The role of peer groups in adolescent students´
academic self-concept and math performance
Korhonen Johan (Åbo Akademi University, Finland); Karin Linnanmäki (Åbo
Akademi University, Finland) and Pirjo Aunio (University of Helsinki, Finland)
Abstract
This paper reports a longitudinal study on math performance and academic selfconcept on student and peer group level. Adolescent students (N=1,152) were
followed up from the end of compulsory education to secondary education.
Preliminary results indicate that a reciprocal relationship exists between academic
self-concept and math performance on student level. Multilevel structural equation
modelling will be used to examine if peer group achievement has a negative effect on
student-level academic self-concept, as is the case with school level achievement and
student-level academic self-concept (the-big-fish-little-pond-effect). Furthermore,
prior research has found that academic self-concept mediate the relationship between
perceptions of friends’ academic behaviors and academic performance. This study
aims to advance the current knowledge base by examining the mediating role of
academic self-concept between friend´s actual academic behavior and student
academic performance. This will expand our understanding on how peer groups affect
adolescent students math performance, and if this relationship is mediated by
academic self-concept.
Extended summary
Introduction
Mathematics is an important skill in everyday life. Students´ mathematics
achievement and positive self-concept are quality indicators of work-force preparation
in the global marketplace (Evans, 2005). Students´ self-perceptions of their
competence or ability are at the core of several psychological theories aimed at
explain learning gains and achievement-related choices. Academic self-concept is
defined as mental representations of one’s abilities in academic domains (Marsh &
Craven, 1997). Academic self-concept has been found to be highly related to
achievement, even after controlling for prior achievement (Eccles, Wigfield, Harold,
& Blumfeld, 1993; Trautwein, Ludtke, Köller, & Baumert, 2006). There is also
convincing evidence that student-level achievement has a positive effect, and schoolaverage achievement has a negative effect (the-big-fish-little-pond-effect, BFLPE) on
student-level academic self-concept (Marsh, 1987; Nagengast & Marsh, 2011). The
adolescent peer environment is a multilevel, multifaceted environment that includes
three main types of peer interactions: dyads (individual relationships), cliques (small
groups of peers who regularly interact with each other), and crowds (adolescents who
share the same reputation, but who may not necessarily interact with each other
(Brown, 2004; Rubin et al., 2006). As noted earlier, extensive research has been done
on how school-average achievement (crowds) affects student-level academic selfconcept but less is known if the same relationship exists between peer groups
(cliques) and academic self-concept.
Korhonen, Linnanmäki, Aunio, abstract and extended summary for NORSMA2013 Aims
The main aim of this study is to examine the role of peer groups in adolescent
students´ academic self-concept and math performance. More specifically we will
test: 1) if peer group math performance (time point 1) has a negative effect on
student-level academic self-concept (T2) (i.e. BFLPE), when controlling for studentlevel math performance 2) a mediation model were student level academic selfconcept (T1) mediates the relationship between peer group math performance (T1)
and student level math performance (T2) controlling for prior student level math
performance (T1).
Methodology
Participants. The participants in this study were 1,152 (576 girls, 576 boys) Swedishspeaking students. All students were measured at the end of ninth grade (mean
age=15.8 years, SD=4.9). The follow-up took place two years later when the students
attended either vocational (N=471) or upper secondary (N=680) education (mean
age=17.9 years, SD=4.9). Of the original sample of 1,152 students, 748 participated in
the second data collection.
Methods of data collection. Students´ Mathematical performance was assessed with
the KTLT-test (Räsänen & Leino, 2005), which is a standardized test for grades 7-9
(13-16 years). The test identifies students with difficulties in basic mathematics and
difficulties in applying these abilities in a context. A low score in the test indicate low
achievement in mathematics. The test consists of 40 tasks. The test is a paper and
pencil test and the students have 40 minutes to do the test. The reliability estimate of
the mathematical achievement test at T1 was 0.89, and 0.92 at T2. Academic selfconcept was measured with items from the Perceived Competence Scale for Children
(Harter, 1982), loading on the academic self-concept subscale (8 items). The
reliability estimate for the academic self-concept subscale at T1 was 0.84, and 0.84 at
T2. In the third data collection participants’ peer groups in grade 9 will be identified
by using a socio-metric procedure developed by Coie, Dodge, and Coppotelli (1982).
The participants will be asked to nominate up to three classmates within their school
(retrospectively) with whom they most liked to spend time in grade 9 (T1).
