Comparisons of adopted and nonadopted adolescents in a large, nationally representative sample

Brent C. Miller, Xitao Fan, Mathew Christensen, Harold D Grotevant, Manfred Van Dulmen

Research output: Contribution to journalArticlepeer-review

113 Scopus citations

Abstract

There are conflicting findings about whether adopted children have more psychological and behavioral problems than nonadoptees. Research results are discrepant partly because many previous studies were based on small clinical samples or on samples biased by self-selection. A nationally representative school survey (Add Health) was used to compare adopted (n = 1, 587) and nonadopted adolescents (total N = 87, 165) across a wide variety of measures. Standardized mean differences show that adopted adolescents are at higher risk in all of the domains examined, including school achievement and problems, substance use, psychological well-being, physical health, fighting, and lying to parents. Demographic and background variable breakdowns show that the effect sizes for differences between adopted and nonadopted adolescents were larger for males, younger or older adolescents, Hispanics or Asians, and adolescents living in group homes or with parents of low education. Distributional analyses revealed approximately a 1:1 ratio of adopted to nonadopted adolescents in the middle ranges of the outcome variables but a ratio of 3:1 or greater near the tails of the distributions. These data clearly show that more adopted adolescents have problems of various kinds than their nonadopted peers; effect sizes were small to moderate based on mean differences, but comparisons of distributions suggest much larger proportions of adopted than nonadopted adolescents at the extremes of salient outcome variables.

Original languageEnglish (US)
Pages (from-to)1458-1473
Number of pages16
JournalChild development
Volume71
Issue number5
DOIs
StatePublished - 2000

Fingerprint Dive into the research topics of 'Comparisons of adopted and nonadopted adolescents in a large, nationally representative sample'. Together they form a unique fingerprint.

Cite this