오경자, 김민영, 김영아, 하은혜, 이진 (2013, June). 10-year changes in behavior problems of Korean adolescents. Poster presented at International Society for Research in Child and Adolescent Psychopathology, Leuven, Belgium.
Time trend studies on children's behavior problems can enhance our understanding of the influence of social changes on children's mental health problems. Recent time trend studies of child behavior problems conducted in Western societies reported varying results with some reporting increases (e.g., Achenbach & Howell, 1993), while others reporting a decreasing trend (e.g., Achenbach, Dumenci, & Rescorla, 2003). Korea has undergone rapid social changes in recent decades and there is a widespread concern that children's mental health might have been affected by such social changes. Recent social statistics such as high suicide rate, steep increase in substance abuse and reported incidents of school violence support the general concern, but very little empirical data are available which directly confirm or disconfirm the concern. The primary aim of the present study is to compare Child Behavior Checklist (CBCL) and Youth Self Report (YSR) data of a large community sample collected in 1997 and in 2007 in order to see whether the behavioral and emotional problems of Korean youths have changed over the 10 year period.
The finding that there was a significant decline across all CBCL scales from 1997 and 2007 might be interpreted to mean that contrary to the widespread concern, Korean adolescents' behavior problems might have actually decreased over the 10-year period. However, in YSR (i.e. self report data), significant 1997-2007 differences were largely limited to internalizing problems and the effect sizes were relatively small compared to the effect sizes in CBCL (i.e. parent report data). Thus, it is possible that the decline in CBCL scores from 1997-2007 might also reflect changes in the threshold for parental perceptions of deviance, or parental willingness to acknowledge their children's problems. The ME-CFA finding that there was a notable difference in intercepts across the two samples for CBCL also supports the above possibility. Need for multi-informant data have been emphasized in recent studies in psychopathology, and the present study highlights the value of multi-informant data in interpretation of time-trend data.