eine aktuelle eigene Arbeit, dass Schüler*innen fachliche Hilfe in Mathematik bevorzugt innerhalb ihrer geschlechts-und leistungshomogenen Freundschaftscliquen suchten (Zander et al., 2019). ...
Article
Full-text available
Hilfe und Unterstützung von Mitschüler*innen kann bei der Überwindung von fachlichen Schwierigkeiten und damit für die Aufrecht-erhaltung von Motivation sowie für den Lernerfolg eine bedeutsame Rolle zukommen. Der vorliegende Beitrag untersucht anhand der Daten einer Mikro-Längsschnitterhebung zweier Klassen einer Regelschule, wie sich fachliche Hilfenetzwerke in Mathematik und Deutsch unter Peers von der 5. bis zur 10. Jahrgangsstufe entwickeln. Dabei analysieren wir unter anderem, wie die Einbindung einzelner Schüler*innen in diese Netzwerke mit deren Noten und fachlichen Selbstkonzepten zusammenhängt. Da es sich bei Mathematik und Deutsch um geschlechtskonnotierte Fachdomänen handelt, in denen die Einschätzungen eigener Kompetenzen von Mädchen und Jungen von Geschlechterstereotypen und -rollen geprägt sind, beleuchten wir genauer, welche Bedeutung das Geschlecht für die Entstehung von und Einbindung in diese(n) Hilfenetzwerke(n) besitzt. In Übereinstimmung mit existierender Forschung zeigten unsere Ergebnisse zunächst stereotyp-konforme Ausprägungen der fachlichen Selbstkonzepte. Mädchen waren zudem in beiden Fächern besser in Hilfenetzwerke eingebunden, was, insbesondere in Mathematik, nicht vollständig durch ihre Kompetenzen erklärbar war. Insgesamt nahm Homophilie, also die individuelle Bevorzugung gleichgeschlechtlicher Helfer*innen, bei Mädchen und Jungen über die Zeit ab. Jungen wählten bereits mit Beginn der mittleren Adoleszenz im Fach Deutsch Mädchen ebenso häufig als Helferinnen wie Jungen als Helfer. Mädchen wählten hingegen erst im letzten Jahr vor Schulabschluss Jungen als Helfer in Mathematik. Individuelle Freundschaftswahlen waren noch stärker als Hilfewahlen von Homophilie geprägt. Implikationen für Lehrpersonen, vor allem mit Bezug zu ihrer Rolle in der Gestaltung von Hilfebeziehungen unter Schüler*innen, werden diskutiert.
... Because of these friendship characteristics, much attention has been directed towards friendship socialization. Initial research in this area considered the longitudinal effect of friend group descriptive norms on changes in students' individual motivation, engagement, and achievement in school (Altermatt & Pomerantz, 2003;Berndt & Keefe, 1995;Ryan, 2001). Friend influence has also been affirmed using different strategies to identify youth's friend groups. ...
... Friend influence has also been affirmed using different strategies to identify youth's friend groups. Friendships have been found to influence academic performance and ability attributions for success and failure (Altermatt & Pomerantz, 2003). Friend group norms also predict changes in students' intrinsic value for school (Ryan, 2001). ...
Article
Full-text available
The goal of our article is to consider the intersection of the peer ecology and teacher practices for students’ academic motivation. We begin by reviewing two perspectives that explain why and how peers matter for students’ motivation. First, the quality of peer relationships and interactions provide affordances for social support. Second, peers are socializing agents, so the content of peer interactions matters for the development of students’ achievement beliefs, values, and goals. Within each of these theoretical frameworks, we discuss three kinds of peer relationships: friendship, social status, as well as the culture of support and norms that characterize the classroom peer group. Throughout, we consider classroom contextual factors that explain why peer relationships matter for students’ motivation and school adjustment. This sets the stage for the key goal of our article, which is to review evidence from the last ten years linking teacher practices to aspects of the classroom peer ecology that are important for students’ motivation in school. We conclude with a discussion of implications for educators and important directions for future research.
... Prior research has indicated that peer relationships are crucial influences in the lives of youths and can have a significant impact on their academic performance and social-emotional wellbeing (Brown & Larson, 2009;Uslu & Gizir, 2017;Yu et al., 2018). In seeking to understand how positive peer relationships are nurtured in schools, researchers have found that a myriad of factors at both the classroom level as well as in schoolwide characteristics and programs may impact the quality of peer interactions (Altermatt & Pomerantz, 2003;Varga & Zaff, 2018;Wentzel et al., 2009). This scholarship, while valuable, has generally treated these systems (classrooms, schools, communities) as separate spheres rather than dynamically interacting to impact the ways in which youth relate to one another in school settings. ...
