Children's Peer Relationships: Longitudinal
Prediction of Internalizing and Externalizing
Problems from Middle to Late Childhood
Shelley Hymel, Kenneth H. Rubin, Lynda Rowden, and
Lucy LeMare
University of Waterloo
HYMEL, SHELLEY; RUBIN, KENNETH H.; ROWDEN, LYNDA; AND LEMARE, LUCY. Children's Peer
Relationships: Longitudinal Prediction of Internalizing and Externalizing Problems from Middle to
Late Childhood. CHILD DEVELOPMENT, 1990,61,2004-2021. In this longitudinal study, the predictive
relations between social difficulties in early childhood (grade 2) and subsequent internalizing as
well as externalizing problems in middle childhood (grade 5) were examined. Of particular interest
was whether early indices of social isolation would predict intemahzing problems 3 years later. A
longitudinal sample of 87 children were assessed in both grades 2 and 5 on a variety of measures,
including sociometric ratings, peer assessments of aggression and isolation, and self-appraisals of
social competence. In the second grade, observations of isolated and aggressive behavior were
made, as well, and teacher ratings of internalizing and externalizing difficulties were obtained. In
the fifth grade, teacher ratings of shy-anxious and acting-out behavior and self-reports of loneliness
and self-esteem were collected. Consistent with previous research, results demonstrated predictive
links between early peer rejection (unpopularity) and aggression and subsequent externalizing
difficulties. Internalizing problems in middle childhood were significantly related to early social
difficulties, particularly those of an internalizing sort, including poor peer acceptance, social isola-
tion, and perceptions of social incompetence. Social isolation, then, may indeed be a risk factor in
early development.
Although previous research has indi- realization may lead to feelings of loneliness
cated that early peer difficulties are predictive (Asher, Hymel, & Renshaw, 1984; Asher, Park-
of later maladjustment (see Kohlberg, La hurst, Hymel, & Williams, 1990), depression
Crosse, & Ricks, 1972; Parker & Asher, 1987, (Strauss, Forehand, Frame, & Smith, 1984;
for reviews), researchers have tended to em- Vosk, Forehand, Parker, & Richard, 1982),
phasize the relations between early peer re- and to negative self-perceptioiis of social com-
jection and/or aggression and subsequent ex- petence and self-esteem (Hymel, 1983), all
temalizing outcomes such as aggression, known concurrent correlates of peer rejection
hostility, acting-out behavior, or delinquency. and unpopularity (see Rubin & Lollis, 1988,
Virtually no data exist concerning the possi- for an extended discussion of the conceptual
bility that early social difficulties may also link between rejection and internalizing dis-
predict internalizing problems. According to orders). The purpose of the present study,
Achenbach (1982; Achenbach & Edelbrock, then, was to examine, prospectively, the asso-
1981), self-related difficulties such as fear- ciation between early social difficulties and
fulness, anxiety, and social withdrawal may indices of both externalizing and internalizing
be labeled "internalizing problems." It seems problems in later childhood,
entirely plausible that a child who is rejected
by peers may come, eventually, to realize that Of particular interest was whether mea-
he or she is disliked or become increasingly sures of overall sociometric acceptance or
aware of his or her poor peer status. Such a more specific indices of internalizing or exter-
Portions of this research were presented at the biennial meeting of the Society for Research in
Child Development, April 1987, Baltimore, MD, and at the first annual meeting of the Society for
Research in Child and Adolescent Psychopathology, February 1989, Miami, FL. The research
reported herein was supported by a grant to authors Rubin and Hymel fi-om Health and Welfare
Canada and by a Killam Research Fellowship to author Rubin from the Canada Council. We are
grateful to the teachers and children in the Waterloo County School Board for their cooperation in
this project. We also thank Anne Emptage and Laurie Addis who coordinated data gathering aspects
of this study. Requests for reprints should be addressed to either author Hymel or Rubin, Depart-
ment of Psychology, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada.
[Child Development, 1990, 61, 2004-2021. © 1990 by the Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/90/6106-0029$01.00]