Children’s Peer Status and Their Adjustment in
Adolescence and Adulthood
Developmental issues in sociometric research
Peter Zettergren
Department of Psychology
Stockholm University
2007
In memory of my parents
Doctoral dissertation 2007
Department of Psychology
Stockholm University
Sweden
Abstract
The present research intended to examine the relationship between childhood stable peer
status and adjustment in midadolescence for both genders, and adjustment in early and middle
adulthood for women. One-year stably peer rejected, popular, and average boys and girls were
identified by an age 10 and age 11 sociometric classification procedure using positive
nominations and rank-ordering. These groups were examined in midadolescence. Findings
indicated that stable peer rejection in childhood was associated with negative school
adjustment and problems in peer relations in adolescence for both genders, and that many
rejected boys were school dropouts. However, rejected children did not associate with deviant
peers or show more antisocial behavior than their counterparts. For the adulthood follow-up,
cluster analysis using LICUR was applied on rank-ordering data from age 10 and age 13 in
order to identify 3-year stably rejected, popular, and average groups of girls. A
methodological comparison with a standard sociometric method showed that the applied
cluster analysis seems to be a useful additional tool in the arsenal of sociometric classification
methods. Furthermore, one seems to arrive at larger stable peer status groups with this method
than with other sociometric classification methods. The longitudinal follow-up into adulthood
showed that rejected girls were at increased risk for criminal offending and alcohol abuse in
young adulthood. Childhood aggressive behavior explained the significant association
between peer rejection and criminality. There were no significant differences between the
stable peer status clusters in the midadulthood adjustment areas of social relations, health,
occupation, and subjective well-being, except that popular girls had achieved more successful
vocational careers (which in its turn was explained by their higher academic achievement in
childhood). Constraints and implications for future research were discussed.
Keywords: sociometric peer status, classification, stability, cluster analysis, longitudinal, longterm adjustment, childhood, adolescence, adulthood
Acknowledgments
This research was made possible by access to data from the longitudinal
research program Individual Development and Adaptation – IDA – once
initiated and led by David Magnusson and since 1996 led by Lars R. Bergman.
The data collections were supported by grants to David Magnusson from the
Bank of Sweden Tercentenary Foundation and the Swedish National Board of
Education, and by grants to Lars R. Bergman from the committee of longitudinal
research at the Swedish Council for the Planning and Coordination of Research
and the Swedish Council for Social Research.
A lot of people have very generously supported me and contributed to the
present work:
First of all, I want to express my profound gratitude to Lars R. Bergman, who
has been my supervisor. He encouraged me to restart my PhD studies, and he
has with his profound scientific knowledge, valuable comments, generous
assistance, and mild, but firm, pressure guided me in the production of this
thesis.
I am, of course, very much indebted to my earlier supervisor Anders Dunér for
many fruitful discussions and comments.
I am also most grateful to David Magnusson, who with his deep wisdom,
insight, and knowledge has inspired me and numerous other students of human
development and adaptation.
Ola Andersson has been of great help and support in guiding me in the IDA data
base and I also want to thank him for his cooperation in the preparation of the
data needed for the cluster classification in this thesis. In my contacts with Viera
Gantner, and in earlier phases of the work with the thesis Luki Hagen, I have got
all the help and assistance I needed and more so. I would also like to thank Lisa
Duffy for her revision of my English.
I have been very much inspired by many research colleagues at the Department
of Psychology. Thank you Margit Wångby (who is also my co-author in one of
the Studies), Daiva Daukantaite, Cornelia Wulff, Chris Magnusson, Kari Trost,
Birgitta Rydén-Lodi, Mia Körlof, Robert Brage, Bassam El-Khouri, and Siv
Lindroth for many both joyful and scientifically rewarding discussions!
Outside of the university, I want to express my gratitude to my former superiors
at the Child Psychiatric Division of the Stockholm County (BUP), Maria
Dunerfeldt and Per-Olof Björck, for their support in letting me continue and
finish my PhD studies.
I am also most grateful to my earlier colleagues at the child psychiatric clinic in
Nynäshamn, who had to put up with my long part-time and full-time study leave
from the clinic. In later years they have been Paula Ahko, Annika Elinder, Kia
Szutkiewicz, Margareta Linell, Berit Gredell-Jansson, Ann-Marie Sellfors, Eva
Mårtensson, Jan Ljungqvist, Anita Nilsson, Ann-Marie Holm, and Inger
Schagerström. I have enjoyed working with you and learning from you!
The same goes for the members of the managerial body of the south-eastern
child psychiatric district who had to struggle with the problems brought about by
my study leave. In that context, I had the privilege to meet with and work with
many experienced and dynamic local leaders, such as Gun Svensson, Elisabeth
Henning, Elsie Arkling, Susanne Walldén, and Grels Hanzon.
Finally, thank you Mattias and Eva for always being there!
Peter Zettergren
Stockholm, February 2007
List of studies
The present thesis is based on the following studies:
Study I
Zettergren, P. (2003). School adjustment in adolescence for
previously rejected, average and popular children. British Journal of
Educational Psychology, 73, 207-221.
Reprinted with permission from the British Journal of Educational Psychology, © 2003
The British Psychological Society.
Study II
Zettergren, P. (2005). Childhood peer status as predictor of
midadolescence peer situation and social adjustment. Psychology in
the Schools, 42, 745-757.
Reprinted by permission of John Wiley & Sons, Inc. from Psychology in the Schools, ©
2005 Wiley Periodicals, Inc.
Study III Zettergren, P. (2007). Cluster analysis in sociometric research: A
pattern-oriented approach to identifying temporally stable peer status
groups of girls. Journal of Early Adolescence, 27, 90-114.
Reprinted with permission from the Journal of Early Adolescence, © 2007 Sage
Publications.
Study IV Zettergren, P., Bergman, L. R., & Wångby, M. (2006). Girls’ stable
peer status and their adulthood adjustment: A longitudinal study from
age 10 to age 43. International Journal of Behavioral Development,
30, 315-325.
Reprinted by permission of Sage Publications Ltd from the International Journal of
Behavioral Development, © 2006 International Society for the Study of Behavioral
Development.
Contents
Contextual and historical background
1
Human development and peer relations
1
The sociometric research tradition
3
Theoretical issues
5
Theoretical underpinnings pertaining to rejected peer status
5
Connection to the present research
8
Methodological issues
9
Sociometric classification methods and peer status groups
9
Introduction
9
Positive and negative nominations
10
Ethical considerations
12
Rank-orderings and peer-ratings
12
Traditional sociometric classification methods-conclusions
13
Temporal stability among peer status groups
14
New directions-cluster analysis
15
The peer context – school, age, and gender
17
Schools and school classes as peer contexts
17
The peer group delimiters of age and gender
18
Longitudinal peer status research
20
Peer status and adjustment-an introduction
20
Peer status and adjustment in adolescence
20
School adjustment
20
Peer adjustment
21
Peer status and adjustment in adulthood
25
Introductory remarks
25
Young adulthood adjustment
25
Midadulthood adjustment
26
Peer status and concurrent correlates – importance for later adjustment
Explanatory models
29
31
Aims of the thesis
34
Summary and methodological integration of the empirical studies
36
Study I
36
Study II
37
Study III
39
Study IV
41
Overview of the measures included in the studies
42
Comparison between the sociometric classifications used in the studies
43
General Discussion
46
Overview
46
Methodological issues
46
Introduction
46
The rank-ordering sociometric measure
46
The peer status classification for the adolescence follow-up (Studies I and II)
47
The peer status classification for the adulthood follow-up (Studies III and IV)
49
Stimulus aspects of sociometric criteria
52
Gender related issues
53
Peer Status and future adjustment
54
The adolescence findings
54
The adulthood findings
57
The context and scope of the reported research: Limitations and constraints
59
The diversity of peer relations research
61
Different types of peer relations
61
Different peer contexts
63
Final comments
64
References
66
Study I
Study II
Study III
Study IV
Contextual and historical background
Human development and peer relations
The theme of the present thesis concerns stable peer status in childhood – with a
special focus on peer rejection – and its developmental significance for
adjustment in adolescence and adulthood. Sociometric classification methods
are also discussed and a classification method based on cluster-analysis is
introduced. The research presented here is embedded in a wide area of peer
relations research and I will begin with a short discussion of this matter.
In different kinds of psychological developmental theories, social factors play an
important role in the development of the individual. From infancy through to
childhood, adolescence, and adulthood, the individual is engaged within a
complex web of relationships with other people. Hartup (1989), among others,
differentiated between vertical and horizontal relationships and argued that both
kinds of relationship are necessary for the child's development. First, the child
must form vertical relationships or attachments, that is, relationships with
individuals who have greater knowledge and social power than her/himself.
These persons are usually adults (first the family caregivers, primarily the
mother in most societies; then other adults, such as preschool teachers) who
provide the child with nurturance, protection, security and basic social skills.
Later, the child can form horizontal relationships with individuals who have
about the same amount of social power as he or she – usually other children
(most often of the same age and gender), with whom the child can develop
social skills, and cooperate and compete in a reciprocal way (e.g., Aboud &
Mendelson, 1996; Fabes, Martin, & Hanish, 2004; Maccoby, 2000).
The contributions and diverse functions of “horizontal” relations with peers in
the socialization process of children have received increased attention in
developmental research and a widespread consensus now exists among social
developmental researchers that children's peers provide unique and essential
contributions to social, cognitive, and emotional development (e.g., Asher &
Coie, 1990; Berndt & Ladd, 1989; Deater-Deckard, 2001; Hartup, 1983, 2005;
Parker, Rubin, Price, & DeRosier, 1995; Rubin, Bukowski, & Parker, 1998).
Damon (1983) made a distinction between the individuation and socialization
functions of development, and McLellan and Pugh (1999), in referring to
Damon’s theory, argued that peers enable the young person to see himself both
as distinct from others and as a member of groups, cultures, and, ultimately, the
greater society. Accounting for the trends of modern society, Asher (1990) assumed that peers might be even more important today than in earlier times. The
number of single-parent families has increased, and in families with two parents,
1
it is usual for both parents to work. Both these family changes have led to an
earlier entry of young children into organized peer group settings such as daycare homes, nursery schools, and day-care centers. Most children also stay in
school for more years than was previously the case, and participate in organized
peer group activities to a higher degree. Asher concluded that this means
children will spend a considerable amount of time with similar-age peers. Two
decades ago, Harkness and Super (1985) noted that “current scientific interest in
peer relations reverses a long-standing bias in western psychology that
overemphasized the role of the parents, especially the mother, as socialization
agents“ (p.219).
Considering the functions that peer relations fulfill in the life of children, not
only in western societies (for a cross-cultural review, see Krappman, 1996), it is
a sad fact that many children lack peers and friends at school and in their leisure
time. Different studies indicate that about 5-10% of the children in elementary
school are not selected as a friend by anyone in their class (Gronlund, 1959;
Hymel & Asher, 1977; Kuhlen & Lee, 1943). If the stronger criterion of
reciprocal friend selection is used, the percentage is even higher (Bukowski &
Hosa, 1989; Hartup & Stevens, 1997; Putallaz & Gottman, 1981). Asher, Parker,
and Walker (1996) estimated that at least 10% of school-age children have
serious difficulties with peer relations.
Children’s peer relations can be classified into two large sectors, namely, groupbased peer interactions and relations, and dyadic peer interactions and relations
(e.g., Gifford-Smith & Brownell, 2003). Inside these sectors, different social
phenomena have been studied. At the group level, there has been a growing
theoretical and empirical interest in children’s peer networks among
psychologists. Social network analysis has a sociological origin and a basic
premise is that children cannot be understood outside their social contexts. The
analysis aims to identify the dynamic processes and structures of children’s peer
groups (e.g., Cairns, Xie, & Leung, 1998; Farmer & Rodkin, 1996; GiffordSmith & Brownell, 2003). However, most research pertaining to the group level
has been concentrated on children’s sociometric peer status or social acceptance
by other group members (e.g., being popular or rejected). Sociometric peer
status has been defined and studied both as continuous dimensions of
popularity/likeability and rejection/dislike (e.g., Buhs & Ladd, 2001; Laird,
Pettit, Dodge, & Bates, 2005), and as categorical, one- or two-dimensional peer
status group classifications (e.g., Coie, Dodge, & Coppotelli, 1982). While peer
status is certainly a group-referent construct and processes leading to popularity
or rejection are group processes, popularity has also been seen as representing a
characteristic of the individual (e.g., Bukowski, 2001; Newcomb, Bukowski, &
Pattee, 1993; Rubin et al., 1998; Terry, 2000). Furthermore, sociometric
popularity has been contrasted to peer-perceived popularity and dominance
social structures (e.g., Parkhurst & Hopmeyer, 1998). At the dyadic level,
2
research interest has been focused on children’s friendships (e.g., Bukowski,
Newcomb, & Hartup, 1996). In later years, other, more negative dyadic
relationships have also been studied, such as mutual antipathies or enemies (e.g.,
Hartup & Abecassus, 2002; Witkow, Bellmore, Nishina, Juvonen, & Graham,
2005). The three major research domains pertaining to children’s social
networks, children’s peer status, and children’s friendships/dyadic relations have
developed in relative independence from each other, but they are theoretically,
conceptually, empirically, and methodologically interrelated (e.g., Gest,
Graham-Bermann, & Hartup, 2001; Gifford-Smith & Brownell, 2003;
Kindermann, 1998; Schneider, Atkinson, & Tardif, 2001).
In Figure 1, I have schematically summarized what has been discussed in this
section about children’s social relations, with focus on peer relations. The
research presented here has its place in the peer-status research domain (the gray
area in Figure 1). To be more specific, categorical peer status groups of rejected,
average, and popular children are studied, where the sociometric measures are
restricted to same-sex class peers.
CHILDREN’S SOCIAL RELATIONS
Vertical social relations
Horizontal social relations
with other children
Group-based peer relations
Peer networks
Dyadic peer relations
Peer status
Figure 1. Hierarchical schema of children’s social relations with focus on peer relations.
The sociometric research tradition
The term sociometry has been connoted with at least two different meanings
(Hallinan, 1981). The first and literal meaning of the term refers to a research
method for the measurement of social relations. It is a technique for gathering
and studying data about interpersonal choices and peer preferences in a group.
Sociometric research tools are often used by social scientists to examine
individual differences in popularity. A second and more general meaning refers
3
to the whole process of analyzing data from sociometric questionnaires to
drawing conclusions. Sociometry then becomes not only a data collection
technique but also a whole body of research pertaining to the study of social
relations and resulting from the analysis of data about preference choices. When
referring to the sociometric research tradition, it is often this second meaning
one has in mind.
The history of sociometry is closely interwoven with the history of peer relations
research, and Moreno (1934) stands out as the most influential researcher in the
early days of sociometry. He designed and introduced the sociometric methods
among researchers studying peer relations during the 1930s. While Moreno's
main interest was the group and its structure, other early investigators used
sociometry to assess individual differences in social acceptance and competence,
and tried to relate characteristics and behaviors of individuals to the positions
they occupied in the peer status structure of a group (e.g., Bonney, 1943;
Gronlund, 1959; Koch, 1933; Lippitt, 1941; Northway, 1946; Young & Cooper,
1944). The participants were mainly preschool children and school children.
Besides documenting the importance of peers for “normal” development, the
research of the last two or three decades has mainly studied problematic peer
relations as antecedents of maladjustment. This research has been characterized
by increased articulation and differentiation pertaining to conceptualization and
methodology, including methods for peer status classification and for prediction
(e.g., Bukowski & Adams, 2005).
The two issues of peer status classification and prospective prediction are at
focus in the present thesis. I will begin with a theoretical discussion about peer
status concentrating on peer rejection and its developmental consequences, after
which I will include a methodological part as a background to the specific
sociometric assessments of the present project. Adhering to the second meaning
of sociometry as a whole research process (Hallinan, 1981), I will then give a
research overview of peer status as a predictor or indicator of later adjustment
and development as a basis for my own longitudinal research efforts, which will
thereafter be accounted for.
4
Theoretical issues
Theoretical underpinnings pertaining to rejected peer status
Sociometric research has often been descriptive and not guided by a single
unifying theoretical perspective. However, some authors have presented
theoretical models of the possible mechanisms and processes that govern the
social interactions of children, and may underlie successful or unsuccessful peer
relationships (e.g., Crick & Dodge, 1994; Parker et al., 1995). In 1990, one of
the most productive and distinguished researchers in the sociometric research
tradition, John D. Coie, outlined a developmental theory of peer rejection that
can be of value as a background to the research presented here. The theoretical
contributions of Coie seem especially suitable for describing and explaining the
emergence of persistent peer rejection and its consequences.
Coie (1990) argues that in examining peer rejection research, it is possible to
trace the outlines of a theory, or at least a model, of the development of disorder
related to childhood peer rejection. To begin with, he places the research on peer
rejection in the larger scientific area of developmental psychopathology (e.g.,
Sroufe & Rutter, 1984), which is concerned with understanding the evolution of
patterns of adaptation and maladjustment throughout the life span, with the
etiology of conditions that are predictive of disorder, and with the development
of coping skills that buffer individuals against disorder. In his theory, Coie
distinguishes between four phases in the process of being rejected by peers:
...a precursor phase, in which the behavior patterns, attitudes, social
expectations, affective response patterns, social goal priorities, and
competences that relate to eventual peer status take shape; an emergent
status phase, in which the child's interactions with a significant peer group
result in the child being rejected by that group; a maintenance phase, during
which time rejection by the group becomes a stable and enduring reality for
some rejected children; and, finally, a consequence phase, in which other
aspects of the individual's life adjustment have deteriorated to the point of
identifiable disorders. The phases, as they are described here, represent
large time periods. Within these large time periods the developmental
processes involved can be described at a finer level of analysis. For
example, there are microprocesses reflecting recursive sequences of
behavior by the child, reactions by the group, affective and cognitive
processes, and even patterns of parental influence that determine a course
of social development that sometimes results in persistent peer rejection.
(p. 365)
5
Coie (1990) then discusses several assumptions that the model is based on. One
assumption is that a child’s social behavior, including interpretations, affective
reactions, and acquired behavioral strategies used in specific social situations, is
primarily responsible for rejection by peers. Nonbehavioral factors such as
appearance, academic achievement, and athletic ability are important, but only
when mediated by specific behavior reactions. Coie presents some research
findings that support the notion of a close relation between the child’s negative
social behavior in the peer group and social rejection (although the causal chain
remains unclear).
The second assumption in Coie’s (1990) model of peer rejection is that the
situation-specific cognitive processes, affective reactions, and behavioral
patterns emerge out of the child's socialization history. Biologically linked
attributes such as temperament, physique, and cognitive abilities also play a role,
but the character of a child's reactions to specific social circumstances is largely
shaped by the history of that child's interactions with parents, siblings, and other
social agents. In taking a developmental perspective on social rejection, this
seems to be an essential and necessary assumption.
A third assumption is that in the process of acquiring social status in the peer
group, the behavior of the child is primary and group behavior is secondary. In
contrast, a stable identity as a rejected person is maintained more by group
dynamics than by the behavior of the child.
Coie’s (1990) fourth and final assumption is that sustained peer rejection has
indirect causal effects on the long-term adjustment of the rejected child.
Although one possibility is that peer rejection may only be a marker of early
adjustment problems that simply continue into adolescence and adulthood,
another hypothesis is that peer rejection has several consequences (e.g.,
inadequate social skills and lack of social support) that contribute significantly
to the incidence of future disorder. This assumption could be said to lie behind
the present, as well as most other, longitudinal approaches in peer-rejection-risk
research (e.g., Bukowski & Adams, 2005) and it is also a guideline for different
specific hypotheses pertaining to expected disordered outcomes.
As an underlying assumption, Coie (1990) points to the fundamental shift in the
socialization influences that take place as the child grows older. Initially, parent
and family values, beliefs, and social interaction patterns are the major
influences on the social orientation of the young child (e.g., Costanzo &
Fraenkel, 1987) and a preparation for later social interactions in the larger world
of non-familial adults and children. Gradually, the peers become increasingly
more important for the child and social failures with peers are both a reflection
and a precursor of individual maladjustment. I have already discussed this
6
assumption in the background part of this thesis, referring to Hartup’s (1989)
distinction between vertical and horizontal relationships.
Finally, Coie (1990), in discussing the precursors to or origins of peer rejection,
argues that there could be two kinds of answers to the question of why children
come to behave in such ways that they get rejected by peers, involving either
proximal or distal causes. Distal causes have to do with the socialization
processes by which children acquire the behavioral, cognitive, or affective
orientations that they bring to the social situation and that are the proximal
causes of peer rejection. Distal causes, such as parental socialization influences
and differences in temperament, seem to play a role in the evolution of peer
status (see also Parker et al., 1995; Putallaz & Heflin, 1990; Rubin et al., 1998;
Rubin, LeMare, & Lollis, 1990). The parent-child relationship could be seen as
the affective prototype of all future relationships, as proposed in attachment
theory or as a training arena for social behaviors and social skills that the child
brings with him to other social relationships (learning theory). Among proximal
causes for peer rejection are the internal cognitive and affective processes of the
child that underlie maladaptive social behavior. Social-cognitive processes can
be seen as determining social behavior, which in turn is a major determinant of
peer status (see also Dodge & Feldman, 1990; Parker et al., 1995; Rubin et al.,
1998). But it is also clear that social cognitions are not affectively neutral. All
cognitions are affectively loaded and the evaluation of alternative strategies is a
process that heavily involves the affective dimension, because these decisions
reflect the individual values and goals of the child. An important task for the
child is to learn to control strong negative emotions so that social interactions
can be carried out in a constructive way. Those who fail to acquire control
mechanisms may be at risk for future peer rejection.
Although Coie (1990) acknowledges the importance of the peer group and the
interaction effects for peer rejection, his main focus is on the behaviors and
internal processes of the rejected child. From a more explicit interactional
perspective, Cillessen, Bukowski, and Haselager (2000) have identified four
explanations for the process of acquiring stable peer status, including both
individual and peer group factors. First, stable child characteristics, such as
aggression or prosocial behavior, might result in stable peer status. Secondly, a
child’s positive or negative interactions with peers may determine the
continuation of his/her earlier acquired peer status. A third explanation pertains
to peers’ social perceptions, attributions and expectations of a child’s
characteristics and behavior based on previous impressions of that child, which
might prevent changes in peer status. Finally, a child’s social self-perceptions
and interpretations of peers’ behavior may contribute to the maintenance of
his/her peer status. Cillessen and colleagues conclude that stable peer status
seems to result from a complex interplay of individual and group factors. Other
authors have also emphasized the importance of the reciprocal nature of
7
children’s social exchanges with their peers, including cognitive and behavioral
components, for peer adjustment (e.g., Parker et al., 1995).
Connection to the present research
Of primary interest for this thesis are the third and fourth phases in Coie's (1990)
process theory of peer rejection, namely the maintenance phase and the
consequence phase. As we have seen, the maintenance phase is a period of
considerable length during which the rejection of the child becomes stabilized
and enduring. The consequence phase includes the short-term and long-term
problems and adversities associated with being rejected by peers. In the research
presented in this thesis, stably rejected children (together with stably popular
and average children) are studied (the maintenance phase) with the aim of
examining the future consequences of being rejected (the consequence phase).
8
Methodological issues
Sociometric classification methods and peer status groups
Introduction. This section focuses on sociometry as referring to a research
method for the measurement of social relations (Hallinan, 1981).
Methodological issues will be discussed with special focus on the stability
problem of peer status groups and on both traditional and new classification
methods, one of which has been used in this thesis (cluster analysis).
The definition of sociometric target groups is partly a method problem; a
question of selecting a suitable sociometric method. Furthermore, included in
the method, a decision has to be made as to which classification scheme and
which numerical criteria should be used for classifying children into sociometric
groups (e.g., Newcomb & Bukowski, 1984; Terry, 2000; Terry & Coie, 1991).
The instrument could be a sociometric questionnaire with general questions
about peer relationship or friendship preferences, such as "Who do you like
most?" or "Who are your best friends?” or more specific, situation-dependent
questions, such as “Who do you want to sit next to in the classroom?” This
distinction goes back to Moreno (1947), who advocated the use of concrete,
task-specific or indirect preference sociometric criteria, such as “play with” or
“sit next to” rather than abstract, direct preference or more multidimensional
criteria, such as “liking” or “friendship” questions (for up-to-date criteria
discussions, see Hymel, Vaillancourt, McDougall, & Renshaw, 2002; Jiang &
Cillessen, 2005; Terry, 2000). Often the choice is limited to certain peers and a
specific number of these peers; for example, three classmates one likes to play
with. Other instruments include a judgment of all peers in a group; for example,
peer ratings (e.g., rating of all peers on a scale from 1 to 5 depending on how
much one wants to play or work with that person) and peer ranking (e.g., rank
ordering of all peers in a school class according to how much one wants them to
join in a class journey). The sociometric use of these instruments, which could
be defined as a process of assessing children’s feelings towards other group
members (“affective sociometry” in the words of Babad, 2001), should be
distinguished from another application, namely, peer assessment. In peer
assessment, the instruments are used to judge characteristics of other group
members, such as traits, behaviors, and achievements (Kane & Lawler, 1978).