... Researchers have found correlations between positive peer relationships and numerous academic and social-emotional outcomes (Stotsky & Bowker, 2018). Supportive relationships among peers are positively associated with academic achievement, academic motivation, prosocial behavior, and school engagement (Altermatt & Pomerantz, 2003;Li et al., 2011). Peers can thus provide one another with a source of emotional support and help youth manage adversities (Moses & Villodas, 2017;Sanders et al., 2017). ...
... This similarity among individuals is known as homophily and is seen in a variety of characters, especially academic characteristics. (Altermatt and Pomerantz, 2003;Kindermann, 2007;Prinstein and Dodge, 2008;Ryan, 2000). Additionally, studies have reported that if you have a good friend or companion who appreciates your academic achievements, serves as a positive factor to achieve motivation. ...
Research
Full-text available
This paper takes into consideration the Covid-19 pandemic situation and provides a brief outlook on the impact on high school students. It mainly investigates the association of peer connection with students' attitude towards school as well as their academic motivation. The study was conducted through quantitative method and a descriptive research design was implemented. Three following scales were used to obtain the data: Flourishing Families Survey of Family Life, Peer Connection Subscale, Rao's School Attitude Inventory and The Academic Motivation Scale. The results highlighted an overall positive and a significant relationship between the variables. The conclusion has strongly emphasized the importance of peer connection.
... Identifying as "a good student" and perceiving someone else the same way might be a more important dimension along which to select similar friends than objective achievement. This is supported by empirical work demonstrating similarity among friends regarding self-perceptions and motivational beliefs (Altermatt & Pomerantz, 2003). Another study based on longitudinal social network modeling found that similarity among students regarding academic self-efficacy and achievement can be attributed to selection processes along with self-efficacy and achievement (Shin & Ryan, 2014). ...
Article
Social comparisons with peers are important sources of self-development during adolescence. Many previous studies showed that students' academic self-concepts (ASC) form by contrasting one's own achievement with the average of one's class or school (the Big-Fish-Little-Pond Effect [BFLPE]). Based on social comparison theory, however, we would expect some peers to be more likely social comparison targets than other peers, for example, because they are more visible or students perceive them as similar to themselves. In this study, we used sociometric data to analyze which peers play the most important role for social comparison effects on ASC. We examined how the average achievement of friends, study partners, peers perceived as popular by the student, as well as same-gender and same-ethnic peers affect the general ASC and how these effects compare to the effect of the classroom's average achievement. The study was based on a German longitudinal sample of 2,438 students (44% no recent immigrant background, 19% Turkish immigrant background, 10% Eastern European immigrant background, 27% other immigrant background) from 117 school classes that were followed from grade 9 to 10. Results from longitudinal social network analysis do not confirm substantial incremental effects of specific types of peers, while class average achievement showed a stable negative effect (confirming the BFLPE). In addition, we could provide evidence for social selection effects based on ASC. We conclude that classrooms provide a specific setting that imposes social comparisons with the "generalized peer" rather than with specific subgroups of peers. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
Article
Conduct problems are increasingly prevalent in girls and they uniquely predict negative outcomes. Yet, few reliable risk factors for aggression and violence in girls and women have been identified. Although preliminary evidence suggests peer relationships may be central to the development of youth conduct problems, especially in girls, rigorous interactive models of peer risk and protective factors for conduct problems are lacking. Based on 3104 10−13‐year‐old girls in the Adolescent Brain and Cognitive Development study, we tested the independent associations of separate peer risk factors (i.e., relational aggression victimization, physical aggression victimization, and deviant peer affiliation) with multidimensional conduct problems, including their moderation by peer support. Being the victim of relational aggression, being the victim of physical aggression, and deviant peer affiliation were each positively associated with conduct problems and perpetration of aggression whereas peer support was negatively associated with youth report conduct problems and perpetration of physical aggression. Further, elevated peer support significantly attenuated the association of being the victim of relational aggression with teacher‐rated conduct problems. These results highlight the sensitivity of conduct problems to peer risk factors and suggest that peer support designates important configurations of risk that differentially relate to conduct problems in girls.
Article
Cultivating critical thinking skills in referencing a plethora of information epitomizes the learning objectives of technology in education. However, different social influences in a classroom, such as academic prestige, friendship, and gender, may impact students’ referencing behavior. Specifically, we investigated how various factors affect pupils’ referencing of peer-generated questions during an online test-construction learning activity and perceptions. Three fifth-grade classes ( n = 75, M age = 11.08, SD age = 0.28) participated in a two-stage mixed methods research for 12 weeks. We found significant effects of friendship and inferred task-specific ability on online referencing behavior based on the multiple regression quadratic assignment procedures, simple regressions, and two-step hierarchical linear regressions. The survey data further revealed that students considered the content of their peers' questions most frequently among a mixture of other factors influencing online referencing decisions. The implications and suggestions for instruction, future studies, and system designs involving behavior connoting intent to endorse are provided to help create meaningful, reflective, and impartial online learning spaces.