Procedure. The present study is part of a longitudinal study following the students
over a period of five years. The first data collection was performed by the researchers
and trained research assistants who conducted the measurements with groups of
students in their own schools during ordinary lessons. The second data collection was
performed by trained research assistants who conducted the measurements with
groups of students after school. The third data collection (January-February 2013)
will be conducted in two steps; first, participants will be contacted by phone and
asked to answer questions concerning friends and current life situation. In the second
step those who gave their e-mail in the phone interview get an electronic survey with
measures of self-concept (Harter). In this study we will use information concerning
students´ peer group from the third data collection.
Findings
The math performance and academic self-concept constructs show strong longitudinal
measurement invariance, which is a prerequisite for meaningful comparisons over
time. The preliminary structural equation modelling with student level data from the
first two measurements show that student level academic self-concept (T1) predicts
Korhonen, Linnanmäki, Aunio, abstract and extended summary for NORSMA2013 later math performance (T2) in secondary education (β = .12, p<.001) even when
controlling for math performance (T1) (β = .86, p<.001), χ2(14)=23.430, p=.0536,
CFI=.997, TLI=.994, RMSEA=.024. In NORSMA 2013 we will present results based
on longitudinal data on student and peer group level with two measurement points by
using multilevel structural equation modelling.
Theoretical and Educational Significance
In comparison to existent research literature this study examines the relationship
between peer groups (cliques) and adolescent students´ academic self-concept and
math performance. We already know that school-average performance has a negative
effect on student-level achievement, which for example speak against tracking of
students into different schools based on performance. This study will add to our
knowledge if they same effects can be found between peer groups achievement and
student-level academic self-concept. Prior research has found that academic selfconcept mediate the relationship between perceptions of friends’ academic behaviors
and academic performance (Jones, Audley-Piotrowski, & Kiefer, 2012). This study
will add to our knowledge if the same holds for friends´ actual academic behavior
using a multilevel longitudinal design. This expands our understanding of the effects
which peer groups have on adolescents´ math performance, and if this relationship is
mediated by academic self-concept.
References
Brown, B. (2004). Adolescents’ relationships with peers. In R. Lehrer & L. Steinberg
(Eds.), Handbook of adolescent psychology (2nd ed., pp. 363–394). New York,
NY: Wiley.
Coie, J. D., Dodge, K. A., Coppotelli, H. (1982). Dimensions and types of social
status: A cross-age perspective. Developmental Psychology, 18, 557–570.
Eccles, J., Wigfield, A., Harold, R., & Blumenfeld, P. (1993). Agent and gender
differences in children’s self- and task perceptions during elementary school. Child
Development, 64(3), 830-847.
Evans, R. (2005). Reframing the achievement gap. Phi Delta Kappan, 86, 582-589.
Harter, S. (1982). The perceived competence scale for children. Child Development,
53, 87-97.
Jones, M. H., Audley-Piotrowski, S. R., & Kiefer, S. M. (2012). Relationship among
adolescents´ perceptions of friends´ behaviors, academic self-concept, and math
performance. Journal of Educational Psychology, 104(1), 19-31.
Marsh, H. W., & Craven, R. G. (1997). Academic self-concept: Beyond the dustbowl.
In G. D. Phye (Ed.), Handbook of classroom assessment: Learning, achievement,
and adjustment, Educational psychology series (pp. 131-198). San Diego, CA:
Academic Press.
Korhonen, Linnanmäki, Aunio, abstract and extended summary for NORSMA2013 Marsh, H. W. (1987). The big-fish-little-pond effect on academic self-concept.
Journal of Educational Psychology, 79, 280-295.
Nagengast, B., & Marsh, H. W. (2011). The negative effect of school-average ability
on science self-concept in the UK, the UK countries and the world: The big-fishlittle-pond effect for PISA 2006. Educational Psychology, 31(5), 629-656.
Rubin, K. H., Bukowski, W., & Parker, J. G. (2006). Peer interactions, relationships,
and groups. In W. Damon & R. Lerner (Series Eds.) & N. Eisenberg (Volume Ed.),
Handbook of child psychology, 6th edition: Volume 3 Social, emotional, and
personality development (pp. 571–645). New York, NY: John Wiley & Sons.
Räsänen, P., & Leino, L. (2005). KTLT. Laskutaidon testi. Opas yksilö- tai
ryhmämuotoista arviointia varten [KTLT. Arithmetic skills test. Manual for
individual or group assessment]. Jyväskylä, Finland: NMI.
Trautwein, U., Lüdtke, O., Köller, O., & Baumert, J. (2006). Self-esteem, academic
self-concept, and achievement: How the learning environment moderates the
dynamics of self-concept. Journal of Personality and Social Psychology, 90(2),
334-349.
Korhonen, Linnanmäki, Aunio, abstract and extended summary for NORSMA2013