Another application is to assess who are perceived as the most popular or
unpopular by the group. It deserves to be mentioned that peer-perceived
popularity has been a rather widely used concept in peer relations research and
has been studied both in its own right and compared with social network
position and sociometric peer status (e.g., Farmer, Estell, Bishop, O’Neal, &
Cairns, 2003; Kosir & Pecjak, 2005; Parkhurst & Hopmeyer, 1998).
9
Over the years a vast amount of literature has shown that sociometric methods
give relatively reliable, stable, and valid information about children's peer
preferences (for a recent meta-analysis, see Jiang & Cillessen, 2005) and that
sociometric choices are consistently related to children's actual interactions with
peers (e.g., Bukowski & Hoza, 1989; Frederickson & Furnham, 1998; Rubin et
al., 1998; Wu, Hart, Draper, & Olsen, 2001). For example, children give positive
sociometric nominations to peers they consider to be prosocial and with whom
they have frequent positive interaction (Gottman, Gonso, & Rasmussen, 1975;
Masters & Furman, 1981; Warden & Mckinnon, 2003), and give considerable
visual attention to (Adams & Roopnarine, 1994; Vaughn & Waters, 1981).
Especially since the early 1980s, there has been a lively discussion about
operational definitions of peer popularity and unpopularity. An essential
drawback of many sociometric risk studies in the past was their confounding of
two conceptually different types of unpopular children: rejected children, who
have no or few peers and who are openly disliked; and neglected children, who
also have no or few peers but who are not unduly disliked (Parker & Asher,
1987). This heterogeneity may yield weaker findings than approaches that separate children into more homogeneous groups (Coie, Dodge, & Kupersmidt,
1990). A large body of research has demonstrated that rejected children, but not
neglected children, are at risk in their social and psychological adjustment (e.g.,
Cantrell & Prinz, 1985; Coie et al., 1990; Cowen, Pederson, Babigian, Izzo, &
Trost, 1973; Ollendick, Weist, Borden, & Greene, 1992; Roff, Sells, & Golden,
1972). It also appears that there is more agreement among peers about the
rejected status than about the neglected status. Crick and Ladd (1989) found that
all sociometric classifications became less accurate when the relative number of
nominators decreased (all classmates, random samples of 90, 80, 70, 60, and
50%), and particularly so for the average and neglected groups. The rejected
group was most resistant to classification errors due to fewer nominators. At the
positive end there has often been confusion between genuinely popular children
and controversial children who are also popular but at the same time disliked by
some peers. Most of the studies referred to above have shown that being popular
among peers is related to good social and psychological adjustment.
Positive and negative nominations. By, for example, adding a measure of
negative peer nominations to the measure of positive nominations, one could
differentiate the actively disliked or rejected children, who get no positive
nominations and many negative nominations, from the neglected children, who
get almost no nominations at all, neither positive nor negative. There also seems
to be, however, difference between neglected and rejected children in the
positive nominations measure. Unlike rejected children, neglected children tend
to be positively nominated by at least one peer (Coie & Dodge, 1983).
10
Furthermore, one can distinguish between the popular group with high “liking”
scores and low “disliking” scores, and the controversial group, which is high in
both dimensions. Roff et al. (1972) have suggested that the controversial
children behave in ways different to the popular children. Bukowski and
Newcomb (1985) found that the controversial group "evidenced higher
variability scores on the liking ratings and on the withdrawal and prosociability
clusters" (p. 1032), which is consistent with their classification.
Finally, there is the large group of children of average popularity in the middle
span of both the positive and the negative nomination dimension. Interestingly,
there is evidence that neglected and controversial peer status groups, identified
by different sociometric classification methods, do not consistently differ from
the average group in their behavioral profiles (e.g., Frederickson & Furnham,
1998).
The validity of the distinction between rejected and neglected children, as well
as between popular and controversial children, is supported by the evidence that
the positive and negative sociometric nomination measures assess relatively
independent dimensions of social status; only low to moderate negative
correlations are typically found between the two measures (e.g., Gottman, 1977;
Wu et al., 2001).
Several multidimensional schemes for classifying children according to their
peer status have been developed (e.g., Asher, 1990; Asher & Dodge, 1986;
Cantrell & Prinz, 1985; Coie et al., 1982; Newcomb & Bukowski, 1983; Peery,
1979; and early precursors such as Dunnington, 1957; Lemann & Solomon,
1952; Thompson & Powell, 1951). By far the most common two-dimensional
classification procedure based on positive and negative nominations is the
standard score method developed by Coie et al. (1982). In that method, the
standardized positive and negative nomination totals are combined to produce an
index of social preference or relative peer likeability (“like most” scores minus
“like least” scores) and an index of social impact or relative social
salience/visibility (“like most” scores plus “like least” scores). These indices are
then used to determine peer status with a classification system based on normal
distribution theory.
Newcomb and Bukowski (1983) criticized the standard score method for not
considering differences in the raw-score distribution between nominating peer
groups. Building on the work of Bronfenbrenner (1945), Newcomb and
Bukowski instead determined social status directly from the positive and
negative nomination totals for each child by a classification based on rare
numbers of liked and disliked nominations according to binomial probability
theory. Other ways of combining the positive and negative nominations in the
classification procedure were developed by Cantrell and Prinz (1985), Cox
11
(1974), and French and Waas (1985), to select a few. It is not surprising that
some discrepancies and incomplete overlap in the peer-status-group
compositions are found as a result of the classification procedure used (e.g.,
Newcomb & Bukowski; Terry & Coie, 1991).
Ethical considerations. The use of negative nominations has, however, elicited
concerns and objections from parents, school officials, practitioners, and some
researchers, and the instrument has occasionally been prohibited by the school
authorities (Putallaz & Wasserman, 1989). They view it as objectionable to
sanction negative nominations and thus, implicitly, negative statements about
others that may increase negative interactions and negative views about rejected
children among peers and lead to increased unhappiness and loneliness (e.g.,
Asher & Dodge, 1986; Rubin et al., 1998).
The few studies that have addressed this issue have not found evidence of
adverse effects on peer relations after the administration of the negative
sociometric measures. This was equally true for preschool children (Hayvren &
Hymel, 1984), as it was for elementary school children (Bell-Dolan, Foster, &
Christopher, 1992; Bell-Dolan, Foster, & Sikora, 1989; Iverson, Barton, &
Iverson, 1997). Thus, the potential ethical risks appear to be less serious than
originally thought. A word of warning is in place because of the limited number
of studies and "therefore, the findings of no obvious harm to children cannot be
considered definitive" (Iverson et al., 1997, p. 111). Furthermore, all studies had
limitations that raise questions about their generality; a fact of which the authors
of the studies were well aware and discussed. Besides, as Bell-Dolan et al.
(1989) noticed, some participants may object to the use of negative nominations
on moral grounds, quite apart from their possible aversive effects on children's
behaviors or feelings. Thus, there seem to be good reasons to proceed cautiously
when considering inclusion of negative nomination measures, as well as good
reasons for a "less controversial procedure" (Asher & Dodge, 1986, p. 448).
Rank orderings and peer ratings. A number of alternative classification
methods are available to distinguish between different sociometric peer groups.
The peer-rating method with a positive-to-negative scale has been a widely
employed choice, either alone or in combination with another measure; in most
cases positive nominations. Asher and Dodge (1986) have found that rejected as
well as popular children could be identified with reasonable accuracy by a
specific transformation of rating-scale scores into negative nominations scores,
which was then combined with positive nominations scores and analyzed
according to the classification algorithm proposed by Coie et al. (1982).
An alternative sociometric method with some advantages is the rank-order
measure, where the individual is asked to rank all members in the sociometric
12
population according to some preference criterion. This method has its place in
sociometric research, although it is not used as frequently as the nomination
measures or the rating scale. Early writers such as Davitz (1955), Horowitz
(1962), and Koch (1933) made use of the rank-ordering scale to differentiate
between low and high sociometric status. Later, Bell-Dolan, Foster, and
Tishelman (1989) developed and used a repeated positive nominations measure
that led to a popularity rank-ordering of pairs of classmates.
The ranking measure gives more information about individual differences than
other methods, because it takes account of all individuals and gives them equal
consideration in the ranking procedure. However, despite its advantages, peer
ranking is seldom used. The method is time consuming and the procedure might
lead to inconsistent placing of participants in the middle of the distribution. The
raters may know those who they like the most and those who they like the least
but be more unsure about the ordering of peers about whom they feel more
indifferent or are unfamiliar with (Hymel et al., 2002; Terry, 2000). Thus, in
using the rank-order method it is important to choose a rather small group, such
as a school class or all same-gender peers in a school class, where all members
know each other relatively well.
Traditional sociometric classification methods – conclusions. Terry and
Coie (1991) have compared several of the most commonly used procedures for
classifying children into peer status groups. The comparison included
classification systems using both negative and positive nominations (Coie et al.,
1982; Newcomb & Bukowski, 1983), the rating scale and positive nominations
(Asher & Dodge, 1986), and solely the rating scale (French, 1988, 1990; Ladd,
1983). The study led to no definitive conclusions, but the results indicated that
some costs in terms of group homogeneity and behavioral distinctiveness are
incurred by using a one-dimensional peer-ratings-based system rather than any
of the two-dimensional systems. The different classification systems gave some
differences in sizes and membership in similar status groups. From their method
comparison, Terry and Coie draw the conclusion that the group selection method
should be dependent on the research purposes it is to serve. For instance, in
prediction studies, with stable peer status groups as predictors, it might be
necessary to select a method that identifies rather large extreme groups.
What seems clear from the above review is the variety of sociometric methods
and classification procedures used in selecting children to different peer status
groups and the relative arbitrary nature of cut-off scores for status inclusion,
which ultimately result in a certain variability, both in group sizes and group
belongings. It seems that cut-off scores for group selections reflect statistical
criteria rather than correspond to an underlying psychological meaning or
functional or conceptual demarcations (e.g., Cillessen & Bukowski, 2000a). As
13
Asher and Dodge (1986) pointed out, status classification systems are just
heuristic devices for studying social adaptation, and the peer groups defined by
sociometric criteria are not discrete entities in the way that, for example,
psychiatric diagnoses are.
Temporal stability among peer status groups. Ollendick et al. (1992)
emphasized that psychometrically sound measures of the construct of
sociometric status and its stability over time are prerequisites for the use of peer
status as a predictive tool. For example, congruent with the theory by Coie
(1990), one might expect that stably rejected children will be of greater shortterm and long-term risk of adjustment problems than temporarily rejected
children (e.g., Kupersmidt, Coie, & Dodge, 1990). A growing body of
confirmatory evidence for this expectation comes from studies of kindergarten
and elementary school children (e.g., Cillessen et al., 2000; DeRosier,
Kupersmidt, & Patterson, 1994; Ladd & Troop-Gordon, 2003; Parke et al.,
1997; Vitaro, Tremblay, Gagnon, & Boivin, 1992).
Many researchers have found that the popular, average, and rejected peer status
categories display a moderate stability that decreases with the total length of the
time span and the number of sociometric data collections (e.g., Bukowski &
Newcomb, 1984; Cillessen et al., 2000; Feinberg, 1964; Ollendick, Greene,
Francis, & Baum, 1991). In one study, stability in peer status decreased from
about 75% to 35% for initially rejected school children when the time interval
increased from 1 month to 2 years (about the same figures were obtained for
popular children) (Newcomb & Bukowski, 1984). In another study examining
preschool boys, half the participants designated as rejected at the beginning of
the preschool year maintained this status at the end of the same preschool year
(Olson & Brodfeld, 1991). In a 5-year longitudinal study covering a period from
late childhood to midadolescence, Coie and Dodge (1983) found that the
rejected status group was more stable than any other extreme group (with the
popular group coming next in stability). The stability of the rejected group
varied from 45% after 1 year to 30% after 4 years. Terry and Coie (1991)
obtained about the same stability percentages in their 2-year follow-up study.
Finally, Vitaro et al. (1992) found that about 20% of rejected children were
stably rejected each year over a 3-year period from kindergarten to grade 2.
While the above studies mainly reported a moderate stability for both the
popular and rejected peer status categories, the controversial and neglected
categories are more volatile over time (Asher & Dodge, 1986; Cillessen et al.,
2000; Frederickson & Furnham, 2001; Newcomb & Bukowski, 1984; Ollendick
et al., 1991; Terry & Coie, 1991). Furthermore, relatively few children are
categorized into these latter groups (Brendgen, Little, & Krappmann, 2000;
Frederickson & Furnham, 2001) and they have not been consistently associated
14
with concurrent and future maladaptive adjustment (Frederickson & Furnham,
1998; Newcomb et al., 1993; Ollendick, et al., 1991). Neglected children tend to
shift in their social status across time to average or sometimes even popular peer
status, but seldom in a more negative direction (Coie and Dodge, 1983). Thus,
the usefulness and representativeness of the neglected and controversial
categories have been questioned (Frederickson & Furnham, 1998, 2001; Hymel
et al., 2002, Newcomb & Bukowski, 1984; Rubin et al., 1990). For example,
Newcomb and colleagues (1993) speculated that neglected children may simply
be selecting a lower level of involvement in the peer group, and, thus, be more
invisible and not so well known by their peers (low social impact). From their
meta-analysis, these authors further suggested that the aggressive behavior of
controversial children appears to be in part balanced and compensated by their
prosocial and cognitive skills.
A few researchers have tried to explore mechanisms and circumstances – such
as social characteristics, self-perceived peer status, locus of control, peer group
norms, and parental monitoring – that enable some rejected children to improve
their social position among peers (e.g., Sandstrom & Coie, 1999). Newcomb and
Bukowski (1984) draw attention to the effects of regression toward the mean
(Nesselroade, Stigler, & Baltes, 1980) as one possible explanation for the lack of
sociometric stability among extreme peer status groups from one data collection
to the next. In any case, the imperfect temporal stability of sociometric
assignments warns us away from using information from only one occasion
when assigning children to sociometric groups. Selecting criteria from two
occasions with a proper time interval between may assure us that stable
sociometric groups will be obtained.
New directions – cluster analysis. Commenting on the moderate stability
figures pertaining to peer status, Bukowski and Hoza (1989) suggested that “to
some extent, ‘at risk’ sociometric status may be either a rather temporary
phenomenon, or it may not be adequately measured by current sociometric
methods” (p. 29). Partly for this reason, sociometric methods have been
reconsidered and new classification methods have been discussed and
introduced into sociometric research (for an introduction, see the editors’ notes
and the articles in Cillessen and Bukowski, 2000b). Among the newly
introduced tools in sociometry are the cluster-analytic methods.
In an effort to circumvent the disadvantages of using traditional sociometric
categorization methods on repeated measurement data, Brendgen, Vitaro,
Bukowski, Doyle, and Markiewicz (2001) applied a semiparametric clustering
technique for continuous longitudinal data, TRAJ (Nagin, 1999). In this way,
they were able to follow the longitudinal course of children’s social standing
among peers over a period of six years, with repeated measurements, and
15
identify longitudinal profiles and three stable groups with different social
preference trajectories. The authors classified their longitudinally stable groups
as the popular group, the average group, and the unpopular group. They were
able to show that there was a relatively good overlap between their social
preference profile groups and the corresponding sociometric categories of
popular, average, and rejected children according to the standard score method
(Coie et al., 1982). Furthermore, their groups differed over the years in
externalizing problem behaviors, so that unpopular children showed the highest
rates of these behaviors and popular children the lowest. The findings of
Brendgen and colleagues are promising, as they found three groups of stable
social standing over time that adhered to earlier sociometric research about peer
status groups and their associations to behavioral problems. One disadvantage
that the authors mention is that the TRAJ procedure assigned everyone in the
sample to one of the identified trajectory groups, which means that the groups
have a rather high within-group heterogeneity.
In another study, Haselager, Van Lieshout, Riksen-Walraven, Cillessen, and
Hartup (2002) used hierarchical clustering according to Ward (1963) to
successfully identify four subgroups of initially rejected boys with different
developmental pathways of prosocial behavior and aggression at three
measurements waves over 5 years. The stable and changing behavioral
trajectories of the different subgroups were associated with different patterns of
sociometric acceptance and rejection over time. The results also showed that the
majority of initially rejected boys were no longer rejected at the end of the
examined time period. Likewise, Frederickson and Furnham (2004) and
Sandstrom and Cillessen (2003) applied cluster analysis in cross-sectional
studies to identify rejected subgroups with different behavioral profiles.
Longitudinal cluster analysis is another way of conceptualizing and empirically
deriving groups of children with different sociometric trajectories compared to
the traditional sociometric classification methods. However, these latter methods
can be applied to validate a cluster approach, as was done in the Brendgen et al.
(2001) study.
A theoretical motivation behind the clustering type of classification-based
methods is given by Bergman and Magnusson (1997). They motivated what they
called a person-oriented approach in the following way: It is based on a model
of individual functioning where the individual is seen from a holistic
perspective. Both internal and external individual factors interact in lawful ways
to optimize individual functioning. In a person-oriented approach, the interest is
on the studied system as a whole, as it is mirrored by the variables under study.
Then each individual’s value profile in these variables comes in focus and this
profile is treated as an indivisible whole in the pattern-oriented statistical
analyses. Since the whole pattern of values is examined and the groups are
16
formed on the basis of having similar patterns within each group, it means that
higher order interactions and nonlinear relationships between the studied
variables are taken into account. This is not the case when using standard
variable-oriented methods. Bergman and Magnusson further argued that there
are often reasons to believe that a small number of typical patterns are in some
sense “optimal” and will therefore often be observed. These can be searched for
using different methods – for instance, by applying a classification method. In
our context, the person-oriented approach leads to an interest in the
classification of the observed longitudinal peer relation patterns into types of
patterns where all subjects belonging to a type have a similar typical pattern. In
the hitherto very few sociometric studies that have taken such an approach
(accounted for above), cluster analysis has been the preferred classification
method. More general examples of the implementation of the person-oriented
approach in empirical psychological research are given by special issues of
Merrill-Palmer Quarterly (Laursen & Hoff, 2006) and New Directions for Child
and Adolescent Development (Peck & Roeser, 2003).
The peer context – school, age, and gender
Schools and school classes as peer contexts. Sociometric peer status
research has been almost exclusively school-based (Gifford-Smith & Brownell,
2003), and classroom and grade sociometry (i.e., social status among classmates
or grade mates, often same-sex) have been the preferred criterion measure. This
criterion has been chosen because schools are important contexts for
socialization where children spend a great deal of time in classrooms and
playgrounds with same-age or near-age peers. Perhaps somewhat provocatively,
Berndt and Ladd (1989) have claimed that “popularity with peers could hardly
be assessed and would probably have little significance if schools did not exist
or did not group large numbers of peers together” (p. 133).
The school milieu can be viewed as a quasi-experimental context, where
children are forced to spend large amounts of time with other children, who they
have not primarily chosen themselves. This is of course especially apparent
when it comes to the school class, where membership is typically an
administrative decision (e.g., Kiesner & Pastore, 2005; Ladd & Kochenderfer,
1996). As time goes by, children develop relations of different kinds with class
peers, and popularity, friendship, and peer network groupings evolve through
intertwined selection and socialization processes (e.g., Kindermann, 1998)
taking place in the particular school class social situation or context. Children’s
close connection to class peers is also reflected in nomination tasks, where it
seems that they predominantly choose class peers in social constructs, such as
popularity or social preference, even when free to choose other school children
17
(e.g., Card, Hodges, Little, & Hawley, 2005; Zakriski & Coie, 1996).
Furthermore, as Zakriski and Coie argued, elementary school children spend
most of their school day with classmates, and, thus, within-class perceptions are
likely to be more stable and reliable than grade-wide perceptions.
The peer group delimiters of age and gender. In sociometric popularity
research, the peers included as raters in the selection criteria are in many cases
of the same gender and close in age. This choice of raters acknowledges the
importance of same-sex and same-age peers and friends in the life and
development of the child (e.g., Aboud & Mendelson, 1996).
Although the nature and structure of children's peer relations may be confused
by similarities in contextual, ecological, or environmental factors (Allen, 1981;
Epstein, 1989; Harkness, 1980), there is considerable evidence supporting the
view that children prefer age-mates as friends (e.g., Ellis, Rogoff, & Cromer,
1981; French, 1984; Furfey, 1927). For example, Hartup (1976) reported that if
there is an adequate number of same-age peers, friendships with age-mates
predominate. Two other studies showed that preadolescent children had more
same-age than cross-age friends across both the classroom and nonschool social
settings (George & Hartmann, 1996; Smith & Inder, 1990).
Likewise, it seems as if a child's primary membership group mostly consists of
same-sex peers (e.g., Benenson, Apostoleris, & Parnass, 1998; Fabes et al.,
2004; Hay, Payne, & Chadwick, 2004; Maccoby, 1990; Smith & Inder, 1990),
and as if there are often more negative evaluations of opposite-sex peers (e.g.,
Duncan & Cohen, 1995; Hayden-Thomson, Rubin, & Hymel, 1987; Singleton &
Asher, 1977). Hartup (1983) stated that children of all ages associate preferentially with members of their own sex, and that this tendency is most pronounced
in middle childhood and early adolescence, but seems to be less dominant later
in adolescence (e.g., Strough & Covatto, 2002). Preschool and school children's
peer contacts are largely sex-segregated, even in contexts where children of both
genders have the opportunity to play together (e.g., Ellis et al., 1981; Fabes et
al., 2004). Card et al. (2005) found that sixth graders nominated same-sex peers
to a higher degree than cross-sex peers on social status items in a school class
context. As with same-age preferences, cultural and contextual factors are
influential (Epstein, 1989), as shown in a study among school children in a rural
community in Kenya (Harkness & Super, 1985).
Hymel et al. (2002) argued that there may be no advantage in sociometric
research to use the more time-consuming practice of including both same-sex
and opposite-sex evaluations. Asher and Hymel (1981), after an empirical
examination of the issue, took the rather firm stand that when examining
children with peer problems, it seems advantageous to use same-sex sociometric
scores, thus excluding the risk that opposite-sex scores might lead to the
18
selection of children who are actually fairly well accepted by their primary
membership group of same-sex peers (or at least not negatively valuated). The
decision to use same-sex scores may also be influenced by a purpose to get as
pure and clearly explicable sociometric groups as possible.
Furthermore, there might be differences in the development and adjustment for
boys and girls, although it seems wise not to over-emphasize gender differences
considering meta-analytic evidence that the genders are more psychologically
similar than dissimilar in most characteristics, with some exceptions pertaining
to primarily aggression (e.g., Hyde, 2005). Newcomb et al. (1993) stressed the
importance of including the evaluation of gender as a moderator variable in
sociometric research “as process-related components of children's peer relations
may vary between the genders, especially in the area of aggressive behavior” (p.
124). This recommendation is consistent with evidence that girls exhibit more
subtle forms of aggression that are more indirect and relational, focusing on
social exclusion and relationship destruction rather than on physical dominance
and control (e.g., Crick & Grotpeter, 1995; Keane & Calkins, 2004). Rys and
Bear (1997) found that, among boys, peer rejection was related to overt
aggression, with relational aggression failing to explain additional variance,
while, among girls, peer rejection was more strongly linked to relational
aggression.
19
Longitudinal peer status research
Peer status and adjustment – an introduction
An essential and significant issue in peer relations research has been to find out
what significance good or bad peer relations have in the life of the child and for
his or her future development and adjustment in adolescence and adulthood.
This issue has, in fact, been a potent stimulus for much of the interest in
children's peer relations (Hartup, 2005; Kupersmidt et al., 1990); as well as for
our longitudinal project.
Peer status and adjustment in adolescence
School adjustment. In acknowledging the importance of the school context as
a social arena for children, it might be assumed that negative class peer relations
are accompanied or later followed by other social and psychological
disturbances in school. Earlier research in this field includes both external
adjustment factors, such as social relations and academic functioning; and
internal adjustment factors, such as psychological well-being and mental health
(for a review, see Rubin et al., 1998).
Cross-sectional and longitudinal studies have demonstrated that peer rejection is
associated with both concurrent and subsequent academic disabilities and
difficulties and below-average intelligence (e.g., Bagwell, Newcomb, &
Bukowski, 1998; Czeschlik & Rost, 1995; Flook, Repetti, & Ullman, 2005;
Ollendick et al., 1992; Wentzel & Asher, 1995). In their meta-analysis of several
studies in the sociometric research area, Newcomb et al. (1993) concluded that
rejected children's academic and intellectual abilities (or cognitive abilities)
were significantly lower than those of other children, while popular children
showed higher levels of cognitive abilities. Although not as extensively
investigated as cognitive factors, athletic competence has also been recognized
as a correlate to popularity. This supports the common notion that athletic
prowess and popularity go together, especially in adolescence and for males
(e.g., Bagwell et al., 1998; Coie et al., 1990; Coleman, 1980; Eitzen, 1975).
From the findings of this research, it seems justified to expect that popular
children might display higher intellectual and athletic skills over the school
years, while rejected children might have academic deficits.