Article
Using a transactional framework, this study explored social relationships in the classroom as mediators of the association between ethnic-racial identity and academic-related outcomes. Participants were 101 fifth graders of diverse backgrounds who completed computer-based questionnaires about their friendships, ethnic-racial identity, and academic engagement. Teachers reported on closeness in their student-teacher relationships. Relationships in the expected direction were evident; positive associations were observed among public regard dimensions of ethnic-racial identity and cognitive engagement in the classroom. Correlational analyses demonstrated higher friendship quality was associated with cognitive engagement, indicating more self-regulated and strategic approaches to learning for both boys and girls. Further, path analyses revealed that the relationship between public regard and cognitive engagement was mediated by student-teacher closeness for the whole sample. Gender differences were evident; for boys, public regard was related indirectly to language arts and math grades through cognitive engagement whereas for girls this indirect effect was not present. Findings highlight the varied contribution of ethnic-racial identity and classroom relationships on achievement-related outcomes, particularly for boys.
Article
Bu araştırmanın amacı, okul öncesi dönem çocuklarının bağımsız öğrenme davranışlarının sosyal konum değişkenleri üzerindeki yordayıcı etkilerinin incelenmesidir Araştırma, ilişkisel tarama yönteminde hazırlanmıştır. Araştırmanın örneklemini, 2021-2022 eğitim öğretim yılında okul öncesi eğitime devam eden 5-6 yaş grubundan 144 çocuk oluşturmuştur. Veri toplama aracı olarak Kişisel Bilgi Formu, Sosyometri Tekniği ve Bağımsız Öğrenme Davranışları Ölçeği (Saraç, Karakelle, Whitebread, 2019) kullanılmıştır. Sosyometri, araştırmacılar tarafından her çocuğa sınıf dışındaki özel bir mekânda ayrı ayrı uygulanmıştır. Kişisel Bilgi Formu ve Bağımsız Öğrenme Davranışları Ölçeği (Saraç, Karakelle, Whitebread, 2019) öğretmenler tarafından her bir çocuk için ayrı ayrı doldurulmuştur. Araştırmanın verileri, SPSS 20.0 istatistik paket programında, Pearson Momentler Çarpımı Korelasyon Katsayısı ve Basit Doğrusal Regresyon Analiz teknikleri ile analiz edilmiştir. Bulgulara göre bağımsız öğrenme davranış düzeyleri ile akranları tarafından sevilme düzeyleri arasında olumlu yönde akranları tarafından sevilmeme arasında olumsuz yönde ilişkinin olduğu belirlenmiştir. Bağımsız öğrenme davranışları, akranları tarafından sevilme ve sevilmeme düzeylerini anlamlı biçimde yordamaktadır. Bağımsız öğrenme davranış düzeyleri ile sosyal etki arasında anlamlı bir ilişki bulunmaz iken sosyal tercih düzeyleri arasında olumlu yönde, orta düzeyde anlamlı bir ilişki olduğu belirlenmiştir. Bağımsız öğrenme davranışları, sosyal etkiyi anlamlı biçimde yordamamakta, sosyal tercihi anlamlı biçimde yordamaktadır.
Article
Full-text available
In this chapter a theory of motivation and emotion developed from an attributional perspective is presented. Before undertaking this central task, it might be beneficial to review the progression of the book. In Chapter 1 it was suggested that causal attributions have been prevalent throughout history and in disparate cultures. Studies reviewed in Chapter 2 revealed a large number of causal ascriptions within motivational domains, and different ascriptions in disparate domains. Yet some attributions, particularly ability and effort in the achievement area, dominate causal thinking. To compare and contrast causes such as ability and effort, their common denominators or shared properties were identified. Three causal dimensions, examined in Chapter 3, are locus, stability, and controllability, with intentionality and globality as other possible causal properties. As documented in Chapter 4, the perceived stability of a cause influences the subjective probability of success following a previous success or failure; causes perceived as enduring increase the certainty that the prior outcome will be repeated in the future. And all the causal dimensions, as well as the outcome of an activity and specific causes, influence the emotions experienced after attainment or nonattainment of a goal. The affects linked to causal dimensions include pride (with locus), hopelessness and resignation (with stability), and anger, gratitude, guilt, pity, and shame (with controllability).
Article
Full-text available
On the basis of a new model of motivation, we examined the effects of 3 dimensions of teacher (n = 14) behavior (involvement, structure, and autonomy support) on 144 children's (Grades 3-5) behavioral and emotional engagement across a school year. Correlational and path analyses revealed that teacher involvement was central to children's experiences in the classroom and that teacher provision of both autonomy support and optimal structure predicted children's motivation across the school year. Reciprocal effects of student motivation on teacher behavior were also found. Students who showed higher initial behavioral engagement received subsequently more of all 3 teacher behaviors. These findings suggest that students who are behaviorally disengaged receive teacher responses that should further undermine their motivation. The importance of the student-teacher relationship, especially interpersonal involvement, in optimizing student motivation is highlighted.
