A meta-analytic study of predictors for loneliness during adolescence

Nurs Res. Sep-Oct 2006;55(5):308-15. doi: 10.1097/00006199-200609000-00003.


Background: Extensive research has been completed to investigate loneliness in adolescents over the past several decades; identification of key predictors for loneliness was needed using meta-analytic techniques.

Objectives: To identify predictors for loneliness in adolescents through a comprehensive review of the literature and to use quantitative meta-analysis to determine the magnitude of the relationships between each predictor and loneliness.

Methods: The literature reviewed included 242 studies published or unpublished between 1980 and 2004, of which 95 met the inclusion criteria. Eleven key predictors for loneliness were identified in the 95 studies. A meta-analysis was conducted on each of the 11 predictors in relation to loneliness.

Results: Four predictors (gender, depression, shyness, and self-esteem) had large effect sizes, four predictors (social support, social anxiety, maternal expressiveness, and paternal expressiveness) had large medium to medium effect sizes, two predictors (stress and self-disclosure) had low effect sizes, and one predictor (age) had a very low effect size.

Discussion: Theories of loneliness served as a framework for interpreting the findings of the meta-analysis. The most powerful predictors can be used in intervention studies aimed at reducing loneliness in adolescents.

Publication types

  • Meta-Analysis
  • Review

MeSH terms

  • Adolescent
  • Child
  • Female
  • Humans
  • Loneliness*
  • Male
  • Psychology, Adolescent*
  • Risk Factors