Another sign of school adjustment is how comfortable the child feels when at
school; the child’s school well-being. How do children of different peer status
experience their school situation, what attitudes do they have towards school,
20
and what meaning do school and schoolwork have for them? Cross-sectional
studies have found that rejected children feel uncomfortable and show signs of
loneliness and social dissatisfaction (e.g., Asher & Wheeler, 1985; Boivin,
Poulin, & Vitaro, 1994; Crick and Ladd, 1993). Recently, Flook and colleagues
(2005) performed a longitudinal study and found that teacher-rated peer
acceptance in grade 4 predicted academic self-concept and psychological wellbeing in grade 5, which in turn predicted academic achievement the following
year. These findings give some support to the hypothesis that peer acceptance
has an impact on children’s internal well-being and psychological adjustment in
school over time (as well as on academic performance), although there were also
indications of reciprocal causal paths. The assumption about the subjective
school distress of rejected children needs further corroboration.
An important indicator of social maladjustment is premature school dropout.
School dropout indicates grave school discomfort and distress, and is bound to
affect future adjustment in a negative way. The incidence rate has been shown to
be affected by a multitude of individual, family, peer, and school factors (for a
review, see Farmer, Estell, Leung, Trott, Bishop, & Cairns, 2003). Peer rejection
seems to be one of these factors. Follow-up studies have demonstrated that peer
rejection is associated with later school dropout (e.g., Kupersmidt et al., 1990;
Ollendick et al., 1992; Parker & Asher, 1987), although there are indications that
only those rejected youth who are high on antisocial behavior are at increased
risk for leaving school early (e.g., French & Conrad, 2001). It must be pointed
out that dropping out of school has predominantly been a boys' problem and
therefore seems most suited to differentiate between male participants. For
example, in the large-scale longitudinal project IDA, from which the participants
of this thesis’ studies were selected, about 7% of the boys were early school
dropouts and 1% of the girls (Magnusson, Dunér, & Zetterblom, 1975).
However, some recent research findings suggest that there may have been a
certain leveling out of the gender differences in incidence rate in later years
(e.g., Farmer, Estell, Leung et al., 2003).
Peer adjustment. Peers and friends come forward as very important and
powerful socialization agents and resources for the individual in the social
transitions that take place during adolescence, when youngsters go from
childhood dependency on the family to growing independence and autonomy in
young adulthood (e.g., Fasick, 1984; Hartup, 1995; Hindy, 1980; Sullivan,
1953). It has been found that demographic characteristics, such as age, gender,
and school grade, are attributes that strongly relate to the types of friendship that
are formed in late childhood and up to mid adolescence (Duck, 1975; Kandel,
1978; Kupersmidt, DeRosier, & Patterson, 1995; Smith & Inder, 1990).
Adolescents typically associate with peers that are proximal to themselves in
such demographic characteristics. One might say that same-age, same-gender,
21
same-class, and same-grade peers make up the common and strictly
conventional types of peers and friends of adolescents.
A conventional peer network seems to be one prerequisite for an age-normative
pattern of adolescence adjustment and development into psychosocial maturity
(e.g., Stein & Newcomb, 1999). Thus, there are reasons to be worried about
adolescents who lack these kinds of peers or who associate with
nonconventional peers, such as younger, older, working or antisocial peers.
Associating with younger peers may slow down the process of maturing, while
associating with older and/or working peers may accelerate or deform the same
psychosocial transition process. On the other hand, chronological age is but one
determining factor of psychosocial maturity, and differences in chronological
age do not necessarily imply differences in psychosocial maturity. Furthermore,
having few or no conventional peers may indicate that the adolescent is slow or
ahead of these peers in her or his psychosocial and/or biological development
and, thus, prone to associate with those nonconventional peers that are more
similar to her or him in maturity. Associating with nonconventional peers may
also be an attempt to compensate for difficulties with conventional peers,
whether related to differences in maturity or not. As this short discussion shows,
the question about nonconventional peer relations and psychosocial maturity is
complicated and offers no simple answers. All the same, with the knowledge
that the vast majority of peers and friends are conventional peers, association
with nonconventional peers seems to be of special interest when studying the
relation between different peer contacts and psychosocial adjustment. This is
especially apparent when studying deviant or antisocial peers.
Because of their popularity, popular children have more opportunities to find
friends than other children do, and, with their wealth of positive traits and
positive social actions, they also appear to have the behavioral repertoire that
promotes success with peers (Newcomb et al., 1993; Roistacher, 1974). In
contrast, rejected children are viewed as lacking positive social actions, positive
social traits, and friendship relations (Deptula & Cohen, 2004; Newcomb et al.,
1993). Thus, one might hypothesize that rejected children would have fewer
conventional peers of different kinds in adolescence than other children, while
popular children would have more conventional peers. In a second step,
rejection by classmates may lead to compensatory attempts to associate with
nonconventional peers – in terms of age, educational or vocational situation, and
social characteristics – as a substitute for the problems with conventional peers
(e.g., Bagwell, Coie, Terry, & Lochman, 2000; Brendgen, Vitaro, & Bukowski,
1998; Hoff, DuPaul, & Handwerk 2003).
Furthermore, focusing specifically on nonconventional peers that are antisocial,
early peer rejection has been pointed out as one important antecedent to deviant
or antisocial peer associations in adolescence, (e.g., Bagwell et al., 2000;
22
Brendgen et al., 1998; Coie, Lochman, Terry, & Hyman, 1992; Laird et al.,
2005). As peer rejection has also been shown to affect later antisocial
maladjustment (e.g., Deater-Deckard, 2001; Deptula & Cohen, 2004; Hay et al.,
2004; Laird, Jordan, Dodge, Pettit, & Bates, 2001), the association between
early peer rejection and deviant peer affiliations in adolescence may at least in
part be related to the teenager’s own antisocial behavior (in which case
antisocial peers appear to be rather similar to the rejected child and in that sense
are “conventional” peers). There has been some support for this confounding
selective process in which adolescents prone to antisocial behavior tend to
associate with deviant peers (e.g., Fergusson, Swain-Campbell, & Horwood,
2002; Heinze, Toro, & Urberg, 2004; Kiesner & Pastore, 2005). This might be a
circular or transactional chain (e.g., Sameroff, 1975) where rejected children’s
association with deviant peers may, in turn, amplify their tendencies toward
antisocial behavior in adolescence (e.g., Dishion & Dodge, 2005; MillerJohnson, Coie, Maumary-Gremaud, Bierman, & Conduct Problems Prevention
Research Group, 2002; Parker et al., 1995); a maladaptive process that seems to
be nurtured by the tendency of rejected (and aggressive) children to be biased
towards overestimating the antisocial behaviors of their friends (e.g., Prinstein &
Wang, 2005) and the general tendency to adjust to the aggression mean level of
the group (Boxer, Guerra, Huesmann, & Morales, 2005).
Recent cross-cultural research, involving children aged 7 to 12, has given
support to the usefulness of a complex enhancement model for the development
of antisocial behavior as a combination of affiliation with deviant peers
(according to the selection model) and rejection by the normative peer group (in
line with the socialization model) (van Lier, Vitaro, Wanner, Vujk, & Crijnen,
2005). However, there were also indications of somewhat different pathways
into deviancy for boys and girls. Boys’ antisocial development was highly in
accordance with the enhancement model, while girls’ development was more in
accordance with the socialization model, as they affiliated to a lower degree
with deviant peers but were affected by rejection by nondeviant peers to the
same degree as boys. The lesser importance of deviant peer associations for
females’ future delinquency compared to males has also been corroborated for
middle adolescents in a study by Piquero, Gover, MacDonald, and Piquero
(2005). On the other hand, Laird and colleagues (2001) found no evidence of
this type of gender difference. Instead they found that especially chronic peer
rejection in childhood had a much more pervasive impact on later externalizing
behavior problems in adolescence for both genders, than association with
deviant peers had. The findings in this research area do not converge into any
clear-cut standpoint about causal inferences, but do give support to the
assumption that parallel to being more prone to antisocial development in
adolescence, rejected children might have more deviant peers, possibly with
some gender differences.
23
According to the model presented by Coie (1990), the maintenance phase of
peer rejection has to be of considerable length in order to lead to negative
adjustment in the consequence phase. Thus, a central question is whether the
sociometric status among same-gender classmates undergoes any change or if
the childhood sociometric groups remain clearly separated in popularity in
adolescence. As we have seen, stability estimates have been found to be
moderately high for rejected, average, and popular groups, which would suggest
that peer status remains rather stable up to adolescence, at least for groups with
earlier established peer status stability in childhood.
Several studies have shown that children often have more negative valuations of
opposite-gender peers than of same-gender peers (e.g., Duncan & Cohen, 1995;
Hayden-Thomson et al., 1987) or at least have a greater liking for same-gender
peers (Bukowski, Gauze, Hoza, & Newcomb, 1993). This might indicate that
there are rather small differences in how well-liked the participants in different
peer status groups are by their opposite-gender class peers, owing to the less
positive view most children seem to have of the other gender. On the other hand,
there seems to be consistency between same-gender and opposite-gender
sociometric assessment and the resulting sociometric status groups (e.g., Hymel
et al., 2002; Terry & Coie, 1991). Further study is needed to examine whether,
for example, rejected children are differently preferred by opposite-gender peers
than they are by same-gender peers in adolescence.
Self-perceived peer support and acceptance have in the literature often been
discussed and studied as mediators between actual peer status and social
adjustment. The findings point to the important role of these self-perceptions for
adjustment (e.g., Guerra, Asher, & DeRosier, 2004; Hymel & Franke, 1985;
Pardini, Barry, Barth, Lochman, & Wells, 2006; Wentzel, 2003). For example,
Sandstrom and Coie (1999) found that rejected preadolescents, especially boys,
who believed they were well-liked by their peers actually improved their social
standing among peers over time. Thus, self-perceived peer popularity seems to
be important to include as an additional indicator of adolescence peer
adjustment. Furthermore, this indicator gives an opportunity to examine how
accurate adolescents are in their social information processing when it comes to
judging their actual peer position. Self-perceived peer status has been found to
be relatively correct from third grade (Krantz & Burton, 1986) and, thus, one
might expect it to be fairly accurate for adolescents. However, in a more recent
study, Pardini et al. (2006) found that the association between self-perceived
peer standing and sociometric status was significant but rather low for
aggressive-disruptive fourth graders. As their sample was not a normative group
of school children it seems difficult to draw any far-reaching conclusions from
this finding.
24
Peer status and adjustment in adulthood
Introductory remarks. The adulthood years are of developmental significance
although they have not been as thoroughly explored in developmental
psychology as the childhood and adolescent years (note, however, the increasing
interest in lifespan psychology in recent decades, e.g., Baltes, Staudinger, &
Lindenberger, 1999; Berry & Jobe, 2002; Magnusson, 1996). Counted in time,
adulthood is the longest period in most peoples’ lives and in research contexts, it
is often divided into different age spans or developmental stages, such as young
or emerging adulthood, middle adulthood, and old age (e.g., Arnett, 2000;
Levinson, 1986). Furthermore, adulthood has turned out to be a period of more
individual change than previously supposed (e.g., Helson & Soto, 2005).
While Campbell and Cluss, as early as 1982, suggested that early peer problems
could be among the most robust predictors of poor adult adjustment, the large
majority of prospective longitudinal studies about future effects of problems in
childhood peer relations has so far concentrated on follow-ups in childhood and
adolescence. However, preliminary support for the suggestion of Campbell and
Cluss has come from some relatively recently published studies about the late
adolescence and early adulthood development and adjustment in different life
areas (Bagwell et al., 1998; Bagwell, Schmidt, Newcomb, & Bukowski, 2001;
Nelson & Dishion, 2004; Prinstein & La Greca, 2004; Reinherz, Giaconia, Hauf,
Wasserman, & Paradis, 2000; Woodward & Fergusson, 1999, 2000). For the
development into more mature adulthood, the research situation is quite
different. Because of lack of longitudinal follow-ups for that age group, one has
to rely on findings for younger age groups and from these findings derive
hypotheses about how sociometric status in childhood might be associated with
midadulthood adjustment.
Young adulthood adjustment. Young adulthood is often seen as an era of
freedom, identity exploration, possibilities, and instability. It has its costs in the
form of increased rates of many risky and antisocial behaviors, which seem to
peak during that time period and then decline substantially with the familial,
parental, and occupational responsibilities of more mature adulthood (e.g.,
Arnett, 2000). These antisocial and other risk behaviors seem to be important
indicators of adjustment in young adulthood. Earlier empirical studies have
found an association between childhood peer status and future antisocial
behavior – such as criminality and substance abuse – up to late adolescence and
the early twenties for both females and males. This association seems mainly to
be due to aggressive and externalizing behavior problems in childhood. Children
with elevated aggression may find it difficult to gain acceptance among peers
and their antisocial problems can grow over time as a result of their rejection
and lead to long-term adjustment problems (e.g., Deater-Deckard, 2001; Nelson
25
& Dishion, 2004; Reinherz et al., 2000; Rubin et al., 1998; Woodward &
Fergusson, 1999).
Mental health problems, especially internalizing problems, are far more
common among women than among men from late adolescence onwards (e.g.,
Crick & Zahn-Waxler, 2003; Culbertson, 1997; Nolen-Hoeksema & Girgus,
1994) and, thus, are an important indicator of female adjustment. The empirical
support is rather mixed for a link between childhood peer status and
psychopathology in late adolescence and early adulthood (e.g., Bagwell et al.,
1998; Bagwell et al., 2001; Reinherz et al., 2000; Woodward & Fergusson,
1999). Considering the seriousness of psychiatric problems, further studies are
motivated. In connection with this problem area, it seems worth mentioning that
an eventual association between peer rejection and subsequent internalizing
problems might partly be explained by withdrawn behavior with peers (e.g.,
Deater-Deckard, 2001; Rubin et al., 1998).
Midadulthood adjustment. Midlife is often characterized by a striving to
succeed in vocational life, at the same time as handling the tasks of daily family
life. Most adults experience having children and seeing them grow up and
gradually strive for independence. They may also have to be concerned about
the needs of their elderly parents. Furthermore, middle-aged adults may have to
face the prospect of giving up some of their early dreams and aspirations and
many adults may experience personal shortcomings and perceptions of personal
inadequacy “making one feel socially unaccepted and rejected and thus lonely”
(Rokach, 2000, p. 77). In short, midadulthood is a period with many demands,
challenges, rewards, and decisions of crucial importance to development and
adjustment, or what in the tradition of Erikson’s (1963) theory of life stages
could be seen as developmental challenges and tasks to be solved in the
generativity life stage.
This may be especially salient for women, who, in Western societies, still have
the main responsibility for the home and the children, as well as for elderly
relatives, at the same time as they are fully available on the labor market (for a
review of the changes in women’s roles in the last few decades, see Aube,
Fleury, & Smetana, 2000; also Halpern, 2005; Meleis & Lindgren, 2002;
Rokach, 2000; Schultheiss, 2006; Woods & Mitchell, 1997). Research findings
have pointed to a negative association between work-family conflict and job and
life satisfaction, particularly among women (e.g., Kossek & Ozeki, 1998; Ozer,
1995; Peake & Harris, 2002). The life satisfaction of midadulthood women
seems to be an important aspect of their adjustment and self-identification (e.g.,
Meulemann, 2003), and closely associated with important life areas such as
family and work.
26
Cross-sectional and longitudinal evidence links having friends with
psychological well-being and life satisfaction from childhood to old age (Hartup
& Abecassis, 2002; Hartup & Stevens, 1997; Ladd, 1999). Having friends seems
to be associated with feelings of self-worth or self-competence, to ward off
feelings of loneliness, and to provide children with competences for future close
relationships (e.g., Gifford-Smith & Brownell, 2003; Hartup & Abecassis, 2002;
Laursen, 1996; Smollar & Youniss, 1982; Sullivan, 1953). Furthermore, the
association between rejected peer status and inadequate interpersonal or social
skills has been firmly established by many research findings (e.g., Cantrell &
Prinz, 1985; Casiglia, Lo Coco, & Zappulla, 1998; Coie et al., 1990; Rys &
Bear, 1997). It is, thus, quite conceivable that childhood peer rejection may be
particularly associated with difficulties in those future adjustment tasks that
require adequate social skills, as in the task of relating to other people.
As we have seen, satisfaction with family life is vital for midlife women, and
Parker and Asher (1987) have suggested marital adjustment as one of the
outcomes of interest in future peer status risk research. Earlier research
pertaining to this issue is sparse and does not specifically focus on rejected
children. Existing evidence suggests that children with peer problems either do
not marry, or if they marry, experience difficulties in their marriages that may
lead to divorce (for a literature review, see Parker & Asher, 1987). Kupersmidt
et al. (1990) argued that marital adjustment and the ability to sustain intimate
friendship are forms of behavior that are most affected by social deficiencies.
Ladd (1984) has suggested that:
...if it were the case that low, stable scores on a friendship nomination
measure were a sign of a child's failure to form friendships and learn many
of the skills needed to establish and maintain intimate relationships (e.g.,
loyalty, trustworthiness, conflict resolution), then information from this
measure might also be useful for predicting his or her success in adult
intimate peer relationships (e.g., marriage). (p. 319)
In accounting for research about social relationships and their influence on
primarily physical, but also mental, health, Cohen (2004) argued that different
aspects of social relationships, especially social support, social integration, and
negative interaction, are associated with health outcomes. She recapitulates
findings from prospective and retrospective studies of both student and adult
samples, and concludes that there is often a significant association between
social environment variables and health, depending on both health-promoting,
stress buffering mechanisms and disease-promoting mechanisms. In her own
studies, she found that this association was reduced when controlling for social
personality traits, especially sociability, but also that it remained substantial and
significant nevertheless. Although the findings accounted for by Cohen cannot
be seen as direct evidence for a relationship between childhood peer status and
27
adulthood health, they do give some strength to such a hypothesis, supplemented
by the assumption (not tested in this thesis) that mediators might be later social
relationships up to adulthood.
There is some evidence in support of the view that low-accepted children have
problems with their later occupational adjustment ( Janes, Hesselbrock, Myers,
& Penniman, 1979; and the literature review by Parker & Asher, 1987).
Kupersmidt et al. (1990) are of the opinion that different problems at the place
of work might be associated with prior social rejection. Existing evidence of the
academic problems of rejected children (e.g., Newcomb et al., 1993) indicates
that they finish school earlier than other children and therefore get rather
unqualified jobs. Some recent empirical findings (e.g., Bagwell et al., 1998,
2001; Woodward & Fergusson, 2000) have extended our knowledge by showing
that childhood peer status is associated with educational and job success in late
adolescence and early adulthood.
The above empirical findings and theoretical suggestions lead to the conclusion
that childhood peer relations may be associated with subjective well-being and
life satisfaction in important middle adulthood life areas pertaining to family
relations and other social relations, as well as to work and health. However, one
argument against our suggestions about an association between childhood peer
status and midadulthood adjustment is that the more stable life situation in
middle adulthood for most women (Arnett, 2000; Levinson, 1986) should result
in lower rates of adjustment problems compared to those experienced in young
adulthood, and that any potential impact from childhood peer relations would be
diluted. To test such a proposition, one would need to have indicators in at least
one adjustment area for both young adulthood and midadulthood.
In the light of the multiple outcomes that have been of interest in sociometric
research, a plausible hypothesis is that vulnerabilities from past life may
resurface as different symptoms and difficulties in handling various aspects of
the individual’s current adult social life. To be more specific: Unsuccessful peer
relations in childhood and adolescence may instigate social vulnerabilities that
carry over into negative implications of different kinds for midadulthood
adjustment. For example, Kupersmidt and Coie (1990) found that peer rejection
in childhood was not a strong predictor of late adolescence specific outcomes,
but they found a stronger link between rejection and some kind of negative
outcome. They interpreted their findings as consistent with arguments that,
because peer rejection may provide a measure of stress operating on children, it
will be better at predicting the outbreak of later disorder rather than its specific
form (e.g., Sroufe & Rutter, 1984; see also Parker et al., 1995).
28
Peer status and concurrent correlates – importance for later adjustment
In sociometric research there has been a considerable interest in relating
children’s peer status assignments to contemporaneous behavioral reactions
(e.g., Asher & Coie, 1990; Coie et al., 1990; Furman & Buhrmester, 1985;
Hymel et al., 2002; Ladd, 1999, 2006; Newcomb et al., 1993; Pope, Bierman, &
Mumma, 1991). Two problem behaviors have been frequently related to status
among peers and shown to be major obstacles to peer acceptance – aggression
and shyness/withdrawal – and it seems well worth to dedicate some space here
to a discussion of these behaviors.
Research findings regarding aggressive behaviors and withdrawal have largely
been supportive of the first and second assumptions put up by Coie (1990) as a
basis for his theoretical model about the peer rejection process (see above).
According to the first assumption there is a close association between negative
social behaviors and peer rejection. The second assumption includes a statement
that temperamental differences might play a role for these behaviors (e.g., Hay
et al., 2004; Walker, Berthelsen, & Irving, 2001), albeit Coie places more
importance on the child’s socialization history. Rubin et al. (1990) have
presented two developmental pathways or precursors to peer rejection that can
be seen as more fine-graded accounts of what Coie proposed in these two
assumptions and in the two first phases of his theory – the precursor phase and
the emergent status phase. One pathway is closely connected to early aggression
and the other to a more withdrawn behavior pattern. In interaction with negative
family patterns, these developmental pathways may create a platform for the
development into being rejected by peers. Although speculative, the
developmental scenarios suggested by Rubin et al. point to the importance of
early aggression and withdrawal as potential determinants of peer rejection.
Overt and relational aggressive behaviors have been associated with problematic
friendship relations and peer rejection among both boys and girls (Brendgen,
Vitaro, Turgeon, & Poulin, 2002; Grotpeter & Crick, 1996; Vaillancourt &
Hymel, 2006; Werner & Crick, 2004). The combination of peer rejection and
aggressive behavior seems to be extremely negative for future adjustment (e.g.,
Coie et al., 1992; Hymel et al., 2002; Kupersmidt & Coie, 1990; Ladd, 2006;
Parker & Asher, 1987; Parkhurst & Asher, 1992). Withdrawn, anxious behaviors
seem to be especially salient to the peer group from late childhood onwards and
have consistently been linked to peer rejection, especially for girls (e.g., Coie et
al., 1990; French, 1990; Hartup & van Lieshout, 1996; Ladd, 2006; Rubin,
Burgess, & Coplan, 2002; Rubin et al., 1998) and in different cultures (Valdivia,
Schneider, Chavez, & Chen, 2005).
Negative behavioral reactions, such as aggressive behavior or social withdrawal,
may also alienate peers and restrict a child’s access or possibility to make
29
friends (Schneider et al., 2001). Lately, some findings have indicated that both
verbal (communicative) and behavioral aspects of withdrawal and aggression
are important for peer acceptance (and self-perceived social competence), both
concurrently and longitudinally, among adolescents (Chang et al., 2005).
Furthermore, it seems that the negative and prosocial behaviors that predict
membership of a specific peer status category also predict stable membership in
that category, so that, for example, aggressive behavior predicts both incidental
and chronic rejection (e.g., Cillessen et al., 2000). Externalizing and
internalizing behavior problems together with peer rejection – and also other
peer problems, such as harassment and victimization (e.g., Cullerton-Sen &
Crick, 2005) – seem to form an intriguing and interrelated developmental
pattern, where concordance has not yet been reached about causal paths and
where bidirectional and reciprocal influences seem to be the most likely
explanation (e.g., Zimmer-Gembeck, Geiger, & Crick, 2005). For example, in
the victimization process, some rejected children might turn out to be either
bullies or victims, depending to some extent on their level of aggressive or
withdrawn behavior (e.g., DeRosier & Thomas, 2003; Warden & Mackinnon,
2003). The associations between peer rejection, aggression or withdrawal, and
victimization (or the bullies/victims problem) are an important and somewhat
neglected research task with potential implications for interventions. However,
these developmental associations are not the research topic of this thesis. As we
have emphasized, the focus is here on the aspect of peer relations pertaining to
peer status and its role for adjustment.
Coie (1990) stressed the importance of deficits in academic achievement and
athletic ability for peer rejection, especially when mediated by specific negative
behavior reactions. Research findings have repeatedly associated cognitive
abilities, and academic achievement and motivation with peer relations (e.g.,
Ladd & Kochenderfer, 1996; Newcomb et al., 1993). Peer rejected and, to some
extent, friendless children have exhibited lower academic success than other
children (e.g., Green, Vosk, Forehand, & Beck, 1981; Hatzicristou & Hopf,
1996; Ladd, 1990) and more negative school attitudes and dislike of school
(Ladd, 1990; Ladd & Coleman, 1997).
A child’s family social background or socioeconomic class might affect both the
peer status of the child and the strength of an association between childhood
peer relations and later adjustment (Eisenberg, 1998; Vitaro, Tremblay, Gagnon,
& Pelletier, 1994; Woodward & Fergusson, 2000). For example, Woodward and
Fergusson found that associations between childhood peer problems and later
educational and job disadvantage were in part explained by sociofamilial
background variables, such as socioeconomic adversity.