Book
I: Background.- 1. An Introduction.- 2. Conceptualizations of Intrinsic Motivation and Self-Determination.- II: Self-Determination Theory.- 3. Cognitive Evaluation Theory: Perceived Causality and Perceived Competence.- 4. Cognitive Evaluation Theory: Interpersonal Communication and Intrapersonal Regulation.- 5. Toward an Organismic Integration Theory: Motivation and Development.- 6. Causality Orientations Theory: Personality Influences on Motivation.- III: Alternative Approaches.- 7. Operant and Attributional Theories.- 8. Information-Processing Theories.- IV: Applications and Implications.- 9. Education.- 10. Psychotherapy.- 11. Work.- 12. Sports.- References.- Author Index.
Chapter
This chapter presents new directions for research on peer influence. Questions concerning the influence on students of friends and peers have long been of interest to sociologists and educators. An even larger body of research compares attitudes and behaviors of students to those of their peers and infers peer influences from observed similarities. Influence can be defined as any factor that affects the formation of a person's attitudes and opinions by acting directly on his or her beliefs. As such, influence is only one of many ways by which interactions produce results. Other ways include inducement, in which rewards are promised for compliance, and deterrence, in which sanctions are threatened for noncompliance. Both of these kinds of interactions are based on social power. The extent of a person's willingness to accept information from another is determined by the extent to which individuals believe they will not be deceived when they accept the information. When an individual and the person from whom information is received are not in a relationship of solidarity or are not pursuing the same ends, the individual will be open to influence only if the normative system that governs the actions of the person transmitting the information can be trusted.
Chapter
This chapter discusses the relationship between friendship selection and peer influence. The relatively impersonal influence of nonselected others has been conceptualized in a number of ways: (1) an interpersonal environment, (2) an elite, (3) a school or peer context, and (4) a value climate. A student's interpersonal environment consists not of selected friends but of everyone registered at the college whose name the student recognized. The second type of nonselected peers who are believed by some to have influence are school elites. A third type of influence that short-circuits the friendship selection process is the influence of peer contexts. Contextual effects, like interpersonal environments, have been said to affect students by establishing a certain ambience. This commentary assembles empirical evidence that confirms the link between selection and influence and provides a framework for codifying the research findings in future research.
Article
Helpless children show marked performance decrements under failure, whereas mastery-oriented children often show enhanced performance. Current theories emphasize differences in the nature of the attributions following failure as determinants of response to failure. Two studies with 130 5th-grade children explored helpless vs mastery-oriented differences in the nature, timing, and relative frequency of a variety of achievement-related cognitions by continuously monitoring verbalizations following failure. Results reveal that helpless children made the expected attributions for failure to lack of ability; mastery-oriented children made surprisingly few attributions but instead engaged in self-monitoring and self-instructions. That is, helpless children focused on the cause of failure, whereas the mastery-oriented children focused on remedies for failure. These differences were accompanied by striking differences in strategy change under failure. The results suggest that in addition to the nature of the attribution one makes, the timing or even occurrence of attributions may be a critical individual difference. (18 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).
Article
A correlational study examined relationships between motivational orientation, self-regulated learning, and classroom academic performance for 173 seventh graders from eight science and seven English classes. A self-report measure of student self-efficacy, intrinsic value, test anxiety, self-regulation, and use of learning strategies was administered, and performance data were obtained from work on classroom assignments. Self-efficacy and intrinsic value were positively related to cognitive engagement and performance. Regression analyses revealed that, depending on the outcome measure, self-regulation, self-efficacy, and test anxiety emerged as the best predictors of performance. Intrinsic value did not have a direct influence on performance but was strongly related to self-regulation and cognitive strategy use, regardless of prior achievement level. The implications of individual differences in motivational orientation for cognitive engagement and self-regulation in the classroom are discussed.
Article
Many theorists have suggested that students' motivation to achieve in school depends on their expectancies for success and the value they attach to success. There are few data, however, on the relation between expectancies and values or their relative contribution to achievement. To examine these issues, we asked 153 seventh graders to complete multiple measures of academic expectancies and values. We used students' report card grades and academic track placements in English and math as indicators of their achievement. Covariance structure analyses showed that students' expectancies were more strongly related to their achievement than were their values. Nevertheless, both expectancies and values made significant, independent contributions to achievement. In addition, the constructs for expectancies and values were positively correlated. Boys and girls had similar expectancies, but boys appeared to value academic success less than did girls.