In addition to being associated with peer status, all of the above discussed, and
potentially confounding, categories (aggression, withdrawal, academic
30
achievement, and family social background) have been found to be unique and
combined predictors of later adjustment in different life areas (e.g., Hinshaw,
Lahey, & Hart, 1993; Ladd, 2006; see also handbooks such as Cicchetti &
Cohen, 1995; Damon & Eisenberg, 1998). In conclusion, these adjustment
dimensions correlate to peer status and, furthermore, have independent
explanatory values in relation to children’s future adjustment. Thus, in including
them as confounding factors, it is possible to shed some light on the
longstanding issue as to whether peer relations have an independent significance
for future adjustment or whether it is mainly a marker of more basic, underlying
behavioral and social processes (Bagwell et al., 1998; Eisenberg, 1998;
Kupersmidt et al., 1990; Parker & Asher, 1987; Parker et al., 1995; Vitaro,
Larocque, Janosz, & Tremblay, 2001; Woodward & Fergusson, 2000). This
issue was also addressed above in the discussion about the peer rejection model
proposed by Coie (1990). I will continue this discussion in the next section.
Explanatory models
Coie’s (1990) fourth assumption, underlying his theoretical model, suggests that
rejection causes disorder or sets up conditions leading to disorder. Yet, as he
also mentions, it is equally plausible to argue that peer rejection serves as a
marker variable for the psychological antecedents of disorder and, thus, not as a
direct cause. However, Coie maintains – for the purpose of debate as he says –
that prolonged experience of peer rejection is a major contributor to subsequent
disorder. He develops this standpoint referring to Albee's equation for the
incidence of disorder (Albee, 1984). Albee proposes that the incidence of
disorder increases as a result of stress, organic variables, and exploitation
factors, and that coping skills, self-esteem, and social support serve to reduce
disorder. In the Albee equation, peer rejection can be interpreted as contributing
to disorder in several ways. Peer rejection is in itself an element of stress for the
child (rejected children tend to be badly treated by peers and consequently may
retreat from social activities with peers, but continue to serve a role for the
group as a group deviant). Rejection can also contribute to having fewer
resources for countering the factors leading to disorder (resources such as
support groups, coping skills, and self-esteem). Popular children, on the other
hand, might have plenty of social resources as a buffer against and tool to reduce
the risks of a troubled development.
However, although research findings point to an association between peer
problems in childhood, especially rejection, and maladjustment in adolescence
and early adulthood, this need not be interpreted as a causal chain. What is
known is that a correlational continuity exists, but the mechanisms behind it are
fairly unknown (Bukowski & Adams, 2005; Kupersmidt et al., 1990). It could
31
be that early peer rejection reflects the child's difficulties in a more general sense
and that these difficulties also interfere with functioning in other life areas and
in different ways over time, or, in the words of Parker and Asher (1987), that
some underlying, continuously present behavioral process is directly or
indirectly responsible for both problematic peer relationships and for later
maladaptive functioning. In such an interpretative pattern or incidental model,
peer rejection might be seen as a marker, but not as a causal variable, which was
also discussed by Coie (1990).
In other words, which model best explains the associations between peer
relations and later adjustment – a causal model or an incidental model? An
example supporting an incidental model is the findings of Vitaro et al. (2001),
which suggest that unpopularity and friendlessness do not directly contribute to
school dropout, but are markers of personal problems such as aggression and
difficulties at school (see also French & Conrad, 2001). Some of the findings in
a study by Henricsson and Rydell (2006) also point to the possibility that peer
rejection is at least partially a result or marker of internalizing or externalizing
problems. On the other hand, some other studies have found unique links
between peer rejection and later psychological and academic adjustment
problems (e.g., Coie et al., 1992; DeRosier et al., 1994). Woodward and
Fergusson (2000) found that child and family confounding factors substantially
reduced the relation between teacher-rated peer problems in childhood and late
adolescence educational and occupational disadvantages, but that there was still
a significant association between early peer problems and later academic
difficulties. The authors’ conclusion was that their results were in support of
both the incidental model and the causal model. In another study about
childhood peer problems and late adolescence psychosocial adjustment,
Woodward and Fergusson (1999) found more clear-cut evidence for the
incidental model. Ladd (2006), by building on earlier research evidence, has
found support for an additive model (controlling for some transactional and
disorder-driven models), where both peer group rejection and aggressive or
withdrawn behavior make at least partially independent contributions in an
additive way to later externalizing or internalizing problems over seven
consecutive one-year intervals across ages 5-12.
Bagwell et al. (1998) have discussed both the causal and the incidental models,
but also an intermediate model, which they see as potentially more useful. In
this model, it is claimed that children's experiences with peers moderate the
relation between risk variables and maladjustment or that peer problems interact
with other risk factors and increase or decrease the risk for negative outcomes
(also Bierman, 2004; Dodge et al., 2003; Masten, 2005; Parker et al., 1995;
Rubin et al., 1998; Salmivalli & Isaacs, 2005). Peer rejection may interact with
existing difficulties of the biologically or socially vulnerable child and thus
increase the risk of negative outcomes. Good peer relations, on the other hand,
32
may be a protective and preventive factor decreasing the influences from other
more basic risk variables of genetic, social or ecological origin. Peer relations
can be seen as both predicting later adjustment, and as having a history and
background that has led to the present peer situation. Or, as Bukowski and
Adams (2005) have pointed out, the influence that peers have on individual
development derives from a wide range of sources and processes inside a peer
system with multiple levels of social complexity that are linked in multiple
ways.
33
Aims of the thesis
The present thesis focuses primarily on different aspects of adjustment in
adolescence and young and middle adulthood for individuals differing in stable
sociometric standing among peers in elementary school; that is, being of stably
rejected, popular or average peer status. Data from the Swedish research
program Individual Development and Adaptation (IDA) (e.g., Magnusson et al.,
1975) were used and the included research has a longitudinal and prospective
direction. The longitudinal perspective is of considerable length. The first period
has a time span of about five years covering the period from the early
elementary school years to midadolescence and the second period has a time
span of 33 years from early elementary school to midadulthood. In the choice of
time periods for selecting sociometric groups, I am mainly in agreement with a
common opinion in peer relations research that peer acceptance is most
important over the years from middle childhood into early adolescence, and that
peer rejection during that time period seems to be associated with potential
disorder in adolescence and adulthood (e.g., Coie, 1990; Ladd, 1999; Parker &
Asher, 1987).
The following specific aims were focused on:
1. The first aim was to prospectively examine the midadolescence
academic, psychological, social, and peer adjustment of one-year stably
peer rejected, popular, and average children (Studies I and II). Earlier
research findings indicate that rejected children would develop more
adjustment problems in these life areas than children with better peer
status. It was further expected that popular children would develop fewer
problems than other children.
2. The associations between three-year stable childhood peer status and
young and middle adulthood adjustment for women were investigated
(Study IV). The young adulthood adjustment in the examined antisocial
domains of criminality and alcohol abuse was expected to be worse for
rejected girls than for other girls, while a positive adjustment was expected
for popular girls. For young adulthood mental health, no hypothesis was
formulated due to the mixed findings of earlier research. The examined
midadulthood outcomes were subjective well-being and life satisfaction in
different life domains (job, social relations, family, general health), alcohol
consumption, and vocational career. Due to the lack of earlier sociometric
at-risk studies for midlife adjustment, the expectations of a negative
adjustment for rejected girls (and a positive adjustment for popular girls) in
these life domains could only be tentative and based on findings from
34
other, more short-term studies. Childhood aggression, withdrawal,
academic achievement, and SES, which were included as control variables,
might explain some of the eventual associations and, in that case, give
support to an incidental explanation of the impact of childhood peer status
on adulthood adjustment.
3. The three-year, stable peer status groups for the adulthood follow-up
were identified by introducing a modern clustering technique (e.g.,
Bergman, 1998) into sociometric research. This person-oriented approach
has been very sparsely applied in sociometric research. Thus, as a third
aim, the usefulness of cluster analysis as a sociometric classification
instrument was examined (Study III). The clustering technique was
validated against the frequently used sociometric classification method
developed by Coie et al. (1982), as well as against some control variables
that have been shown to be associated with peer status in childhood.
35
Summary and methodological integration of the
empirical studies
Study I: School adjustment in adolescence for previously rejected,
average, and popular children
Introduction and methods. The central issue of the present study pertained to
the midadolescent school adjustment of children of stable same-sex rejected,
average, and popular class peer status in late childhood. The school domains
examined were academic achievement, subjective school experiences, attitudes
towards school, and school dropout. The general hypothesis was that rejected
children in adolescence would show a poorer adjustment in different school
domains compared to other children. In line with earlier research about the
beneficial effects of good peer relations, a further hypothesis was that popular
children would display a positive school adjustment in adolescence.
The participants of this study were boys and girls from the main group of the
research program IDA (Magnusson et al., 1975). Sociometrically homogeneous
peer status groups of one-year stably rejected, popular, and average boys and
girls were identified by Adebäck (1969) in a procedure using a sociometric
ranking measure and a positive nominations measure in grade 3 (mean age 10
years) and a positive nominations measure in grade 4 (mean age 11 years). Each
peer status group consisted of 15 participants. This classification method has
certain original elements, but is also in line with traditional classification
methods in the sociometric research domain (for reviews, see Terry, 2000; Terry
& Coie, 1991). The midadolescence school adjustment measures were taken
from the large investigation that was carried out in grade 8 (mean age 15 years)
for the whole main group of the IDA research program (Magnusson et al.,
1975). School marks in Swedish and physical training, a standardized
achievement test in Swedish, intelligence scales, and self-reports about attitudes
to school were used as outcome measures. Premature school dropout rate was
also included, but only for boys (as this has mainly been a male problem).
ANOVAs and chi2 tests were applied to compare the peer status groups.
Attrition analysis showed that the midadolescence missing cases did not differ
significantly in sociometric values from the mean value of the status group they
belonged to.
Main results and conclusions. The academic performance and intelligence
level of rejected boys and girls were short of the standards of children from the
other status groups, while the scores of popular boys and girls were of superior
standard. Post-hoc analyses showed that rejected and popular groups of both
36
genders differed significantly from each other in most cases. While some studies
have found gender differences in academic outcomes, the findings here were
more in line with other prospective studies that have found no gender
differences in long-term academic achievement for different peer status groups.
There were slight, non-significant indications that rejected girls (but not rejected
boys) had negative attitudes towards school and schoolwork, and that popular
girls had positive attitudes. The school dropout rate of rejected boys was
significantly higher than that of other boys. Furthermore, those rejected boys
who were school dropouts seem to have had a lower peer status initially (in
grade 3) even when compared to other rejected boys, and to have been absent
from school more frequently before they dropped out. These might be signs of a
negative social outcasting process resulting ultimately in their dropping out of
school.
The findings indicated that the rejected children are a risk group for school
problems over a longer period of time. The causal chain is unclear, but peer
rejection combined with other negative school factors, might jointly function as
an ostracizing mechanism and drive the individual into an outsider position.
Considering the important developmental aspects of the adolescence years, there
appear to be good reasons to worry about the future adulthood adjustment of
peer rejected children. Unfortunately, the small group sizes are a limitation that
reduces the power of the statistical analysis.
Study II: Childhood peer status as predictor of midadolescence peer
situation and social adjustment
Introduction and methods. The participants and sociometric groups of this
study were the same as in Study I. The intention was to examine conventional
and unconventional peer relations in midadolescence (age 15), including deviant
peer relations and sociometric peer status among pupils of both genders in the
class. Antisocial outcomes in midadolescence were also studied. The general
hypothesis was that rejected children would have fewer conventional peers of
different kinds in adolescence than other children, while popular children would
have more conventional peers. Rejected children might try to compensate by
associating with unconventional peers such as different age and working peers,
but especially with antisocial peers. As association with antisocial peers has
been found to be related to an individual’s own antisocial activity in
adolescence, it was further hypothesized that rejected children would be more
antisocial than other children. Stability of same-gender sociometric status up to
midadolescence was also examined, as well as opposite-gender sociometric
status and self-perceived peer status. Finally, gender differences were studied.
37
The midadolescence peer relations and social adjustment measures were, as in
Study I, taken from the large follow-up that was carried out in grade 8
(Magnusson et al., 1975). A peer rank-ordering instrument was used to obtain
adolescent same-gender and opposite-gender sociometric status in the school
class. Furthermore, the adolescents estimated their own ranking according to
how they thought their same-gender classmates had ranked them. From a selfreport instrument about peer network, it was possible to obtain the number of
conventional and unconventional peers of different kinds. Information about
antisocial peers and the participants’ own antisocial activity was obtained from a
norm questionnaire (Chinapah, 2000; Henricson, 1971; Magnusson et al., 1975),
where values, intentions, and behaviors related to norm-breaking situations were
measured. In addition, an index of the self-experienced peer situation (index of
friendship, loneliness, and popularity) was constructed from items in the norm
questionnaire. In order to examine sociometric status and gender differences in
midadolescence outcome variables, two-way analyses of variance (ANOVAs)
and logistic regression analyses were conducted. Midadolescence attrition cases
did not differ significantly in childhood sociometric values from the mean value
of their peer status group.
Main results and conclusions. The findings revealed that rejected children
maintained their lower popularity with same gender classmates by differing
significantly from popular and average children in midadolescence, while
popular children had a significantly higher peer status compared to children in
the other peer status groups. An important conclusion was that the different peer
status groups seem to follow different sociometric paths and, at the group level,
preserve their peer status position over several years (this result was one
motivation for extending the classification period for stable peer status to 3 years
in the adulthood follow-up in Study IV).
The same peer status group differences were also found for opposite-gender
popularity in midadolescence (with somewhat smaller but nevertheless high
effect sizes). Furthermore, rejected children rated themselves as less popular
among same-gender class peers in midadolescence than average and popular
children. Accordingly, rejected children seem to be aware of their low status
among peers at age 15.
Adolescents of both genders had most of their peers in conventional peer
categories, that is, among schoolmates, same-age peers, and classmates. Some of
the peer status group differences pertaining to relations with conventional peers
were significant. Rejected participants reported fewer school peers than both
average and popular participants and fewer same-age peers than popular
participants. These findings are indications that peer rejection in late childhood
has an impact on the conventional peer network in midadolescence. As
38
adolescents, rejected children may not succeed in finding a place among those
peers that are close to them in terms of social characteristics and age.
Furthermore, they experienced their peer situation as worse than average
children and as much worse than popular children. These findings, together with
the findings that they were still rejected in adolescence, give a picture of a rather
negative peer situation for rejected children in adolescence, as judged by both
peers and themselves.
There were, however, no indications that rejected children tried to compensate
for these negative peer features by associating to a higher degree with
nonconventional peers – in terms of age, occupation, or antisocial characteristics
– than other children did. Furthermore, rejected children did not report more
norm-breaking deviancies in adolescence than other children did. While some
authors have found significant associations between a child’s own delinquent
behavior and the delinquency of its peers and that childhood peer rejection
seems to trigger such a development in adolescence, this was not the case for the
rejected children of the present study. This is a hopeful result. At least up to
their midadolescence years, rejected children did not seem to have developed
strong antisocial tendencies or associations with deviant peers.
No significant gender differences or interaction effects between gender and peer
status were found for midadolescence sociometric or peer network variables.
These findings point to the possibility that the same social processes are at work
for both genders and give similar outcomes. For data about own and peers’
antisocial tendencies there were no interaction effects but significant gender
effects in all variables. Boys were more deviant in the measured antisocial
aspects, and they also judged their peers as more antisocial than girls did. This
result was not unexpected, as boys are generally more prone to antisocial
attitudes and activities than girls (e.g., Barriga, Morrison, Liau, & Gibbs, 2001;
Laird et al., 2005).
Study III: Cluster analysis in sociometric research: A pattern-oriented
approach to identifying temporally stable peer status groups of girls
Introduction and methods. A prerequisite for the utility of sociometric peer
status as a predictive tool seems to be its stability over time. In an effort to
circumvent the possible disadvantages for stability of using traditional
sociometric categorization methods on repeated measurement data, a few
researchers have applied some kind of cluster analysis to identify longitudinal
profiles and stable groups with different social preference trajectories (Brendgen
et al., 2001; Haselager et al., 2002). An overarching aim of the present study
was to follow a similar selection procedure by using a clustering technique to
39
group girls with stable and volatile sociometric peer status over the late
childhood and early adolescence years.
Cluster analysis within the framework of the LICUR procedure (Bergman,
1998) was applied to the group consisting of all girls in the main cohort of the
IDA research program (Magnusson et al., 1975) who had sociometric rankordering data at age 10 (grade 3) and at age 13 (grade 6) (N = 445). The
sociometric ranking scores, transformed into “like most” and “like least” scores,
were the basis for the longitudinal cluster analysis. The cluster solution was then
validated against the peer status categories obtained by the frequently used CDC
classification method (Coie & Dodge, 1983; Coie et al., 1982). In a further
validation effort, the clusters were associated with some childhood control
variables that have been frequently and significantly related to differences in
peer status: aggression and withdrawal (teacher ratings), academic success
(grade point averages), and family background (SES).
Main results and conclusions. The cluster analysis resulted in three
temporally stable clusters identified as rejected, popular, and average, which
were larger than corresponding CDC classified groups. A further finding was
that inside their stable peer status assignments, rejected/popular girls became
increasingly rejected/popular from age 10 to age 13. While no stable clusters of
neglected or controversial girls were identified, four temporally volatile clusters
were found. In validating the cluster solution, it was shown that all stably
rejected and all (but one) stably popular CDC girls were members of the
corresponding cluster. Furthermore, rejected and popular clusters showed the
same associations to the childhood correlates as has been documented by earlier
sociometric research: Girls in the rejected cluster showed the most negative
adjustment (and lowest SES) and girls in the popular cluster showed the most
positive adjustment (and highest SES) in these control variables.
Most traditional sociometric classification methods have arbitrary and
conventional rules and cut-off scores, and could be seen as some kind of
heuristic devices for assigning social status to a given individual (Asher and
Dodge, 1986; Cillessen & Bukowski, 2000a; Rubin et al., 1998; Terry and Coie;
1991), which might lead to rather artificial groupings. By using the clustering
technique, one makes use of other information from the studied variables,
including interactions and nonlinear relationships. The clusters are, in a way,
“natural” groupings, with membership in a cluster as a consequence of the
individual’s “holistic” place or value profile in the space of included variables.
The crucial point in analyzing the results of a cluster analysis is to scrutinize the
cluster solution in order to arrive at theoretically meaningful interpretations of
the clusters with reasonable correspondence to other results in the specific
research area. This scrutiny has been the main purpose and content of the
40
present study and the findings support the empirical and theoretical
meaningfulness of applying cluster analysis as a classification method in
sociometric research.
Study IV: Girls’ peer status and their adulthood adjustment: A
longitudinal study from age 10 to age 43
Introduction and methods. Although there has been a change in recent years,
the psychology of peer relations is still to a high degree based on boys’
development; the role of girls’ peer relations for their development and
adjustment is less well explored and understood (e.g., Bagwell et al., 2000;
Ladd, 1999). From what we know, girls’ peer relations cannot be seen as
blueprints of boys’ peer relations. For example, girls seem to be more allocentric
than boys, that is, they tend to be affiliate more, be more cooperative, and
oriented towards sources of social support (e.g., Dayan, Doyle, & Markiewicz,
2001; Yamaguchi, Kuhlman, & Sugimori, 1995). This motivates an interest in
girls’ development and to assemble evidence on how good and deviant peer
relations affect girls’ later adjustment. The present study intended to contribute
to this endeavor by examining the importance of girls’ stable peer status for
adjustment in young and middle adulthood; a follow-up period that has been rare
or absent in past sociometric at-risk research.
Stable age 10 to age 13 peer status clusters (see Study III for details about the
applied cluster analysis and peer status stability) of girls were identified from the
main group of girls in the IDA research program (Magnusson et al., 1975).
Adulthood adjustment was examined in mostly the same life areas that earlier
sociometric research has focused on for adolescence and early adulthood
adjustment. Adulthood data was taken from official registers (young adulthood)
and from the large intensive investigation for women aged 43 that was carried
out in the IDA program (Bergman, 2000). From theoretical accounts and
temporal extrapolations of earlier empirical findings, a number of hypotheses
were formulated: Peer rejection was expected to be associated with higher rates
and peer popularity with lower rates of young adulthood alcohol abuse and
criminality; the examination of the association between peer status and young
adulthood mental health problems was exploratory due to the mixed findings of
earlier research; alcohol consumption was also followed up into midadulthood to
test the hypothesis that any potential impact from childhood peer relations
would be diluted or changed over this long time perspective, considering the
more stable life situation in midadulthood; peer rejection was supposed to have a
negative impact on subjective well-being and life satisfaction in a variety of
important life domains in midadulthood (job, social relations, family, general
health), as well as on educational and job success up to midadulthood (while the
41
reverse association was hypothesized for peer popularity). Four potentially
confounding behavioral and social correlates from the childhood period were
included: aggression, withdrawal, academic achievement, and SES.
Main results and conclusions. Only a few hypotheses were supported. Results
showed that, by young adulthood, rejected girls were at increased risk of
criminal offending and alcohol abuse (2 and 8 times increased risk compared to
the average status cluster, respectively). In midlife, the only significant finding
was that popular girls had obtained higher-qualified jobs than other girls. For all
other midlife adjustment areas, such as social relations, subjective well-being,
health, and alcohol consumption no significant associations were found.
Childhood confounding variables were introduced and explained two of the
three significant results. The significant association for higher-qualified jobs in
midlife was explained by childhood academic achievement, and young
adulthood criminality was explained by childhood aggression. Thus, these
significant findings more or less seem to fit into an incidental model, where peer
problems are seen as markers of other problems. However, alcohol problems in
young adulthood were significantly related to childhood peer rejection, even
after the childhood correlates were introduced. This finding points to a causal
model: The results of the study were also shown to hold if the method of
analysis is switched to a conventional, variable-oriented one, applying
regression analysis to predict adulthood adjustment from childhood peer
relations.
The unusually long time period might be seen as a strength in the sense that we
investigated far-reaching effects of childhood peer relations. However, it is also
a weakness and a limitation, as potential developmental influences might be
diluted over time and without more measurement points in between, we have no
knowledge about what happens at different time periods and possible
developmental turning points. In future investigations, it is also of importance to
include both men and women, and to compare their development and adjustment
in the light of childhood peer problems.
The absence of any appreciable impact of negative childhood peer relations on
the adjustment in midlife is hopeful. However, the relatively small peer status
groups and attrition problems lead to limited ability to detect group differences
in cases where the effect sizes are small.
Overview of the measures included in the studies
The measures that were used in the studies are presented in Table 1. The
intention is to give the reader an overview of what measures have been used in
42
Table 1.
Overview of the measures in Studies I to IV
Study
Sociometric measures
Control measures
Outcome measures
I
Grade 3 (age 10)
-Positive nominations PR
-Rank orderings PR
Grade 4 (age 11)
-Positive nominations PR
Grade 3 (age 10)
-(Intelligence T)
II
-See Study I
Grade 3 (age 10)
-(Aggression TR)
-(Withdrawal TR)
-(Intelligence T)
III
Grade 3 (age 10)
-Rank orderings PR
Grade 6 (age 13)
-Rank orderings PR
IV
-See Study III
Grade 3 (age 10)
-Family SES PaR
Grades 3 and 6 (age 10/13)
-Academic achievement R
-Aggression TR
-Withdrawal TR
-See Study III
Grade 8 (age 15)
-Intelligence T
-Marks in Swedish and physical training R
-Test in Swedish T
-School attitudes SR
-School dropout (boys) R
Grade 8 (age 15)
-Rank orderings PR
- Self-estimated peer ranking SR
-Peer network measures SR
-Measures about antisociality SR
-Peer situation SR
-
Young adulthood
-Alcohol abuse R
-Criminality R
-Mental health R
Midadulthood
-General health SR
-Alcohol consumption SR
-Subjective well-being SR
-Measures about satisfaction with social
relations SR
-Job satisfaction SR
-Vocational career SR
Note. PR = peer-report; SR = self-report; TR = teacher-report; PaR = parent-report; R = register data; T = test.
Measures in brackets are accounted for in the text of the study, but are not formally treated as control variables.
which study and how the studies are connected to each other in that respect.
Study III stands out from the other studies as it is a methodological study and
not a prospective study with outcome measures.
Comparison between the sociometric classifications used in the studies
In this section, I intend to compare and connect the cluster solution in Studies III
and IV with the classification method used in Studies I and II for identifying
stable peer status groups. The latter one-year longitudinal classification
procedure is described in detail in Adebäck (1969), in Magnusson et al. (1975),
and in Study I, and it is here referred to as the AM method. Bearing in mind the
discussion above about arriving at rather similar results with different
43
Table 2.
Girls in AM peer status groups classified into specific clusters
AM peer status groups
Clusters
Average
Popular
Rejected
1. Average
7
-
1
2. Popular
1
8
-
3. Rejected
1
-
9
4. Near average at age 10 with a positive change
4
-
-
5. A marked positive change into average at age 13
-
-
3
6. A negative change into near average at age 13
2
4
-
7. A marked negative change into near average at age 13
-
2
-
All AM girls
15
14
13
Note. AM peer status groups refer to sociometric categories according to Adebäck (1969), Magnusson et al.,
(1975), and Study I. The clusters are from the sociometric classification in Study III.
sociometric measures and classification methods, it seems probable that a
considerable overlap will be found in sociometric group belongings identified by
the two classification methods. In addition, the classification time periods are
overlapping, although differing in length (one year and three years,
respectively), and rank ordering at age 10 is used as a sociometric measure in
both classifications.
Table 2 account for how the girls in the three AM groups of one-year, stably
rejected, average, and popular peer status are distributed over the different
clusters. An exact, cell-wise probability test (Bergman & El-Khouri, 1987)
showed that the cell with members belonging both to the rejected cluster and the
rejected AM peer status group, the cell with members belonging both to the
popular cluster and the popular AM peer status group, and the cell with
members belonging both to the average cluster and the average AM peer status
group indicated, in all three cases, significant types at p< .001. A more detailed
descriptive account of the figures in Table 2 shows that a majority of the girls in
the rejected AM peer status group belongs to the rejected cluster and the rest of
them, but one, to the volatile Cluster 5, which is characterized by low peer status
at age 10 (see Study III for cluster identifications). Likewise, most of the
popular AM peer status girls are members of the popular cluster, and the rest of
them are members of the volatile Clusters 6 and 7, characterized by high status
at age 10. The same tendency was found for the average AM peer status girls.
Most of them belong to the average cluster or in second hand to Cluster 4,
44
characterized by average status at age 10, while the rest of them are spread out
in other clusters.
In conclusion, the congruence between the two methods used in this thesis to
assign girls into different peer status groups is rather high for the girls included
in the AM classification. It also seems logical that the correspondence is highest
at age 10, where both classification methods had the first of their selection
points, while the second selection point was separated in time between the two
classifications (at age 11 and age 13, respectively). Judging from our findings,
some age 10 rejected and popular girls, while still rejected and popular at age 11
(according to AM), have changed their peer status at age 13 (according to the
cluster solution). In referring to the maintenance phase of the Coie (1990)
model, it might be suggested that one year is too short a period to capture the
developmental processes taking place in that phase of peer rejection and that
three years is a more appropriate time period to separate stable peer rejection
from more transient states of peer rejection. Ideally, one should wish to have
more than two time points for the longitudinal sociometric measurements.
45
General discussion
Overview
In the general discussion, I will concentrate in more detail on the findings of the
empirical studies and point to some problems and implications. Three main parts
will be included. The first two parts pertain to the two-folded meaning of
sociometry as both a methodology and a whole research domain (as connoted by
Hallinan, 1981). First, a methodological part will focus on the sociometric
methods used in the present research and on some implications of the
introduction of the clustering classification approach. Then, in the second part,
the adolescent and adulthood adjustment consequences of childhood stable peer
status will be discussed, as mirrored by the empirical findings of this thesis. To
bring the thesis to a close, a final comment will be included pertaining to
constraints of the present research and the plea for future research.
Methodological issues
Introduction. By our sociometric classifications over time, we have tried to
mirror the maintenance phase of peer rejection, which according to Coie (1990),
is a long time period during which peer rejection becomes a stable reality for
some children. In our case, the sociometric measurement period is one year for
the adolescence outcomes studies (Studies I and II) and three years for the
adulthood outcome study (Study IV). The first one-year time period is well in
line with many other studies examining stable peer status as an adjustment
predictor, while the second three-year period is rather long in comparison to
most other research.
The rank-ordering sociometric measure. First, let me state that to secure
good cooperation from families and school staff, negative nomination measures
were omitted from the present research program and replaced by a rank-ordering
measure combined with positive nominations. From the beginning, the research
program aimed to examine the included cohorts in a long-term prospective
perspective, and, for that reason and for general ethical considerations, it was
important not to endanger the program by choosing potentially unethical and
questionable measures. As a result, we then run the risk that our peer status
groups include, in part, different members compared to groups where the
negative nomination criterion or, for that matter, some other criterion is used. As
we have seen, a variety of sociometric methods and classification procedures
have been adopted in placing children in different status groups. Some
46
discrepancies and incomplete overlap in the peer-status-group compositions are
to be expected as a result of these classification differences (e.g., Terry & Coie,
1991). However, it seems that different two-dimensional approaches give rather
similar rejected and popular status group belongings with similar predictive
utility (e.g., Asher & Dodge, 1986; Terry & Coie, 1991).
The choice of the rank-ordering measure is based on its psychometrical quality
to force the child to place every peer in the group in a unique position on the
ranking that differs from every other peer. This is a crucial difference in
comparison with the rating scale, where a child can give as many or as few
extreme ratings to his peers as he wishes (Asher & Dodge, 1986; Asher, 1985).
Further, the rank ordering differs from the positive and negative nominations
(Coie et al., 1982; Newcomb & Bukowski, 1983) in as far as the measure is not
asking for a number of liked or disliked children. To obtain clearly positive to
negative endpoints on the scale, the children in our project were instead given a
same-sex class roster and asked to rank every classmate of the same sex in the
order they wanted them to stay with the class if it was to move to a new
classroom where there is not room for everyone. This is equal to asking the
children to hypothetically exclude some classmates from the class. Of
importance is that, in contrast to the negative nominations measure, the rankordering measure used in this project has a positive selection focus (those
classmates one wants to stay). Thus, the child is encouraged to think about how
much he or she prefers different peers, rather than to a direct selection of
nonpreferred peers.
The peer status classification for the adolescence follow-up (Studies I
and II). The rank-ordering measure that was administered in grade 3 was used
as one of the classification criteria. The rank-order values (from 0 to n-1 where n
is the number of same-sex pupils in the class) given by each child were
standardized for each class and sex (transforming ordinal scores to normalized
z-scores according to Fisher & Yates, 1953; Josephson, 1967), which made
values from different classes with different numbers of same-sex pupils in them
more comparable. The score for each child is the mean of his or her standardized
values. This classification procedure implicates that children in heterogeneous
classes, where there is a clear hierarchy of peer likings and dislikings, are
represented in the extreme peer status groups of children. In contrast, children in
homogeneous classes, with no extreme status positions (i.e., more even scores
for different children), are excluded in accordance with the purpose of
identifying children with extreme peer status positions. For the other criteria,
positive nominations given in grades 3 and 4, classification cut-offs were based
on a statistical model for rare scores developed by Bronfenbrenner (1945). This
classification model is rather robust against differences in group sizes that are
found in, for instance, school classes. Thus, independently of number of
47
nominators, the same cut-offs were used for all school classes. Like for the rankordering, this classification method emphasized representation from status
heterogeneous classes by its consideration of differences in raw-score
distribution between nominating peer groups (see discussion in Newcomb &
Bukowski, 1983, whose classification model is based on Bronfenbrenner). The
boys and girls who fulfilled the criteria for having a negative peer status in all
three included sociometric measures (rank-ordering and positive nominations in
grade 3 and positive nominations in grade 4) were classified as rejected. After
that, the most popular in the same classes according to the criteria for being
popular were selected and finally the average status children were selected. The
procedure is described in detail in Adebäck (1969) and in Study I.
As was accounted for above, research has shown that rejected and popular
groups exhibit moderate sociometric stability, while neglected and controversial
status groups are highly volatile, usually changing into average status (e.g.,
Bukowski & Newcomb, 1984; Coie & Dodge, 1983; Ollendick et al., 1991).
This was the reason for adding the positive nominations at the end of grade 4 as
a single criterion, to the criteria measures at grade 3 (rank orderings and positive
nominations). It secures the aim of the classification procedure to attain stable
peer status groups of popular and rejected boys and girls for the adolescence
follow-up, because neglected and controversial children do not usually remain in
their extreme status assignments over time. It is worth to add that the stability
differences between extreme status groups open up for an alternative
classification system (to my knowledge not yet realized) using only positive
nominations from two time points, to identify, rather pure, stably rejected (no or
almost no positive nominations at both occasions) and popular (many positive
nominations at both occasions) peer status groups.
By the applied selection procedure, with its time interval of one year, much of
the regression effect was eliminated (Nesselroade, Stigler, & Baltes, 1980;
Newcomb & Bukowski, 1984), which reduced the problem of instability in the
assignment of children into the sociometric groups. Furthermore, the stability of
the groups was deliberately strengthened by the choice of time interval. In
advancing from grade 3 to grade 4, the pupils move from the beginning level
(grades 1 to 3) to the middle level (grades 4 to 6) of the Swedish school system,
which implies some rather significant changes (a new teacher, a new classroom
– sometimes in another school building). As those potentially influencing factors
were included in the time span, their effect on the sociometric groups was
eliminated, which ought to contribute to the peer status stability of the identified
groups and increase the validity of the findings.
Another methodological implication of our selection procedure for the
adolescence outcome studies, with its school class matching component, is that
the members of the popular groups are not necessarily the most popular children
48
in the population. The group selection is based on finding the most extremely
rejected boys and girls according to the criteria, and then to select the most
popular boys and girls in the same classes. Thus, there might be children in other
school classes who are more popular, as judged by the sociometric
measurements. Our classification procedure aimed at controlling for school class
differences in, for example, teacher influences, peer constellations and
interactions, and the school context, which were judged as more important than
recruiting the absolutely most popular boys and girls to our status groups.
The peer status classification for the adulthood follow-up (Studies III and
IV). A more modern classification procedure for establishing stability in peer
status was introduced for the follow-up into adulthood. One of the issues that the
method discussion in the literature has focused on are the problems of
artificiality of sociometric groups, where classification is mainly achieved by
statistical considerations rather than by conceptual reasons and underlying
psychological meaning (Asher & Dodge, 1986; Cillessen & Bukowski, 2000a;
Rubin et al., 1998; Terry & Coie; 1991; for theoretical and methodological
discussions, see, for example, the work edited by Cillessen & Bukowski, 2000b;
for discussion of different cut-off scores, see Hymel, Bowker, & Woody, 1993;
and Zakriski & Coie, 1996). By using more person-oriented methods, such as
cluster analysis, one may arrive at more “natural” groupings, with membership
in a cluster dependent on the individual’s value profile in the space of included
variables. Thus, we saw it as a methodological challenge to apply a clusteranalytic technique in sociometric research in the hope of identifying valid, stable
peer status groups for use in longitudinal follow-ups.
Another reason for changing the methodological approach is that for extremely
long-term follow-ups, attrition might be a substantial problem. While interesting
because they represent extreme social standing among peers, the small peer
status groups in the adolescence follow-up studies are very vulnerable. These
groups also represent a smaller percentage of the population than is common in
most other sociometric follow-up studies. Thus, it would be preferable to have a
larger number of participants in the peer status groups for the follow-up into
adulthood. There was a need to reconsider the use of the sociometric
classification methods and furthermore to decide which criteria might be best
applied to a new research design. It seems that the sociometric rank-ordering
measure – which was given in grade 3 and again in grade 6 – has a key position.
In an effort to identify different peer status groups that are both temporally
stable over a long time period and sufficiently large for long-term follow-up
purposes, a modern clustering technique was applied to the scores derived from
these age 10 and age 13 rank orderings. The study by Brendgen and colleagues
(2001) accounted for above was an inspiration, because large clusters of
popular, average, and unpopular children were identified in that study.
49
As a first step, it seems important to reflect on how the sociometric data one
includes should be properly represented or classified. As we have seen, the
standard score method has been criticized for not considering the raw score
distribution of nominations before standardizing. This will cause almost equal
distributions and rather fixed-sized proportions of status group members from
different groups (in our case, same-sex class peers), even if the groups differ
widely in their raw score distributions. To circumvent this problem, the authors
of some studies have based their definitions of sociometric status on the number
of positive or negative nominations received by a child divided by the number of
nominators (e.g., Green et al., 1981; Kistner, Balthazor, Risi, & Burton, 1999;
Pakaslahti & Keltikangas-Järvinen, 2001; Salmivalli, Kaukiainen, & Lagerspetz,
2000; Wu et al., 2001). The same solution for representing “like most” and “like
least” sociometric data was applied in the cluster-analytic studies of the present
thesis (Studies III and IV).
The findings from the validation procedure of Study III supported the empirical
and theoretical meaningfulness of cluster analysis as a classification method in
sociometric research. As was hypothesized, the cluster solution included three
stable clusters identified as average, popular, and rejected. These longitudinally
stable clusters affiliated in important respects with the corresponding peer status
groups that were identified by the CDC method (with cut-off limits introduced
by Coie and Dodge, 1983, which are the most widely used in sociometric
studies). However, the rejected and popular clusters were substantially larger
than the corresponding CDC peer status groups, with at least twice as many
members. Thus, it seems that cluster analysis helps to overcome some of the
problems related to traditional sociometric methods in the identification of
sufficiently large homogeneous popular and rejected groups with stable
trajectories. Note that the comparison between the cluster classification and the
sociometric classification (AM), which was accounted for in a previous section
and Table 2, gave similar results as the Study III validation procedure.
The findings for the basic sociometric measures derived from the age 10 and age
13 rank-orderings indicated that the additional girls in the rejected cluster were
less rejected than those that also belonged to the stably rejected group according
to CDC. However, as was also found, these two groups of rejected girls did not
differ significantly in the childhood adjustment correlates. The findings for
popular girls were mainly the same (Study III, p. 104). It is worth noticing that
age 10 aggression was not significantly related to rejected status. Besides the
explanation given in Study III (p. 108) pertaining to multiple significance
testing, another hypothesis is that the aggression measure might not adequately
indicate girls’ aggression. On the other hand, the same measure was applied at
age 13 resulting in a significant association with rejection, and, furthermore, it
was constructed as a rather “global” and general indicator of aggression to fit
both girls and boys (see Magnusson et al., 1975).
50
An interesting in-group status development was detected for the rejected and
popular clusters, with a significant increase in degree of rejection and
popularity, respectively, from age 10 to age 13. Thus, the cluster analysis
produced a sociometric classification with stable extreme clusters that mainly
seem to include girls that, while stable in their group membership, became more
negatively/positively judged by their peers over time. Perhaps temporal stability
in an extreme peer status position might indicate an interaction with peers that
not only conserves this position, but also leads to increased rejection or
popularity over time. As we have seen, Coie (1990), in his phasic developmental
theory of peer rejection, included a third phase, the maintenance phase, ”during
which time rejection by the group becomes a stable and enduring reality for
some rejected children” (p. 365). Among the developmental processes involved
in that phase may be some that not only contribute to maintenance of peer status
(e.g., Cillessen et al., 2000; Parker et al., 1995), but also to increased rejection.
One may, for example, imagine increasingly negative perceptions, stereotypes,
and behaviors as included among those processes and similar, but benign,
processes for the popular child. Coie described several routes into a negative
cycle of peer difficulties in his emergent phase of peer rejection (e.g., social
characteristics, self-perceived peer standing, expectations about social success or
failure, and the role of the peer group) that might be equally applicable in a
process of aggravated peer rejection in the maintenance phase. While Sandstrom
and Coie (1999) have recently studied factors associated with improved peer
standing among elementary school children, including modest improvements
within an unchanged rejected peer status, it seems that the negative process
found here remains to be examined more closely.
As expected, the temporarily volatile clusters that were identified in the cluster
analysis could be interpreted as including average or “near average” status either
at age 10 or at age 13 (Study III, p. 100). According to previous research
findings, average peer status seems to be central in children’s status translations
and the developmental paths of children with transient peer status mostly start
from or lead to the average status position (Coie & Dodge, 1983; Newcomb &
Bukowski, 1984; Parker et al., 1995; Terry & Coie, 1991). However, transient
peer status trajectories, which may include both linear and nonlinear processes
over time, are difficult to analyze in detail without a more fine-graded timescale.
While such an analysis was not a goal of the studies included in the present
thesis, one may, nevertheless, speculate that these trajectories might not be
captured by specific sociometric groups or clusters but are fluid characteristics
inherent in the peer group dynamics and interactions affecting the peer status of
individual children at different points in time (see also the extensive discussion
in Wu et al., 2001, about stable and fluid characteristics of children’s peer status
and the reliability of sociometric instruments). Thus, the volatile clusters may
come out as accidental gatherings of girls that might have dissolved or changed
51
into other groupings, at least pertaining to included members, if other or more
time points had been included in the cluster analysis. Furthermore, stepping up
or down one level in the hierarchy of cluster solutions (to the eight-cluster
solution and the six-cluster solution, respectively) affects only volatile clusters,
which adds to the evidence of their more accidental nature.
The cluster analysis did not identify any stable clusters of neglected or
controversial peer status, which is in line with the expectations, and earlier
empirical findings and theoretical discussions (e.g., Frederickson & Furnham,
1998). While the CDC classification method did not identify a longitudinally
stable controversial group either, it identified a stable neglected group (Study
III, p. 103). In the cluster analysis, most of these CDC neglected girls were
assigned to the stably average cluster. Newcomb and colleagues (1993), in their
meta-analysis, found that neglected children displayed the fewest differences
from average children of all extreme sociometric groups. They speculated that
neglected children may simply select a lower level of involvement in their peer
group, and, thus, are more invisible and not so well known by their peers.
Perhaps they represent a special group of average children with a lower profile,
and accordingly they may have a natural place in “average status” clusters.
From a purely methodological perspective, one might state that there is no prior
way of knowing what clusters you will obtain in a cluster analysis – it depends
entirely on the associations between the included variables (the individual value
profiles). However, one may form hypotheses (as in the cluster analysis in the
present thesis) built on theory and empirical findings from earlier research in
which the same or other classification methods have been used. Then, one has to
apply the criteria outlined for the clustering method and try to interpret the
centroid values of the clusters in the solution (somewhat analogous to
interpreting factors in a factor analysis) and validate the interpretation. In most
cases this is possible and meaningful only for some of the clusters. For the other,
more “mixed” clusters, one has to search for information from the specific
research area in question, which may give some hints about how these clusters
might tentatively be interpreted.
Stimulus aspects of sociometric criteria. It seems appropriate to draw some
attention to the phrasing of the questions in the sociometric measures in the light
of the fact that a number of different criteria have been used in sociometric
research (for up-to-date reviews, see Hymel et al., 2002; Terry, 2000). The
criteria applied here in both the positive nominations and the rank-ordering
measures were in adherence with a tradition in sociometric measurement that
goes back to Moreno (1947) and which advocates the use of concrete, taskspecific, situation-specific or indirect preference questions – such as play with,
sit next to, invite to a birthday party or go on a trip together – rather than
52
abstract, situation-general, direct preference or more multidimensional questions
– such as “liking” or “friendship” questions. In a recent meta-analytic review,
Jiang and Cillessen (2005) noticed that peer nomination methods in different
studies varied substantially with regards to specificity in the aforementioned
sense. These authors examined the short- and long-term stability of different
dimensions of sociometric status and found that reliability and validity were not
influenced by sociometric technique, nor in the sense of general or situationspecific questions, nor pertaining to peer ratings or nominations. Thus, it seems
that the specific choice of sociometric questions may be of limited significance
for the identification of stable peer status groups in Studies I and II or of stable
peer status clusters in Study III. Furthermore, our finding of a high correlation
between age 10 positive nominations and the “like most” scores of the age 10
rank-ordering is also in line with the Jiang and Cillessen findings.
Gender related issues. As in several other studies, same-sex peer status was
examined here. As was discussed above, same-sex peers are the nearest, most
familiar, and most well-known peers in childhood and early adolescence for
most children and children associate to a very high degree only or mostly with
those peers (e.g., Fabes et al., 2004; Maccoby, 2000). Thus, it seems justified to
primarily measure children’s standing among same-sex peers because they are
much more familiar with them than with cross-sex peers. On the other hand, our
adolescence findings indicate that status among same-sex and cross-sex peers is
rather similar (Study II), which might mean that the choice of nomination group
pertaining to gender does not influence peer status membership very much.
One essential limitation of the methodological study (III) is that it only included
girls. However, in a meta-analytic review, Hyde (2005) found strong evidence
for the gender similarities hypothesis. Furthermore, judging from the
overwhelming majority of findings pertaining to stable peer status groups of
both genders (see the introduction), it seems probable that a cluster analysis may
identify the same typical stable groupings for boys as for girls, and may also
lead to similar results in the validating process. Most importantly, Brendgen and
colleagues (2001), in their cluster-analytic study, presented findings that were
similar for boys and girls, and Haselager et al. (2002) showed that cluster
analysis was an effective tool in classifying rejected boys into different
subgroups. As another, admittedly not very strong defense of the “only girls”
position, one might point out that there has been a relative shortage of
sociometric and other psychological studies about girls in the past, so in that
sense, if one has to make a choice, concentrating on girls is the right decision. In
addition, the cluster-analytic studies (III and IV) are included in a project about
females’ longitudinal adjustment up to adulthood and it was in this context that
they were planned and implemented. Finally, in the methodological study (III),
the concentration is on the clustering method in itself and whether it is of
53
interest in sociometric research. One needs a relevant, large sample of children
and if the method works with that sample one can go on to validate it in other
samples and cultures.
On the other hand, in their meta-analysis, Jiang and Cillessen (2005) found some
evidence that a larger proportion of boys in the samples might be associated
with lower sociometric stability. Thus, future studies addressing the issue of
whether cluster analysis is a reliable and valid classification technique for
identifying stable peer status groups of both genders, would be valuable. In a
validation of cluster analysis as a method in sociometric research, it is also
important to examine different age periods from early childhood onwards,
perhaps including more than two time points for the sociometric measurements,
and to use a more modern data set than was used in the present study.
Peer status and future adjustment
The adolescence findings. The findings pertaining to midadolescence
adjustment are mainly in adherence with earlier findings in the sociometric
research area. First, in connecting to the methodological discussion above, one
may notice that the sociometric measurements show that rejected children of
both sexes seem to maintain their lower popularity with classmates of the same
sex also in middle adolescence and that the popular children were still the most
popular. Of course, this finding does not exclude the possibility that some
children may have changed their social standing among peers; the results pertain
to groups of children, and some individuals show different patterns of
development (as was indicated by the comparison between AM groups and
clusters above, see Table 2). However, the most important conclusion seems to
be that the different peer status groups follow different developmental paths and
as groups preserve their status position among peers over several years, and
probably over the entire obligatory school period. Compared to most other
studies, our time period from grade 3 to grade 8 is unusually long and could in
that respect be compared to the study of Coie and Dodge (1983), where rather
high stability was found for the rejected group over a 5-year period. One might
speculate about whether this long-standing status stability might also be
extended backwards to the preschool years. A study by Waldrop and Halverson
(1975) lends support to such a supposition. These authors found a stability in
peer relations from the early preschool years to the early school years, which, if
applied to our peer status groups, suggests that many participants might have
had the same status position in the preschool years as they had in their school
years. In the nomenclature of Coie (1990), the precursor phase followed by the
emergent status phase of peer rejection may start early in life (during the
preschool years) and then the maintenance phase and the consequence phase
54
develop and affect the child’s development during the school years and into
adolescence.
Similar patterns were found both for same-gender and cross-gender peer status
in midadolescence (Study II). This similarity was especially pronounced for
rejected children and the findings also indicated that the children were aware of
their low age 15 peer status. Thus, it seems that the rejected children’s low
popularity is largely gender-independent, which could be a still more
aggravating fact for their adjustment than being of low popularity only among
same-gender peers. Particularly during adolescence, the peer group comes
forward as an important alternative in the liberation from earlier dependencies
and in the search for an identity of one’s own. Furthermore, the awakening of
sexuality and the sexual maturity process in adolescence increase the importance
of opposite-gender peers (e.g., Maccoby, 1990). Negative reactions from peers
of both genders during adolescence might therefore be obstacles in the process
of forming one’s own identity and for one’s self-esteem.
Overall, the findings from both self and peer reports about peer relations during
adolescence indicate that rejected children had a more negative situation among
conventional peers (that is, among schoolmates, same-age peers, and classmates)
in adolescence than other children. In their adolescence years, rejected children
may not be as successful as other adolescents in finding a place among those
peers that are close to them in terms of social characteristics and age. However,
they do not seem to compensate by associating to a higher degree with
nonconventional peers (in terms of age, occupation, or antisocial characteristics)
or to be more antisocial themselves than other adolescents. These findings
pertain to midadolescence and there may be a change in late adolescence, when
antisocial activity seems to peak (e.g., Agnew, 2003). Furthermore, the early
aggressive tendencies of the rejected children (documented in Study II, p. 753,
for age 10) may find other expressions in midadolescence than the normbreaking behaviors that were in focus in the present project as examples of
antisocial behavior (e.g., French & Conrad, 2001). Also, the information about
own and peers’ norm-breaking behaviors in this project came from self-reports.
Perhaps data from other sources – for example, register data – might have given
another picture (see Laird et al., 2005, for a short discussion of this issue).
However, the self-report measure that was employed was partly validated by the
expected finding that boys were more antisocial and also judged their peers as
more antisocial than girls (e.g., Barriga et al., 2001).
Differences in academic performance and intellectual capacity were in the
predicted direction and adhered to earlier research documenting a positive
association between peer status and school achievement. As was discussed in
Study I (p. 216), the causal direction is unclear because no control was made for
academic achievement or intelligence in the early school years. However, in a
55
separate examination, a rather high similarity was found in intelligence score
differences between the peer status groups at grade 3 and grade 8. This finding
gives some support to an incidental or intermediate explanation of peer status
influences on academic ability. Furthermore, other authors have found that
controlling for cognitive capacities decreases the association between childhood
peer status and later academic achievement (e.g., Bagwell et al., 1998; Vitaro et
al., 1994).
Peer status did not predict general attitudes towards school either for boys or for
girls; only slight tendencies were found for girls that were congruent with the
school achievement findings. As this is a relatively unexplored research field,
more investigations – perhaps with other self-report methods, such as interviews
– are needed before more definitive conclusions can be reached about the
relation between peer status and subjective school experiences.
One third of the rejected boys dropped out of school without finishing the last
obligatory school year. School dropout is an important indicator of
maladjustment and gives further evidence of the rejected boys’ exposed position
in school. Other boys (or, for that case, girls independent of peer status group
belonging) had a much smaller or nonexistent dropout rate. Furthermore, the
rejected school dropouts seemed to be more gravely rejected initially than other
boys in the same peer status group and they were absent from school more often
before they dropped out than other boys and than girls. Thus, their definitive
dropping out of school might be a rather late step in a social outcasting process,
taking place during the maintenance phase and consequence phase of the
development of peer rejection (Coie, 1990). Earlier steps in that process include
long-term rejection by peers, high levels of school absence, and low school
achievement. One may speculate whether these indications of a possible social
outcasting process are an effect more of the dropout boys’ initial extreme
rejected status (compared to other rejected boys) than of consistent rejection, in
which case it comes close to an incidental interpretation of the impact of stable
rejection on school adjustment for this subgroup of rejected boys (e.g., Cillessen
et al., 2000). It seems important in future research to control for initial peer
position, even inside a specific peer status group, to separate the effects of stable
peer status from the effects of initial peer status.
An alternative explanation may be that peer rejection does not directly
contribute to school dropout, but is a marker of personal problems, such as the
behavioral problems pertaining to aggression and withdrawal, and the school
problems that were documented in Studies I and II. This explanation is in line
with the findings of Vitaro and colleagues (2001), that early disruptiveness and
poor academic performance (as well as socio-familial adversity) directly
predicted school dropout for boys, while unpopularity and friendlessness did
not. Likewise, French and Conrad (2001) found that only rejected adolescents
56
who were antisocial had a higher rate of early school dropout. Finally, Farmer,
Estell, Leung, and colleagues (2003) identified three paths leading to school
dropout for aggressive youth. Some aggressive youth seem to be isolated and
socially unconnected in school and their dropping out may be associated with a
lack of such social affiliations that could have promoted engagement in school.
Farmer, Estell, Leung et al. did not include sociometric status measures in their
study, but the dropout process of their aggressive and socially isolated youth
may well coincide with the same process for the rejected dropouts in the present
thesis.
In conclusion, peer status is related to future school adjustment factors, such as
academic achievement and, for boys, school dropout. How important peer status
is for school adjustment is, of course, a difficult question to answer, but one
might expect that average or high popularity could function as a safety net that
gives strength and support when one fails in other life areas. Rejected children
of both genders seem to have problems in at least some important school
adjustment areas, which may reinforce and uphold their rejected status in an
interactional causal chain. As we have discussed, peer rejection combined with
other negative school factors might jointly function to drive the individual into
an outsider position and as an ostracizing mechanism. Thus, some of our
research findings indicate that peer situation together with other adjustment
factors seem to form an intriguing social pattern of developmental importance
for the future life situation. This interpretation of the effect of peer rejection on
adolescent adjustment comes close to an intermediate model, that is, a model
between the causal and the incidental models (for a description of these models,
see above).
The adulthood findings. In Study IV, we have prospectively investigated how
late childhood and early adolescence stable peer status relates to adjustment in
different life domains for women in young and middle adulthood. Previous
longitudinal studies have been limited to follow-ups into late adolescence and, in
a few cases, young adulthood. Thus, our study has a time span that considerably
extends what has been common in this research field.
The young adulthood follow-up revealed that women belonging to the rejected
cluster had been exposed to legal and treatment measures from the authorities
due to their drinking behavior much more frequently than women belonging to
the average cluster (almost eight to one). This strong significant association
remained even after included childhood correlates had been controlled for. Thus,
one might assume that peer rejection in childhood plays a role in later antisocial
drinking over and beyond important childhood confounders such as aggression.
As we hypothesized, rejected females were also significantly overrepresented in
the criminal records across young adulthood into the beginning of middle
57
adulthood. However, this effect disappeared, when childhood adjustment factors
– and aggression in particular – were included in the analysis. Thus, aggression
might be a powerful underlying factor that affects antisocial behavior, while
rejection seems to be more of a marker. This is well in line with earlier research,
accounted for above, about the role of peer rejection and aggression in
explaining later antisocial behavior and criminality. Finally, no significant peer
status differences were found for the young adulthood mental health indicator,
which is in accordance with some previous studies (e.g., Woodward &
Fergusson, 1999).
What is most striking when reviewing the findings pertaining to adjustment in
midlife is the absence of evidence for any sweeping or appreciable impact of
childhood peer relations. Women who were rejected or popular in childhood did
not differ much in their adjustment 30 years later from women who had average
peer relations as children. However, earlier research has only examined
outcomes up until young adulthood and, thus, more research for the middle
adulthood period is needed to corroborate our findings.
The only significant difference found was that popular females had achieved a
higher career-level compared to average status females (while, in contrast to
previous research findings up to early adulthood, rejected females did not differ
significantly from average females). As we hypothesized, the significant
difference between popular and average females was explained by the childhood
correlates, especially academic achievement. Thus, a possible causal model
seems to be excluded in favor of an incidental model, where positive peer
relations are markers of other more basic strengths; in this case, academic
achievement and probably intellectual capacity.
Middle-aged women’s satisfaction and well-being experiences in different life
sectors do not seem to be associated with childhood peer status. The stable
clusters did not differ significantly in subjective well-being, satisfaction with
adulthood social relations, self-reported health (including alcohol consumption)
or job satisfaction. This might indicate that childhood or later factors, either
environmental and/or of a more heritable kind, are of greater importance for
women’s general life satisfaction or else may dilute the effects of childhood peer
problems. The long time span between childhood and midadulthood might, for
example, give women with a lack of positive peer experiences in childhood and
early adolescence plenty of time to repair and acquire more satisfying
experiences of social relationships.
Finally, in the only adjustment area where we have measures from both young
and middle adulthood, namely drinking behavior, it was possible to examine
developmental adjustment over two adulthood life phases. In contrast to the
significant young adulthood association between childhood peer status and
58
alcohol abuse, the midadulthood self-declared alcohol consumption of women
that were rejected in childhood did not exceed the consumption of other women
(which was rather modest, about 30 grams/week, regardless of their cluster
assignment). A possible explanation for the divergent findings over time for
rejected women might be that different indicators were used. Furthermore,
young adulthood represents a rather turbulent life period (Arnett, 2000;
Levinson, 1986), and perhaps especially so for those who have had previous
adjustment problems, such as being peer rejected. One may see that period as
partly a continuation of the adolescence period with its negative experiences for
rejected girls. In middle adulthood, most of the rejected females reported that
they had a comparatively “good” life, indicating a rather traditional and
“normal” life. So the problematic behaviors of some of them, such as heavy
drinking, might have diminished and been replaced by more acceptable
behaviors.
The context and scope of the reported research: Limitations and
constraints
A number of limiting conditions pertaining to the research included in the
present thesis have, in part, been accounted for in the studies and at different
places in the preceding text. However, it may be of interest to discuss some of
these specific features conjointly and point to some important constraints that
need consideration in future research.
The small group sizes are a limitation, especially for the adolescence studies,
which implies reduced power of the statistical analyses. However, this limitation
may be almost unavoidable when, as in this case, examining children with
extreme and relatively homogeneous peer difficulties selected from a population
that is not very large. Although larger than the peer status groups in the
adolescence studies (I and II), the clusters in the adulthood study (IV) are of
moderate sizes (especially at the midadulthood follow-up due to attrition), which
could well have given non-significant results for small effect sizes in our
analyses. Furthermore, the dropout at age 43 was higher for the rejected cluster
than for the average cluster. However, an attrition analysis of the rejected cluster
(mainly adhering to the steps that, according to Barry, 2005, are important to
take in an examination of attrition in longitudinal research) revealed that this
might not have affected our group differences in adulthood outcomes. We share
the small group sizes with most other long-term prospective sociometric studies
and this situation seems to be ubiquitous in this kind of research.
At the same time, some of the adolescence adjustment differences between peer
status groups are relatively strong. Furthermore, where there are compatible
59
results from other studies (with samples from different time periods and cultural
settings), they are mainly in harmony with the results presented here, which lend
support to the validity of the findings. In contrast, the findings of largely no
substantial relationships between childhood peer relations and midlife
adjustment are at some variance to those found in our studies for
midadolescence and young adulthood adjustment, and by others for late
adolescence and young adulthood adjustment. There are some circumstances
pertaining to the adulthood follow-up that deserve to be mentioned in the present
context.
Our participants seem to be a reasonably representative sample of Swedish
females from age 10 to age 43, born in 1955, but, as always, the specific features
of the sample are of importance in discussing potential generalizations of the
findings. Important questions are whether and how the specific time period
studied has affected the results we obtained and whether these results apply to
young girls of today, to males or to other cultures. These issues appeal for future
studies of different samples from different cultures (e.g., Masten, 2005).
Another methodological limitation pertaining to the adulthood follow-up is that
the life situation in midadulthood was mainly considered from the perspective of
the participants. On the other hand, Jung (1984), for example, emphasized the
importance of subjective measures of social support, since one might be more
affected by the social support one thinks is present than by real, but
unrecognized, social support. However, our significant findings in adulthood
referred to register data in young adulthood and to a self-report of a rather
“objective” criterion at age 43 about job qualification level, while none of our
other self-report indicators at age 43 gave significant results. This serves as a
warning against using only one kind of informant or method. A multi-method,
multi-informant design, with the same indicators at different time periods, is
recommended for future studies.
Notwithstanding the limitations pertaining to the adulthood study, it is also a
fact that almost no other sociometric study has covered such a long period of
time as we did. This fact makes it difficult to validate and generalize our
findings. However, it seems likely that the relationships over time are strongly
attenuated over a period of 30 years. Attenuation over long periods of time is
found in almost any area of study (Mednick, Harway, & Finello, 1984; cf. the
longitudinal law formulated by Clarke & Clarke, 1984), so in that sense there is
no contradiction to the large body of studies that usually report findings for
follow-up periods of 3-15 years. To use an analogy: We can predict the weather
fairly accurately three or even five days ahead but as soon as the time period
becomes longer than a week, the prediction becomes poor. Chaos in the form of
infinite sensitivity to initial conditions (Barton, 1994) and intervening events not
included in the initial equation may be responsible for this. To a certain extent,
60
this line of reasoning is also valid when predicting midlife adjustment from
childhood peer relations. Differences in peer status in late childhood were
associated with significant differences in school achievement and peer relations
in the predicted direction five years later, while there were almost no significant
findings pertaining to midadulthood adjustment.
In a methodological article about longitudinal research and different research
strategies, Collins (2006) discussed some disadvantages when applying a
longitudinal panel design, as was the case here. Panel designs are characterized
by relatively long intervals between occasions of measurement. There may be
considerable curvilinearity or transitions that are obscured or not observed by
this paucity of measurement waves. For example, Helson and Soto (2005) found
marked nonmonotonic developmental changes in middle age for women, which
might make interpretations of findings difficult if only one or two measurement
points are included. Thus, to get a better understanding of the developmental
trajectories and important predictors of adulthood development and how they
interact at different time periods and developmental turning points, there is a
need for more frequent observations from childhood up to middle adulthood
(although there are, of course, practical and economical obstacles to such an
approach).
The diversity of peer relations research
Different types of peer relations. In this thesis, I have concentrated on one
aspect of children’s peer relations – sociometric peer status – as a predictor of
future adjustment. In recent years, some studies have also included friendship
status as an important predictor. It seems that an overwhelming part of
childhood friendship relations are beneficial and have positive effects for future
development (e.g., Newcomb & Bagwell, 1996). As a contrasting example,
allying with antisocial friends might lead to antisocial development (Berndt,
1996), which we have put forward as one possible explanation for the antisocial
drinking behavior in young adulthood among our rejected females (Study IV). In
any case, it might be important in future studies to include friendship status,
friendship quality, and/or characteristics of friends together with peer status as
important predictors for development and later adjustment (e.g., Berndt, 1999,
Hartup, 1996). Likewise, the opposite of mutual friendship status, namely
mutual antipathy between children, seems to be of interest to include together
with peer rejection in studies examining the maladaptive outcomes of inadequate
peer relations (e.g., Witkow et al., 2005).
Another of the perspectives that have received increased interest among
psychological researchers in recent years is the study of peer networks.
61
However, already back in the 1930s, Moreno (1934) described sociometric
methods as a mean for understanding group structure and group processes.
Using sociometric research tools to only measure an individual’s peer status or
friendship status leaves unexplored other features of the information about social
relations that might be collected from sociometric measurement (e.g., Cairns et
al., 1998; Johnson, Ironsmith, & Poteat, 1994). Important pieces of information
pertain to the group belongings of the individual and “the nature of the social
network within which the person is embedded, including the number, size, and
composition of the component social groups and their relationship to one
another” (Cairns et al., 1998, p. 30). For a comprehensive and insightful analysis
of the social network perspective – and a comparison and integration with the
individual, sociometric status perspective used in this thesis – see Cairns et al.
In a recently published essay review, Laursen and Mooney (2005) argued that
friendship experiences are confounded with other social experiences and that it
seems difficult to attribute unique variance to friendship apart from variance
attributed to family and peer group relations. They see peer rejection as a better
predictor of most adjustment outcomes and, furthermore, friendship quality as a
stronger predictor than participation in a friendship. Salmivalli and Isaacs (2005)
examined prospective relations among peer rejection, friendlessness,
victimization, and children’s self and peer perceptions. Their findings indicate
that peer rejection has a central importance for negative self and peer
perceptions, is highly stable over time, and predicts increases in victimization
and friendlessness. The authors concluded that their findings confirm prior
evidence that peer rejection increases the risk of multiple forms of peer
adversities, such as victimization and friendlessness. Gest et al. (2001)
investigated the association of mutual friendships, social network centrality, and
sociometric status with concurrent behavioral correlates. They found that all
three dimensions of classroom social position were uniquely associated with
some prosocial or antisocial behavioral styles, but that only being
sociometrically disliked was uniquely associated with the full range of social
behaviors and, furthermore, to some degree correlated with the other two
dimensions of classroom social standing. Finally, Laird et al. (2001) found that
peer rejection was a unique predictor of subsequent externalizing behavior
problems in childhood and adolescence, while the influence of antisocial friends
disappeared when controlling for concurrent externalizing problem behaviors.
In our own research, we have in some preliminary analyses (not included in the
studies of the present thesis) compared females having a reciprocal best
friendship at both age 10 and age 13 with females that were friendless at both
ages. We arrived at about the same findings as for peer status clusters pertaining
to significant and nonsignificant adjustment differences in young and middle
adulthood. The significant findings were, however, somewhat weaker and, as for
peer status groups, they mostly disappeared when childhood correlates were
62
inserted in the analyses. In line with what has been found by other researchers
(e.g., Rubin et. al., 1998) the “friended” group was significantly associated with
the popular cluster and a similar significant association was found between the
friendless group and the rejected cluster (that is, they had many common
members). The weaker findings for friendship groups and the high correlation
between them and corresponding peer status clusters give some support to the
suggestions in the preceding paragraph about the comparatively low additional
value of including friendship participation in the sociometric at-risk studies.
Thus, it seems that the contribution of including other peer problems might not
add substantially to peer rejection in predicting later adjustment outcomes. Peer
rejection seems to co-vary with other peer problems of importance for
maladjustment, including, “negative quality” friendships, friendlessness,
victimization, and/or antisocial network belonging, and, in addition, seems to be
a good general “peer problems” indicator for maladjusted outcomes.
The above discussion shall not be interpreted as a justification for choosing
sociometric status as the only or even main focus for studying children’s peer
relations, but only as a suggestion that in research on maladapted outcomes, peer
rejection may be chosen as an important indicator or predictor signaling rather
general peer problems. Focusing on different aspects of a child’s position in the
peer group besides social status – such as friendship, mutual antipathy, self- and
peer-perceived popularity, and peer network – seems to be of the utmost
importance in endeavoring to understand the development and importance of
peer relations, both at the individual level and at the group or structural level.
Different peer contexts. Another important issue is how representative peer
relations in the school class are as a mirror of the general peer situation of the
child, characterized by different but partly overlapping peer groups
(Kindermann, 1998) and peer contexts (Kiesner, Kerr, & Stattin, 2004). In the
method sections above, I have cited Berndt and Ladd (1989), who attached the
utmost importance to the school context for popularity research, because school
groups large numbers of peers together. But this issue needs further elaboration.
Although classrooms and the school and preschool contexts are primary
socialization milieus in childhood, they nevertheless represent a narrow view of
the social contexts of children. Children associate with peers and friends during
their leisure time and in different organized activities, which may potentially
compensate for negative peer relations in school (e.g., Hymel et al., 2002). In
contrast, already early in the sociometric research tradition there were both
explicit and implicit assumptions in theoretical and empirical writings that a
child’s social standing may mirror some kind of personal characteristic and that
peer status, therefore, is relatively independent of the specific peer context (e.g.,
see discussion in Kindermann, 1998; Newcomb et al., 1993; Rubin et al., 1998).
63
A recent interesting example is a sociometric measurement model introduced by
Terry (2000) where he proposes a latent trait of “likeability” but also stresses the
importance of simultaneously considering the perceived and the perceiver (who
actually made the choices) in a perceptual framework and in their association
with the latent trait. Terry gives an empirical example, but acknowledges that his
latent trait model of interpersonal perception (LaTRIPP) is in need of more
empirical testing to evaluate “the extent to which it provides a fruitful paradigm
for further research” (p. 51).
Research findings indicate that classroom peer acceptance and out-of-school
acceptance are moderately related, at least when using peer reports in both
contexts. This relation seems to decrease in adolescence when at least mild
forms of antisocial behavior – which continue to be negatively valued in the
school class and related to rejection – become more accepted among some outof-school peers, who tend to nominate each other (see Kiesner & Pastore, 2005).
As Rubin and colleagues (1998) have emphasized, the association between a
specific form of behavior and popularity depends on whether that behavior is
normative for a group. Our own findings pertaining to the adolescence peer
situation differ somewhat from the findings of Kiesner and Pastore, as our
findings indicate at least a partial association between peer status in the class
and number of in-school and out-of-school peer contacts, and, further, that
rejected adolescents do not associate to a higher degree with antisocial peers
than other adolescents do (Study II). These somewhat divergent research
findings strengthen the need of expanding future sociometric studies to a wider
range of peer groups, including those outside the school context.
Final comments
In this final part, let us go back to the title of the thesis for some reflections.
How important is childhood peer status for later adjustment? It seems that
troublesome peer relations over a long time period in childhood have a
detrimental influence on adjustment in the adolescence years in different life
areas, as multiple research findings have documented. The contribution of this
thesis pertains to school adjustment and relations to peers in adolescence. An
exception is that no increased risk for antisocial behavior or association with
deviant peers was found for rejected boys and girls. As was concluded above, it
does not seem that rejected children’s aggressive tendencies in childhood
develop into antisocial behavior or deviant peer associations in midadolescence.
This is a beneficial finding, and one which may indicate rather normative
behavior patterns, in spite of other difficulties, and may also partly explain the
later development into an acceptable level of adjustment in midadulthood that
was found for rejected females. On the other hand, the rejected children may
64
display aggressive behavior in other ways than being explicitly antisocial, and it
is also possible that they develop antisocial behaviors and deviant peer contacts
later in adolescence and early adulthood, as the register findings indicate.
However, the paucity of sociometric studies with follow-ups into adulthood
makes any conclusions about this topic rather uncertain.
In connection with these final reflections and as a continuation of them, let us
listen to some conclusions and viewpoints of one of the most experienced
researchers and theorists in the sociometric and peer relations research domain –
Willard W. Hartup. Recently Hartup (2005; see also Hartup, 1999) summarized
what is known about peer interaction among children as a source of behavior
change and the dynamics of peer influence. He considered six issues as
especially important to reflect upon: (a) characteristics of the influence source;
(b) characteristics of the child being influenced; (c) the relationship between
influencer and influenced; (d) developmental considerations; (e) the process
problem; and (f) domain constraints. Hartup argued that peer influences have
complex effects on children’s behavior and that numerous moderators and
mediators are essential forces in these developmental processes. Longitudinal
studies are necessary to unveil developmental continuities and changes.
Furthermore, he would also like to see more field experiments and interventions
incorporated into multivariate longitudinal studies in order to enhance our
knowledge of the social processes going on between children or adolescents.
Dishion and Dodge (2005) have proposed an ecological framework in the
tradition of Bronfenbrenner as a useful heuristic to understand conditions and
mechanisms (moderators/mediators) at individual, group, and context levels
under which beneficial and harmful peer contagion effects are more or less
likely.
In the present research I have, in a variety of ways, considered circumstances
that pertain to the six issues that Hartup (2005) pointed out as important to
reflect upon in the study of peer relations. What is needed in longitudinal
sociometric research about peer relations and adjustment is a more systematic
approach to these issues and an integration of theoretical models with temporal
designs and applicable statistical models (see the discussion in, for example,
Collins, 2006). And last but not least, as Hartup also stressed, intervention
studies are needed to give us tools to help children improve their peer situation
and to increase our understanding of the processes of behavior change in the
peer domain.
65
References
Aboud, F. E., & Mendelson, M. J. (1996). Determinants of friendship selection and quality:
Developmental perspectives. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup
(Eds.), The company they keep: Friendship in childhood and adolescence (pp. 87-112).
New York: Cambridge University Press.
Adams, G. R., & Roopnarine, J. L. (1994). Physical attractiveness, social skills, and same-sex
peer popularity. Journal of Group Psychotherapy, Psychodrama and Sociometry, 47,
15-35.
Adebäck, B. (1969). Sociala relationer i skolan: Hemmiljö och familjerelationer [Social
relations at school: Home environment and family relations]. Unpublished licentiate
thesis, Stockholm University, Department of Psychology.
Agnew, R. (2003). An integrated theory of the adolescent peak in offending. Youth and
Society, 34, 263-299.
Albee, G. W. (1984). Prologue: A model for classifying prevention programs. In J. M. Joffe,
G. W. Albee, & L. D. Kelly (Eds.). Readings in primary prevention of
psychopathology: Basic concepts (pp. 228-245). Hanover, NH: University Press of New
England.
Allen, V. L. (1981). Self, social group, and social structure: Surmises about the study of
children's friendships. In S. R. Asher & J. M. Gottman (Eds.), The development of
children's friendships (pp. 182–203). New York: Cambridge University Press.
Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through
the twenties. American Psychologist, 55, 469-480.
Asher, S. R. (1985). An evolving paradigm in social skill training research with children. In
B. H. Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children's peer relations:
Issues in assessment and intervention (pp. 157-174). New York: Springer Verlag.
Asher, S. R. (1990). Recent advances in the study of peer rejection. In S. R. Asher, & J. D.
Coie (Eds.), Peer rejection in childhood (pp. 3-14). New York: Cambridge University
Press.
Asher, S. R., & Coie, J. D. (Eds.) (1990). Peer rejection in childhood. New York: Cambridge
University Press.
Asher, S. R., & Dodge, K. A. (1986). Identifying children who are rejected by their peers.
Developmental Psychology, 22, 444-449.
Asher, S. R., & Hymel, S. (1981). Children's social competence in peer relations: Sociometric
and behavioral assessment. In J. D. Wine & M. D. Smye (Eds.), Social competence (pp.
125-157). New York: Guilford.
Asher, S. R., Parker, J. G., & Walker, D. L. (1996). Distinguishing friendship from
acceptance: Implications for intervention and assessment. In W. M. Bukowski, A. F.
Newcomb, & W. W. Hartup (Eds.), The company they keep: Friendship in childhood
and adolescence (pp. 366-405). New York: Cambridge University Press.
Asher, S. R., & Wheeler, V. A. (1985). Children's loneliness: A comparison of rejected and
neglected peer status. Journal of Consulting and Clinical Psychology, 53, 500-505.
Aube, J., Fleury, J., & Smetana, J. (2000). Changes in women’s roles: Impact on and social
policy implications for the mental health of women and children. Development and
Psychopathology, 12, 633-656.
Babad, E. (2001) On the conception and measurement of popularity: more facts and some
straight conclusions. Social Psychology of Education, 5, 3-30.
66
Bagwell, C. L., Coie, J. D., Terry, R. A., & Lochman, J. E. (2000). Peer clique participation
and social status in preadolescence. Merrill-Palmer Quarterly, 46, 280-305.
Bagwell, C. L., Newcomb, A. F., & Bukowski, W. M. (1998). Preadolescent friendship and
peer rejection as predictors of adult adjustment. Child Development, 69, 140-153.
Bagwell, C. L., Schmidt, M. E., Newcomb, A. F., & Bukowski, W. M. (2001). Friendship and
peer rejection as predictors of adult adjustment. In D. W. Nangle & C. A. Erdley (Eds.),
The Role of Friendship in Psychological Adjustment. New directions for child
development, No 91 (pp. 25-50). San Francisco: Jossey-Bass.
Baltes, P. B., Staudinger, U. M., & Lindenberger, U. (1999). Lifespan psychology: Theory
and application to intellectual functioning. Annual Review of Psychology, 50, 471-507.
Barriga, A. Q., Morrison, E. M., Liau, A. K., & Gibbs, J. C. (2001). Moral cognition:
Explaining the gender difference in antisocial behavior. Merrill-Palmer Quarterly, 47,
532-562.
Barry, A. E. (2005). How attrition impacts the internal and external validity of longitudinal
research. Journal of School Health, 75, 267-270.
Barton, S. (1994). Chaos, self-organization, and psychology. American Psychologist, 49, 515.
Bell-Dolan, D. J., Foster, S. L., & Christopher, J. S. (1992). Children’s reactions to
participating in a peer relations study: An example of cost-effective assessment. Child
Study Journal, 22, 137-156.
Bell-Dolan, D. J., Foster, S. L., & Sikora, D. M. (1989). Effects of sociometric testing on
children's behavior and loneliness in school. Developmental Psychology, 25, 306-311.
Bell-Dolan, D. J., Foster, S. L., & Tishelman, A. C. (1989). An alternative to negative
nomination sociometric measures. Journal of Clinical Child Psychology, 18, 153-157.
Benenson, J., Apostoleris, N., & Parnass, J. (1998). The organization of children’s same-sex
peer relationships. In W. M. Bukowski & A. H. Cillessen (Eds.), Sociometry then and
now: Building on six decades of measuring children’s experiences with the peer group.
New directions for child development, No 80 (pp. 5-23). San Francisco: Jossey-Bass.
Bergman, L. R. (1998). A pattern-oriented approach to studying individual development:
Snapshots and processes. In R. B. Cairns, L. R. Bergman, & J. Kagan (Eds.), Methods
and models for studying the individual (pp. 83-121). Thousand Oaks, CA: Sage.
Bergman, L. R. (2000). Women’s health, work, and education in a life-span perspective.
Technical report 1: Theoretical background and overview of the data collection.
(Reports from the project Individual Development and Adaptation No. 70). Stockholm
University, Department of Psychology.
Bergman, L. R., & El-Khouri, B. M. (1987). EXACON: A Fortran 77 program for the exact
analysis of single cells in a contingency table. Educational and Psychological
Measurement, 47, 155-161.
Bergman, L. R., & Magnusson, D. (1997). A person-oriented approach in research on
developmental psychopathology. Development and Psychopathology, 9, 291-319.
Berndt, T. J. (1996). Exploring the effects of friendship quality on social development. In W.
M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The company they keep:
Friendship in childhood and adolescence (pp. 346-365). New York: Cambridge
University Press.
Berndt, T. J. (1999). Friends’ influence on students’ adjustment to school. Educational
Psychologist, 34, 15-28.
Berndt, T. J., & Ladd, G. W. (1989). Part Two: Peer relationships in the school context. In T.
J. Berndt, & G. W. Ladd (Eds.), Peer relationships in child development (pp. 133-134).
New York: Wiley.
67
Berry, J. M., & Jobe, J. B. (2002) Introduction: At the intersection of personality and adult
development. Journal of Research in Personality, 36, 283-286.
Bierman, K. L. (2004). Peer rejection: Developmental processes and intervention strategies.
New York: Guilford Press.
Boivin, M., Poulin, F., & Vitaro, F. (1994). Depressed mood and peer rejection in childhood.
Development and Psychopathology, 6, 483-498.
Bonney, M. E. (1943). Personality traits of socially successful and socially unsuccessful
children. Journal of Educational Psychology, 34, 449-472.
Boxer, P., Guerra, N. G., Huesmann, L. R. & Morales, J. (2005). Proximal peer-level effects
of a small group selected prevention on aggression in elementary school children: An
investigation of the peer contagion hypothesis. Journal of Abnormal Child Psychology,
33, 325-338.
Brendgen, M., Little, T. D, & Krappmann, L. (2000). Rejected children and their friends: A
shared evaluation of friendship quality? Merrill-Palmer Quarterly, 46, 45-70.
Brendgen, M., Vitaro, F., & Bukowski, W. M. (1998). Affiliation with delinquent friends:
Contributions of parents, self-esteem, delinquent behavior, and rejection by peers.
Journal of Early Adolescence, 18, 244-265.
Brendgen, M., Vitaro, F., Bukowski, W. M., Doyle, A. B. & Markiewicz, D. (2001).
Developmental profiles of peer social preference over the course of elementary school:
Associations with trajectories of externalizing and internalizing behavior.
Developmental Psychology, 37, 308-320.
Brendgen, M., Vitaro, F., Turgeon, L., & Poulin, F. (2002). Assessing aggressive and
depressed children’s social relations with classmates and friends: A matter of
perspective. Journal of Abnormal Child Psychology, 30, 609-624.
Bronfenbrenner, U. (1945). The measurement of sociometric status, structure and
development. Sociometry Monographs, 6.
Buhs, E. S., & Ladd, G. W. (2001). Peer rejection as an antecedent of young children’s school
adjustment: An examination of mediating processes. Developmental Psychology, 37,
550-560.
Bukowski, W. M. (2001). Friendship and the worlds of childhood. In D. W. Nangle & C. A.
Erdley (Eds.) The Role of Friendship in Psychological Adjustment. New directions for
child development, No 91 (pp. 93-105). San Francisco: Jossey-Bass.
Bukowski, W. M. & Adams, R. (2005). Peer relationships and psychopathology: Markers,
moderators, mediators, mechanisms, and meanings. Journal of Clinical Child and
Adolescent Psychology, 34, 3-10.
Bukowski, W. M., Gauze, C., Hoza, B., & Newcomb, A. F. (1993). Differences and
consistency between same-sex and other-sex peer relationships during early
adolescence. Developmental Psychology, 29, 255-263.
Bukowski, W. M., & Hoza, B. (1989). Popularity and friendship. Issues in theory,
measurement, and outcomes. In T. Berndt & G. Ladd (Eds.), Peer relationships in child
development (pp. 15-45). New York: Wiley.
Bukowski, W. M., & Newcomb, A. F. (1984). Stability and determinants of sociometric status
and friendship choice: A longitudinal perspective. Developmental Psychology, 20,
941-952.
Bukowski, W. M., & Newcomb, A. F. (1985). Variability in peer group perceptions: Support
for the "controversial" sociometric classification group. Developmental Psychology, 21,
1032-1038.
Bukowski, W. M., Newcomb, A. F., & Hartup W. W. (1996). Friendship and its significance
in childhood and adolescence: Introduction and comment. In W. M. Bukowski, A. F.
68
Newcomb, & W. W. Hartup (Eds.), The company they keep: Friendship in childhood
and adolescence (pp. 1-15). New York: Cambridge University Press.
Cairns, R., Xie, H., & Leung, M-C. (1998). The popularity of friendship and the neglect of
social networks: Toward a new balance. In W. M. Bukowski & A. H. Cillessen (Eds.),
Sociometry then and now: Building on six decades of measuring children’s experiences
with the peer group. New directions for child development, No 80 (pp. 25-53). San
Francisco: Jossey-Bass.
Campbell, S. B., & Cluss, P. (1982). Peer relationships of young children with behavior
problems. In K. H. Rubin & H. S. Ross (Eds.), Peer relationships and social skills in
childhood (pp. 323-352). New York: Springer Verlag.
Cantrell, V. L., & Prinz, R. J. (1985). Multiple perspectives of rejected, neglected, and
accepted children: Relation between sociometric status and behavioral characteristics.
Journal of Consulting and Clinical Psychology, 53, 884-889.
Card, N. A., Hodges, E. V. E., Little, T. D., & Hawley, P. H. (2005). Gender effects in peer
nominations for aggression and social status. International Journal of Behavioral
Development, 29, 146-155.
Casiglia, A. C., Lo Coco, A., & Zappulla, C. (1998). Aspects of social reputation and peer
relationships in Italian children: A cross-cultural perspective. Developmental
Psychology, 34, 723-730.
Chang, L., Lei, L., Li, K. K., Liu, H., Guo, B. Wang, Y., et al. (2005). Peer acceptance and
self-perceptions of verbal and behavioural aggression and social withdrawal.
International Journal of Behavioral Development, 29, 48-57.
Chinapah, E. (2000). Tonårsnormer i förändring [Are adolescent norms changing?]
Unpublished doctoral dissertation, Stockholm University, Department of Psychology.
Cicchetti, D. & Cohen, D. J. (Eds.). (1995). Developmental Psychology: Vol. 2. Risk,
disorder, and adaptation. New York: Wiley.
Cillessen, A. H. N., & Bukowski, W. M. (2000a). Conceptualizing and measuring peer
acceptance and rejection. In A. H. N. Cillessen & W. M. Bukowski (Eds.), Recent
advances in the measurement of acceptance and rejection in the peer system. New
directions for child development, No 88 (pp. 3-10). San Francisco: Jossey-Bass.
Cillessen, A. H. N., & Bukowski, W. M. (Eds.). (2000b). Recent advances in the
measurement of acceptance and rejection in the peer system. New directions for child
development, No 88. San Francisco: Jossey-Bass.
Cillessen, A. H. N., Bukowski, W. M., & Haselager, G. J. T. (2000).Stability of sociometric
categories. In A. H. N. Cillessen & W. M. Bukowski (Eds.), Recent advances in the
measurement of acceptance and rejection in the peer system. New directions for child
development, No 88 (pp. 75-93). San Francisco: Jossey-Bass.
Clarke, A. D. B., & Clarke, A. M. (1984). Constancy and change in the growth of human
characteristics. Journal of Child Psychology and Psychiatry, 25, 191-210.
Cohen, S. (2004). Social relationships and health. American Psychologist, 59, 676-684.
Coie, J. D. (1990). Toward a theory of peer rejection. In S. R. Asher, & J. D. Coie (Eds.),
Peer rejection in childhood (pp. 365-401). New York: Cambridge University Press.
Coie, J. D., & Dodge, K. A. (1983). Continuities of children's social status: A five year
longitudinal study. Merrill-Palmer Quarterly, 29, 261-282.
Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of social status: A
cross age perspective. Developmental Psychology, 18, 557-569.
Coie, J. D., Dodge, K. A., & Kupersmidt, J. B. (1990). Peer group behavior and social status.
In S. R. Asher, & J. D. Coie (Eds.), Peer rejection in childhood (pp. 17-59). New York:
Cambridge University Press.
69
Coie, J. D., Lochman, J. E., Terry, R., & Hyman, C. (1992). Predicting early adolescent
disorder from childhood aggression and peer rejection. Journal of Consulting and
Clinical Psychology, 60, 783-792.
Coleman, J. C. (1980). Friendship and the peer group in adolescence. In J. Adelson (Ed.),
Handbook of Adolescent Psychology (pp. 408-431). New York: John Wiley & Sons.
Collins, L. M. (2006). Analysis of longitudinal data: The integration of theoretical model,
temporal design, and statistical model. Annual Review of Psychology, 57, 505-528.
Costanzo, P. R., & Frankel, P. (1987). Social influence, socialization, and the development of
social cognition: The heart of the matter. In N. Eisenberg (Ed.), Advances in
developmental psychology (pp. 190-215). New York: Wiley.
Cowen, E. L., Pederson, A., Babigian, H., Izzo, L. D., & Trost, M. A. (1973). Long-term
follow-up of early detected vulnerable children. Journal of Consulting and Clinical
Psychology, 41, 438-446.
Cox, S. H. (1974). The association of peer acceptance-rejection with children's perception of
parental behaviors. Psychology in the Schools, 11, 222-225.
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social informationprocessing mechanisms in children's social adjustment. Psychological Bulletin, 115, 74–
101.
Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and socialpsychological adjustment. Child Development, 66, 710-722.
Crick, N. R., & Ladd, G. W. (1989). Nominator attrition: Does it affect the accuracy of
children's sociometric classifications? Merrill-Palmer Quarterly, 35, 197-207.
Crick, N. R., & Ladd, G. W. (1993). Children's perceptions of their peer experiences:
Attributions, loneliness, social anxiety, and social avoidance. Developmental
Psychology, 29, 244-54.
Crick, N. R., & Zahn-Waxler, C. (2003). The development of psychopathology in females and
males: Current progress and future challenges. Development and Psychopathology, 15,
719-742.
Culbertson, F. M. (1997). Depression and gender: An international review. American
Psychologist, 52, 25-31.
Cullerton-Sen, C., & Crick, N. R. (2005). Understanding the effects of physical and relational
victimization: The utility of multiple perspectives in predicting social-emotional
adjustment. School Psychology Review, 34, 147-160.
Czeschlik, T., & Rost, D. H. (1995). Sociometric types and children's intelligence. British
Journal of Developmental Psychology, 13, 177-189.
Damon, W. (1983). Social and Personality Development: Infancy through Adolescence. New
York: Norton.
Damon, W. (Series Ed.) & Eisenberg, N. (Vol. Ed.). (1998). Handbook of Child Psychology:
Vol. 3. Social, emotional, and personality development (5th ed). New York: Wiley.
Davitz, J. R. (1955). Social perception and sociometric choice of children. Journal of
Abnormal and Social Psychology, 50, 173-176.
Dayan, J., Doyle, A.-B., & Markiewicz, D. (2001). Social support networks and self-esteem
of idiocentric and allocentric children and adolescents. Journal of Social and Personal
Relationships, 18, 767-784.
Deater-Deckard, K. (2001). Annotation: Recent research examining the role of peer
relationships in the development of psychopathology. Journal of Child Psychology and
Psychiatry, 42, 565-579.
Deptula, D. P., & Cohen, R. (2004). Aggressive, rejected, and delinquent children and
adolescents: a comparison of their friendships. Aggression and Violent Behavior, 9, 75104.
70
De Rosier, M. E., Kupersmidt, J. B., & Patterson, C. J. (1994). Children's academic and
behavioral adjustment as a function of the cronicity and proximity of peer rejection.
Child Development, 65, 1799-1813.
DeRosier, M. E., & Thomas, J. M. (2003). Strengthening sociometric prediction: Scientific
advances in the assessment of children’s peer relations. Child Development, 75, 13791392.
Dishion, T. J., & Dodge, K. A. (2005). Peer contagion in interventions for children and
adolescents: Moving towards an understanding of the ecology and dynamics of change.
Journal of Abnormal Child Psychology, 33, 395-400.
Dodge, K. A., & Feldman, E. (1990). Issues in social cognition and sociometric status. In S.
R. Asher, & J. D. Coie (Eds.), Peer rejection in childhood (pp. 119-155). New York:
Cambridge University Press.
Dodge, K. A., Lansford, J. E., Salzer Burks, V., Bates, J. E., Pettit, G. S., Fontaine, R., &
Price, J. M. (2003). Peer rejection and social information-processing factors in the
development of aggressive behavior problems in children. Child Development, 74, 374393.
Duck, S. W. (1975). Personality, similarity, and friendship choices by adolescents. European
Journal of Social Psychology, 5, 351-365.
Duncan, M. K., & Cohen, R. (1995). Liking within the peer group as a function of children's
sociometric status and sex. Child Study Journal, 25, 265-287.
Dunnington, M. J. (1957). Behavioral differences of sociometric status groups in a nursery
school. Child Development, 28, 103-111.
Eisenberg, N. (1998). Introduction. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.),
Handbook of Child Psychology: Vol. 3. Social, emotional, and personality development
(5th ed., pp. 1-24). New York: Wiley.
Eitzen, D. S. (1975). Athletics in the status system of male adolescents: A replication of
Coleman's “The Adolescent Society”. Adolescence, 10, 267-276.
Ellis, S., Rogoff, B., & Cromer, C. C. (1981). Age segregation in children's social
interactions. Developmental Psychology, 17, 399-407.
Epstein, J. L. (1989). The selection of friends. Changes across the grades and in different
school environments. In T. Berndt & G. Ladd (Eds.), Peer relationships in child
development (pp. 158-187). New York: Wiley.
Erikson, E. H. (1963). Childhood and society. New York: Norton.
Fabes, R. A., Martin, C. L., & Hanish, L. D. (2004). The next 50 years: Considering gender as
a context for understanding young children’s peer relationships. Merrill-Palmer
Quarterly, 50, 260-273.
Farmer, T. W., Estell, D. B., Bishop, J. L., O’Neal, K. K., & Cairns, B. D. (2003). Rejected
bullies or popular leaders? The social relations of aggressive subtypes of rural African
American early adolescents. Developmental Psychology, 39, 992-1004.
Farmer, T. W., Estell, D. B., Leung, M.-C., Trott, H., Bishop, J., & Cairns, B. D. (2003).
Individual characteristics, early adolescent peer affiliations, and school dropout: an
examination of aggressive and popular group types. Journal of School Psychology, 41,
217-232.
Farmer, T. W., & Rodkin, P. C. (1996). Antisocial and prosocial correlates of classroom
social positions: The social network centrality perspective. Social Development, 5, 174188.
Fasick, F. A. (1984). Parents, peers, youth culture and autonomy in adolescence. Adolescence,
19, 143-157.
Feinberg, M. R. (1964). Stability of sociometric status in two adolescent class groups. Journal
of Genetic Psychology, 104, 83-87.
71
Fergusson, D. M., Swain-Campbell, N. R., & Horwood, L. J. (2002). Deviant peer affiliations,
crime and substance use: A fixed effects regression analysis. Journal of Abnormal Child
Psychology, 30, 419-430.
Fisher, R. A., & Yates, F. (1953). Statistical tables for biological, agricultural and medical
research. London, England: Oliver & Boyd Ltd.
Flook, L. Repetti, R. L., & Ullman J. B. (2005). Classroom social experiences as predictors of
academic performance. Developmental Psychology, 41, 319-327.
Frederickson, N. L., & Furnham, A. F. (1998). Sociometric classification methods in school
peer groups: A comparative investigation. Journal of Child Psychology and Psychiatry,
39, 921-933.
Frederickson, N. L., & Furnham, A. F. (2001). The long-term stability of sociometric status
classification: A longitudinal study of included pupils who have moderate learning
difficulties and their mainstream peers. Journal of Child Psychology and Psychiatry, 42,
581-592.
Frederickson, N. L., & Furnham, A. F. (2004). Peer-assessed behavioral characteristics and
sociometric rejection: Differences between pupils who have moderate learning
difficulties and their mainstream peers. British Journal of Educational Psychology, 74,
391-410.
French, D. C. (1984). Children's knowledge of the social functions of younger, older, and
same-age peers. Child Development, 55, 1429-1433.
French, D. C. (1988). Heterogeneity of peer-rejected boys: Aggressive and nonaggressive
subtypes. Child Development, 59, 976-985.
French, D. C. (1990). Heterogeneity of peer-rejected girls. Child Development, 61,
2028-2031.
French, D. C., & Conrad, J. (2001). School dropout as predicted by peer rejection and
antisocial behavior. Journal of Research on Adolescence, 11, 225-244.
French, D. C., & Waas, G. A. (1985). Behavior problems of peer-neglected and peer-rejected
elementary-age children: Parent and teacher perspectives. Child Development, 56,
246-252.
Furfey, P. H. (1927). Some factors influencing the selection of boys' chums. Journal of
Applied Psychology, 11, 47-51.
Furman, W., & Buhrmester, D. (1985). Children's perceptions of the personal relationships in
their social networks. Developmental Psychology, 21, 1016-1024.
George, T. P., & Hartmann, D. P. (1996). Friendship networks of unpopular, average, and
popular children. Child Development, 67, 2301-2316.
Gest, S. D., Graham-Bermann, S. A., & Hartup, W. W. (2001). Peer experience: Common and
unique features of number of friendships, social network centrality, and sociometric
status. Social Development, 10, 23-40.
Gifford-Smith, M. E., & Brownell, C. A. (2003). Childhood peer relationships: social
acceptance, friendships, and peer networks. Journal of School Psychology, 41, 235-284.
Gottman, J. M. (1977). Toward a definition of social isolation in children. Child
Development. 48, 513-517.
Gottman, J. M., Gonso, J., & Rasmussen, B. (1975). Social interaction, social competence,
and friendship in children. Child Development, 46, 709-718.
Green, K. D., Vosk, B., Forehand, R., & Beck, S. (1981). An examination of differences
among sociometrically identified accepted, rejected, and neglected children. Child Study
Journal, 11, 117-124.
Gronlund, N. E. (1959). Sociometry in the classroom. New York: Harper.
Grotpeter, J. K., & Crick, N. R. (1996). Relational aggression, overt aggression, and
friendship. Child Development, 67, 2328-2338.
72
Guerra, V. S., Asher, S. S., & DeRosier, M. E. (2004). Effect of children’s perceived rejection
on physical aggression. Journal of Abnormal Child Psychology, 32, 551-563.
Hallinan, M. T. (1981). Recent advances in sociometry. In S. R Asher & J. M. Gottman
(Eds.), The development of children's friendships (pp. 91-115). New York: Cambridge
University Press.
Halpern, D. F. (2005). Psychology at the intersection of work and family: Recommendations
for employers, working families, and policymakers. American Psychologist, 60, 397409.
Harkness, S. (1980). The cultural context of child development. In C. M. Super & S.
Harkness (Eds.), New directions for child development: Vol. 8. Anthropological
perspectives on child development (pp. 7-15). San Francisco: Jossey-Bass.
Harkness, S., & Super, C. M. (1985). The cultural context of gender segregation in children's
peer groups. Child Development, 56, 219-224.
Hartup, W. W. (1976). Cross-age vs. same-age peer interactions: Ethological and crosscultural perspectives. In V. Allen (Ed.), Children as teachers: Theory and research on
tutoring (pp. 41-54). New York: Academic Press.
Hartup, W. W. (1983). Peer relations. In E. M. Hetherington (Ed.), Handbook of child
psychology (Vol. 4): Socialization, personality, and social development (pp. 103-196).
New York: Wiley.
Hartup, W. W. (1989). Social relationships and their developmental significance. American
Psychologist, 44, 120-126.
Hartup, W. W. (1995). Three faces of friendship. Journal of Social and Personal
Relationships, 12, 569-574.
Hartup, W. W. (1996). The company they keep: Friendships and their developmental
significance. Child Development, 67, 1-13.
Hartup, W. W. (1999). Constraints on peer socialization: Let me count the ways. MerrillPalmer Quarterly, 45, 172-183.
Hartup, W. W. (2005). Peer interaction: What causes what? Journal of Abnormal Child
Psychology, 33, 387-394.
Hartup, W. W., & Abecassus, M. (2002). Friends and enemies. In P. K. Smith, & C. H. Hart
(Eds.), Blackwell Handbook of Childhood Social Development (pp. 285-306). Malden,
MA: Blackwell Publishers.
Hartup, W. W., & Stevens, N. (1997). Friendships and adaptation in the life course.
Psychological Bulletin, 121, 355-370.
Hartup, W. W., & van Lieshout, C. F. M. (1996). Personality development in social context.
Annual Review of Psychology, 46, 655-687.
Haselager, G. J. T., Van Lieshout, C. F. M., Riksen-Walraven, J. M. A., Cillessen, A. H. N.,
& Hartup, W. W. (2002). Heterogeneity among peer-rejected boys across middle
childhood: Developmental pathways of social behavior. Developmental Psychology, 38,
446-456.
Hatzichristou, C., & Hopf, D. (1996). A multiperspective comparison of peer sociometric
status groups in childhood and adolescence. Child Development, 67, 1085-1102.
Hay, D. F., Payne, A., & Chadwick, A. (2004). Peer relations in childhood. Journal of Child
Psychology and Psychiatry, 45, 84-108.
Hayden-Thomson, L., Rubin, K. H., & Hymel, S. (1987). Sex preferences in sociometric
choices. Developmental Psychology, 23, 558-562.
Hayvren, M., & Hymel, S. (1984). Ethical issues in sociometric testing: The impact of
sociometric measures on interaction behavior. Developmental Psychology, 20, 844-849.
Heinze, H. J., Toro, P. A., & Urberg, K. A. (2004). Antisocial behavior and affiliation with
deviant peers. Journal of Clinical Child and Adolescent Psychology, 33, 336-346.
73
Helson, R., & Soto, C. J. (2005). Up and down in middle age: Monotonic and nonmonotonic
changes in roles, status, and personality. Journal of Personality and Social Psychology,
89, 194-204.
Henricson, M. (1971). Tonåringars normer och normklimat [The norms and the norm climate
of adolescents]. Unpublished master’s thesis, Stockholm University, Department of
Psychology.
Henricsson, L., & Rydell, A.-M. (2006). Children with behaviour problems: The influence of
social competence and social relations on problem stability, school achievement and
peer acceptance across the first six years of school. Infant and Child Development, 15,
347-366.
Hindy, C. G. (1980). Children's friendship concepts and the perceived cohesiveness of same
sex friendship dyads. Psychological Reports, 47, 191-203.
Hinshaw, S. P., Lahey, B. B., & Hart, E. L. (1993). Issues of taxonomy and comorbidity in
the development of conduct disorder. Development and Psychopathology, 5, 31-49.
Hoff, K. E., DuPaul, G. J., & Handwerk, M. L. (2003). Rejected youth in residential
treatment: Social affiliation and peer group configuration. Journal of Emotional and
Behavioral Disorders, 11, 112-121.
Horowitz, F. D. (1962). The relationship of anxiety, self-concept, and sociometric status
among fourth, fifth, and sixth grade children. Journal of Abnormal and Social
Psychology, 65, 212-214.
Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60, 581-592.
Hymel, S., & Asher, S. R. (1977). Assessment and training of isolated children’s social skills.
Paper presented at the biennial meeting of the Society for Research in Child
Development, New Orleans (ERIC Document Reproduction Service No. ED 136 930).
Hymel, S., Bowker, A., & Woody, E. (1993). Aggressive versus withdrawn unpopular
children: Variations in peer and self-perceptions in multiple domains. Child
Development, 64, 879-896.
Hymel, S., & Franke, S. (1985). Children's peer relations: Assessing self-perceptions. In B. H.
Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children's peer relations: Issues in
assessment and intervention (pp. 75-92). New York: Springer Verlag.
Hymel, S., Vaillancourt, T., McDougall, P., & Renshaw P. D. (2002). Peer acceptance and
rejection in childhood. In P. K. Smith, & C. H. Hart (Eds.), Blackwell Handbook of
Childhood Social Development (pp. 265-284). Malden, MA: Blackwell Publishers.
Iverson, A. M., Barton, E. A., & Iverson, G .L. (1997). Analysis of risk to children
participating in a sociometric task. Developmental Psychology, 33, 104-112.
Janes, C. L., Hesselbrock, V. M., Myers, D. G., & Penniman, J. H. (1979). Problem boys in
young adulthood: Teachers' ratings and twelve year follow-up. Journal of Youth and
Adolescence, 8, 453-472.
Jiang, X. L., & Cillessen, A. H. N. (2005). Stability of continuous measures of sociometric
status: a meta-analysis. Developmental Review, 25, 1-25.
Johnson, J. C., Ironsmith, M., & Poteat, G. M. (1994). Assessing children’s sociometric
status: Issues and the application of social network analysis. Journal of Group
Psychotherapy, Psychodrama and Sociometry, 47, 36-48.
Josephson, I. (1967). Sociala relationer och skolprestation: En empirisk studie av barn i olika
åldrar [Social relations and school achievement: An empirical study of children in
different ages]. Unpublished licentiate thesis, Stockholm University, Department of
Psychology.
Jung, J. (1984). Social support and its relation to health: A critical evaluation. Basic and
Applied Social Psychology, 5, 143-169.
74
Kandel, D. B. (1978). Similarity in real-life adolescent friendship pairs. Journal of
Personality and Social Psychology, 36, 313-323.
Kane, J. S., & Lawler, E. E. (1978). Methods of peer assessment. Psychological Bulletin, 85,
555-586.
Keane, S. P., & Calkins, S. D. (2004). Predicting kindergarten peer social status from toddler
and preschool problem behavior. Journal of Abnormal Child Psychology, 32, 409-423.
Kiesner, J., Kerr, M., & Stattin, H. (2004). “Very important persons” in adolescence: going
beyond in-school, single friendships in the study of peer homophily. Journal of
Adolescence, 27, 545-560.
Kiesner, J., & Pastore, M. (2005). Differences in the relations between antisocial behavior and
peer acceptance across contexts and across adolescence. Child Development, 76, 12781293.
Kindermann, T. (1998). Children’s development within peer groups: Using composite social
maps to identify peer networks and to study their influences. In W. M. Bukowski & A.
H. Cillessen (Eds.), Sociometry then and now: Building on six decades of measuring
children’s experiences with the peer group. New directions for child development, No
80 (pp. 55-82). San Francisco: Jossey-Bass.
Kistner, J., Balthazor, M., Risi, S., & Burton, C. (1999). Predicting dysphoria in adolescence
from actual and perceived peer acceptance in childhood. Journal of Clinical Child
Psychology, 28, 94-104.
Koch, H. L. (1933). Popularity among preschool children: Some related factors and a
technique for its measurement. Child Development, 4, 164-175.
Kosir, K. & Pecjak, S. (2005). Sociometry as a method for investigating peer relationships:
what does it actually measure? Educational Research, 47, 127-144.
Kossek, E. E., & Ozeki, C. (1998). Work-family conflict, policies, and the job-life satisfaction
relationship: A review and directions for organizational behavior-human resources
research. Journal of Applied Psychology, 83, 139-149.
Krantz, M. & Burton, C. (1986). The development of the social cognition of social status.
Journal of Genetic Psychology, 147, 89-95.
Krappmann, L. (1996). Amicitia, drujba, shin-yu, philia, freundschaft, friendship: On the
cultural diversity of a human relationship. In W. M. Bukowski, A. F. Newcomb, & W.
W. Hartup (Eds.), The company they keep: Friendship in childhood and adolescence
(pp. 19-40). New York: Cambridge University Press.
Kuhlen, R. G., & Lee, B. J. (1943). Personality characteristics and social acceptability in
adolescence. Journal of Educational Psychology, 34, 321-340.
Kupersmidt, J. B., & Coie, J. D. (1990). Preadolescent peer status, aggression, and school
adjustment as predictors of externalizing problems in adolescence. Child Development,
61, 1350-1362.
Kupersmidt, J. B., Coie, J. D., & Dodge, K. A. (1990). The role of poor peer relationships in
the development of disorder. In S. R. Asher, & J. D. Coie (Eds.), Peer rejection in
childhood (pp. 274-305). 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 demographic characteristics. Journal of Social and
Personal Relationships, 12, 439-452.
Ladd, G. W. (1983). Social networks of popular, average, and rejected children in school
settings. Merrill-Palmer Quarterly, 29, 283-307.
Ladd, G. W. (1984). Expanding our view of the child's social world: New territories, new
maps, same directions? Merrill-Palmer Quarterly, 30, 317-320.
75
Ladd, G. W. (1990). Having friends, keeping friends, making friends, and being liked by
peers in the classroom: Predictors of children's early school adjustment? Child
Development, 61, 1081-1100.
Ladd, G. W. (1999). Peer relationships and social competence during early and middle
childhood. Annual Review of Psychology, 50, 333-359.
Ladd, G. W. (2006). Peer rejection, aggressive or withdrawn behavior, and psychological
maladjustment from ages 5 to 12: An examination of four predictive models. Child
Development, 77, 822-846.
Ladd, G. W., & Coleman, C. C. (1997). Children’s classroom peer relationships and early
school attitudes: Concurrent and longitudinal associations. Early Education and
Development, 8, 51-66.
Ladd, G. W., & Kochenderfer, B. J. (1996). Linkages between friendship and adjustment
during early school transitions. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup
(Eds.), The company they keep: Friendship in childhood and adolescence (pp. 322-345).
New York: Cambridge University Press.
Ladd, G. W., & Troop-Gordon, W. (2003). The role of chronic peer difficulties in the
development of children’s psychological adjustment problems. Child Development, 74,
1344-1367.
Laird, R. D., Jordan, K. Y., Dodge, K. A., Pettit, G. S., & Bates, J. E. (2001). Peer rejection in
childhood, involvement with antisocial peers in early adolescence, and the development
of externalizing behavior problems. Development and Psychopathology, 13, 337-354.
Laird, R. D., Pettit, G. S., Dodge, K. A., & Bates, J. E. (2005). Peer relationship antecedents
of delinquent behavior: Is there evidence of demographic group differences in
developmental processes? Development and Psychopathology, 17, 127-144.
Laursen, B. (1996). Closeness and conflict in adolescent peer relationships: Interdependence
with friends and romantic partners. In W. M. Bukowski, A. F. Newcomb, & W. W.
Hartup (Eds.), The company they keep: Friendship in childhood and adolescence (pp.
186-210). New York: Cambridge University Press.
Laursen, B., & Hoff, E. (Eds.). (2006). Person- centered and variable-centered approaches to
longitudinal data [Special issue]. Merrill-Palmer Quarterly, 52(3).
Laursen, B., & Mooney, K. S. (2005). Why do friends matter? Human Development, 48, 323326.
Lemann, T. B., & Solomon, R. L. (1952). Group characteristics as revealed in sociometric
patterns and personality ratings. Sociometry, 15, 7-90.
Levinson, D. J. (1986). A conception of adult development. American Psychologist, 41, 3-13.
Lippitt, R. (1941). Popularity among preschool children. Child Development, 12, 305-322.
Maccoby, E. E. (1990) Gender and relationships: A developmental account. American
Psychologist, 45, 513-520.
Maccoby, E. E. (2000). Perspectives on gender development. International Journal of
Behavioral Development, 24, 398-406.
Magnusson, D. (Ed.) (1996). The life-span development of individuals: Behavioral,
neurobiological and psychosocial perspectives. Cambridge, England: Cambridge
University Press.
Magnusson, D., Dunér, A., & Zetterblom, G. (1975). Adjustment: A longitudinal study.
Stockholm: Almqvist & Wiksell. New York: Wiley.
Masten, A. S. (2005). Peer relationships and psychopathology in developmental perspective:
Reflections on progress and promise. Journal of Clinical Child and Adolescent
Psychology, 34, 87-92.
Masters, J. C., & Furman, W. (1981). Popularity, individual friendship selection, and specific
peer interaction among children. Developmental Psychology, 17, 344-350.
76
McLellan, J. A., & Pugh, M. J. V. (1999). Editors’ notes. In J. A. McLellan & M. J. V. Pugh
(Eds.), The role of peer groups in adolescent social identity: Exploring the importance
of stability and change . New directions for child development, No 84 (pp. 1-6). San
Francisco: Jossey-Bass.
Mednick, S. A., Harway, M., & Finello, K. M. (1984). Handbook of Longitudinal research.
Volume I. Birth and childhood cohorts. New York: Praeger.
Meleis, A. I., & Lindgren, T. G. (2002). Man works from sun to sun, but woman’s work is
never done: Insights on research and policy. Health Care for Women International, 23,
742-753.
Meulemann, H. (2003). Biographical self-definitions from adolescence to adulthood and
beyond. In J. J. Arnett & N. L. Galambos (Eds.), Exploring cultural conceptions of the
transition to adulthood. New Directions for Child and Adolescent Development, No 100
(pp. 51-62). San Francisco: Jossey-Bass.
Miller-Johnson, S., Coie, J. D., Maumary-Gremaud, A. Bierman, K., & Conduct Problems
Prevention Research Group (2002). Peer rejection and aggression and early starter
models of conduct disorder. Journal of Abnormal Child Psychology, 30, 217-230.
Moreno, J. L. (1934). Who shall survive? A new approach to the problem of human
interrelations. Washington, D.C.: Nervous and Mental Disease Publishing Co. (also
New York: Beacon House, 1953).
Moreno, J. L. (1947). Progress and pitfalls in sociometric theory. Sociometry, 10, 268-272.
Nagin, D. S. (1999). Analyzing developmental trajectories: A semiparametric, group-based
approach. Psychological Methods, 4, 139-157.
Nelson, S. E., & Dishion, T. J. (2004). From boys to men: Predicting adult adaptation from
middle childhood sociometric status. Development and Psychopathology, 16, 441-459.
Nesselroade, J. R., Stigler, S. M., & Baltes, P. B. (1980). Regression toward the mean and the
study of change. Psychological Bulletin, 88, 622-637.
Newcomb, A. W., & Bagwell, C. L. (1996). The developmental significance of children’s
friendship relations. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The
company they keep: Friendship in childhood and adolescence (pp. 289-321). New
York: Cambridge University Press.
Newcomb, A. F., & Bukowski, W. M. (1983). Social impact and social preference as
determinants of children's peer group status. Developmental Psychology, 19, 856-867.
Newcomb, A. F., & Bukowski, W. M. (1984). A longitudinal study of the utility of social
preference and social impact sociometric classification schemes. Child Development,
55, 1434- 1447.
Newcomb, A. F., Bukowski, W. M., & Pattee, L. (1993). Children's peer relations: A metaanalytic review of popular, rejected, neglected, controversial, and average sociometric
status. Psychological Bulletin, 113, 99-128.
Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender differences in
depression during adolescence. Psychological Bulletin, 115, 424-443.
Northway, M. L. (1946). Personality and sociometric status: A review of the Toronto studies.
Sociometry, 9, 233-241.
Ollendick, T. H., Greene, R. W., Francis, G., & Baum, C. G. (1991). Sociometric status: Its
stability and validity among neglected, rejected, and popular children. Journal of Child
Psychology and Psychiatry, 32, 523-534.
Ollendick, T. H., Weist, M. D., Borden, M. C., & Greene, R. W. (1992). Sociometric status
and academic, behavioral, and psychological adjustment: A five-year longitudinal
study. Journal of Consulting and Clinical Psychology, 60, 80-87.
77
Olson, S. L., & Brodfeld, P. L. (1991). Assessment of peer rejection and externalizing
behavior problems in preschool boys: A short term longitudinal study. Journal of
Abnormal Child Psychology, 19, 493-503.
Ozer, E. M. (1995). The impact of childcare responsibility and self-efficacy on the
psychological health of professional working mothers. Psychology of Women Quarterly,
19, 315-335.
Pakaslahti, L., & Keltikangas-Järvinen, L. (2001). Peer-attributed prosocial behavior among
aggressive/preferred, aggressive/non-preferred, non-aggressive/preferred and nonaggressive/non-preferred adolescents. Personality and Individual Differences, 30, 903916.
Pardini, D. A., Barry, T. D., Barth, J. M., Lochman, J. E., & Wells, K. C. (2006). Selfperceived social acceptance and peer social standing in children with aggressivedisruptive behaviors. Social Development, 15, 46-64.
Parke, R. D., O'Neil, R., Spitzer, S., Isley, S. Welsh, M., Wang, S., Lee, J., Strand, C., &
Cupp, R. (1997). A longitudinal assessment of sociometric stability and the behavioral
correlates of children's social acceptance. Merrill-Palmer Quarterly, 43, 635-662.
Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are lowaccepted children at risk? Psychological Bulletin, 102, 357-389.
Parker, J. G., Rubin, K. H., Price, J. M., & DeRosier, M. E. (1995). Peer relationships, child
development, and adjustment: A developmental psychopathology perspective. In D.
Cicchetti & D. J. Cohen (Eds.), Developmental Psychology: Vol. 2. Risk disorder, and
adaptation (pp. 96-161). New York: Wiley.
Parkhurst, J. T., & Asher, S. R. (1992). Peer rejection in middle school: Subgroup differences
in behavior, loneliness, and interpersonal concerns. Developmental Psychology, 28,
231-241.
Parkhurst J. T., & Hopmeyer A. (1998). Sociometric popularity and peer-perceived
popularity: Two distinct dimensions of peer status. Journal of Early Adolescence, 18,
125-144.
Peake, A., & Harris, K. L. (2002). Young adults’ attitudes toward multiple role planning: The
influence of gender, career traditionality, and marriage plans. Journal of Vocational
Behavior, 60, 405-421.
Peck, S. C., & Roeser, R. W. (Eds.). (2003). Person-centered approaches to studying
development in context [Special issue]. New Directions for Child and Adolescent
Development, No 101. San Francisco: Jossey-Bass.
Peery, J. C. (1979). Popular, amiable, isolated, rejected: A reconceptualization of sociometric
status in preschool children. Child Development, 50, 1231-1234.
Piquero, N. L., Gover, A. R., MacDonald, J. M., & Piquero A. R. (2005). The influence of
delinquent peers on delinquency. Does gender matter? Youth & Society, 36, 251-275.
Pope, A. W., Bierman, K. L., & Mumma, G. H. (1991). Aggression, hyperactivity, and
inattention-immaturity: Behavior dimensions associated with peer rejection in
elementary school boys. Developmental Psychology, 27, 663-671.
Prinstein, M. J., & La Greca, A. M. (2004). Childhood peer rejection and aggression as
predictors of adolescent girls’ externalizing and health risk behaviors: A 6-year
longitudinal study. Journal of Consulting and Clinical Psychology, 72, 103-112.
Prinstein, M. J., & Wang, S. S. (2005). False consensus and adolescent peer contagion:
Examining discrepancies between perceptions and actual reported levels of friends’
deviant and health risk behaviors. Journal of Abnormal Child Psychology, 33, 293-306.
Putallaz, M., & Gottman, J. M. (1981). An interactional model of children's entry into peer
groups. Child Development, 52, 986-994.
78
Putallaz, M., & Heflin, A. H. (1990). Parent-child interaction. In S.R. Asher, & J.D. Coie
(Eds.), Peer rejection in childhood (pp. 189-216). New York: Cambridge University
Press.
Putallaz, M., & Wasserman, A. (1989). Children's naturalistic entry behavior and sociometric
status: A developmental perspective. Developmental Psychology, 25, 297-305.
Reinherz; H. Z., Giaconia, R. M., Hauf, A. M. C., Wasserman, M. S., & Paradis, A. D.
(2000). General and specific childhood risk factors for depression and drug disorders by
early adulthood. Journal of the American Academy of Child and Adolescent Psychiatry,
39, 223-231.
Roff, M., Sells, B., & Golden, M. M. (1972). Social adjustment and personality development
in children. Minneapolis: University of Minnesota Press.
Roistacher, R. C. (1974). A microeconomic model of sociometric choice. Sociometry, 37,
219-238.
Rokach, A. (2000). Perceived causes of loneliness in adulthood. Journal of Social Behavior
and Personality, 15, 67-84.
Rubin, K. H., Bukowski, W., & Parker, J. G. (1998). Peer interactions, relationships, and
groups. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of Child
Psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp. 610700). New York: Wiley.
Rubin, K. H., Burgess, K. B., & Coplan, R. J. (2002). Social withdrawal and shyness. In P. K.
Smith, & C. H. Hart (Eds.), Blackwell Handbook of Childhood Social Development (pp.
329-352). Malden, MA: Blackwell Publishers.
Rubin, K. H., LeMare, L. J., & Lollis, S. (1990). Social withdrawal in childhood:
Developmental pathways to peer rejection. In S.R. Asher, & J.D. Coie (Eds.), Peer
rejection in childhood (pp. 217-249). New York: Cambridge University Press.
Rys, G. S., & Bear, G. G. (1997). Relational aggression and pear relations: Gender and
developmental issues. Merrill-Palmer Quarterly, 43, 87-106.
Salmivalli, C., & Isaacs, J. (2005). Prospective relations among victimization, rejection,
friendlessness, and children’s self- and peer-perceptions. Child Development, 76, 11611171.
Salmivalli, C., Kaukiainen, A., & Lagerspetz, K. (2000). Aggression and sociometric status
among peers: Do gender and type of aggression matter? Scandinavian Journal of
Psychology, 41, 17-24.
Sameroff, A. (1975). Transactional models in early social relations. Human Development, 18,
65-79.
Sandstrom, M. J., & Cillessen, A. H. N. (2003). Sociometric status and children’s peer
experiences: Use of the daily diary method. Merrill-Palmer Quarterly, 49, 427-452.
Sandstrom, M. J., & Coie, J. D. (1999). A developmental perspective on peer rejection:
Mechanisms of stability and change. Child Development, 70, 955-966.
Schneider, B. H., Atkinson, L., & Tardif, C. (2001). Child-parent attachment and children’s
peer relations: A quantitative review. Developmental Psychology, 37, 86-100.
Schultheiss, D. E. P. (2006). The interface of work and family life. Professional Psychology:
Research and Practice, 37, 334-341.
Singleton, L. C., & Asher, S. R. (1977). Peer preferences and social interaction among thirdgrade children in an integrated school district. Journal of Educational Psychology, 69,
330-336.
Smith, A. B., & Inder, P. M. (1990). The relationship of classroom organisation to cross-age
and cross-sex friendships. Educational Psychology, 10, 127-140.
79
Smollar, J., & Youniss, J. (1982). Social development through friendship. In K. H. Rubin &
H. S. Ross (Eds.), Peer relationships and social skills in childhood (pp. 279-298). New
York: Springer Verlag.
Sroufe, L. A., & Rutter, M. (1984). The domain of developmental psychopathology. Child
Development, 55, 17-29.
Stein, J. A., & Newcomb, M. D. (1999). Adult outcomes of adolescent conventional and
agentic orientations: A 20-year longitudinal study. Journal of Early Adolescence, 19,
39-65.
Strough, J., & Covatto, A. M. (2002). Context and age differences in same- and other-gender
peer preferences. Social Development, 11, 346-361.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York, NY: Norton.
Terry, R. (2000). Recent advances in measurement theory and the use of sociometric
techniques. In A. H. N. Cillessen & W. M. Bukowski (Eds.), Recent advances in the
measurement of acceptance and rejection in the peer system. New directions for child
development, No 88 (pp. 27-53). San Francisco: Jossey-Bass.
Terry, R., & Coie, J. D. (1991). A comparison of methods for defining sociometric status
among children. Developmental Psychology, 27, 867-880.
Thompson, G. G., & Powell, M. (1951). An investigation of the rating- scale approach to the
measurement of social status. Educational and Psychological Measurement, 11,
440-455.
Vaillancourt, T., & Hymel, S. (2006). Aggression and social status: The moderating roles of
sex and peer-valued characteristics. Aggressive Behavior, 32, 396-408.
Valdivia, I. A., Schneider, B. H., Chavez, K. L., & Chen; X. (2005). Social withdrawal and
maladjustment in a very group-oriented society. International Journal of Behavioral
Development, 29, 219-228.
van Lier, P. A. C., Vitaro, F., Wanner, B., Vujk, P., & Crijnen A. A. M. (2005). Gender
differences in developmental links among antisocial behavior, friends’ antisocial
behavior, and peer rejection in childhood: Results from two cultures. Child
Development, 76, 841-855.
Vaughn, B. E., & Waters, E. (1981). Attention structure, sociometric status, and dominance:
Interrelations, behavioral correlates, and relationships to social competence.
Developmental Psychology, 17, 275-288.
Vitaro, F., Larocque, D., Janosz, M., & Tremblay, R. E. (2001). Negative social experiences
and dropping out of school. Educational Psychology, 21, 401-415.
Vitaro, F., Tremblay, R. E., Gagnon, C., & Boivin, M. (1992). Peer rejection from
kindergarten to Grade 2: Outcomes, correlates, and prediction. Merrill-Palmer
Quarterly, 38, 382-400.
Vitaro, F., Tremblay, R. E., Gagnon, C., & Pelletier, D. (1994). Predictive accuracy of
behavioral and sociometric assessments of high-risk kindergarten children. Journal of
Clinical Child Psychology, 23, 272-282.
Waldrop, M. F., & Halverson, C. F., Jr. (1975). Intensive and extensive peer behavior:
Longitudinal and cross-sectional analyses. Child Development, 46, 19-26.
Walker, S., Berthelsen, D, & Irving, K. (2001). Temperament and peer acceptance in early
childhood: Sex and social status differences. Child Study Journal, 31, 177-192.
Ward, J. H. (1963). Hierarchical grouping to optimize an objective function. Journal of the
American Statistical Association, 58, 236-244.
Warden, D. & Mackinnon, S. (2003). Prosocial children, bullies and victims: An investigation
of their sociometric status, empathy and social problem-solving strategies. British
Journal of Developmental Psychology, 21, 367-385.
80
Wentzel, K. R. (2003). Sociometric status and adjustment in middle school: A longitudinal
study. Journal of Early Adolescence, 23, 5-28.
Wentzel, K. R., & Asher, S. R. (1995). The academic lives of neglected, rejected, popular,
and controversial children. Child Development, 66, 754-763.
Werner, N. E., & Crick, N. R. (2004). Maladaptive peer relationships and the development of
relational and physical aggression during middle childhood. Social Development, 13,
495-514.
Witkow, M. R., Bellmore, A. D. Nishina, A., Juvonen, J., & Graham, S. (2005). Mutual
antipathies during early adolescence: More than just aggression. International Journal
of Behavioral Development, 29, 209-218.
Woods, N. F., & Mitchell, E. S. (1997). Women’s images of midlife: Observations from the
Seattle midlife women’s health study. Health Care for Women International, 18, 439453.
Woodward, L. J., & Fergusson, D. M. (1999). Childhood peer relationship problems and
psychosocial adjustment in late adolescence. Journal of Abnormal Child Psychology,
27, 87-104.
Woodward, L. J., & Fergusson, D. M. (2000). Childhood peer relationship problems and later
risks of educational under-achievement and unemployment. Journal of Child
Psychology and Psychiatry, 41, 191-201.
Wu, X., Hart, C. H., Draper, T. W., & Olsen, J. A. (2001). Peer and teacher sociometrics for
preschool children: Cross-informant concordance, temporal stability, and reliability.
Merrill-Palmer Quarterly, 47, 416-443.
Yamaguchi, S., Kuhlman, D. M., & Sugimori, S. (1995). Personality correlates of allocentric
tendencies in individualist and collectivist cultures. Journal of Cross-Cultural
Psychology, 26, 658-672.
Young, L. L., & Cooper, D. H. (1944). Some factors associated with popularity. Journal of
Educational Psychology, 35, 513-535.
Zakriski, A. L., & Coie, J. D. (1996). A comparison of aggressive-rejected and nonaggressiverejected children’s interpretations of self-directed and other-directed rejection. Child
Development, 67, 1048-1070.
Zimmer-Gembeck, M. J., Geiger, T. C., & Crick, N. R. (2005). Relational and physical
aggression, prosocial behavior and peer relations: Gender moderation and bidirectional
associations. Journal of Early Adolescence, 25, 421-452.
81