본 연구는 청소년 자살생각과 관련된 지난 10년간의 선행연구를 바탕으로 자살생각을 억제하는 변인들의 효과크기를 메타분석을 통해 살펴보았다. 또한 메타회귀분석을 통해 조사시기, 조사대상, 연구대상의 표집 수에 따른 변화추이를 살펴봄으로 청소년 자살생각과 관련된 후속 연구의 방향을 제시하고, 청소년의 자살 예방을 위한 교육 및 상담현장에 실질적인 자료를 제시하고자 하였다. 이를 위해 청소년 자살생각을 종속변인으로 하고, 청소년의 자살생각 억제에 영향을 미치는 변인들을 독립변인으로 하는 학회지 및 석·박사논문 중 상관관계를 통해 독립변인과 종속변인의 관계를 설명하여 효과크기 변환이 가능한 논문 69편을 연구대상으로 선정하였다. 연구결과는 첫째, 억제변인군의 효과크기는 심리 변인군과 가족환경 변인군의 효과 크기가 통계적으로 유의미한 것으로 나타났고, 두 변인군 모두 적은 크기의 효과를 가진 것으로 나타났다. 둘째, 심리 억제변인군의 하위변인 중 사회심리적 지지, 자아존중감, 자아탄력성이 중간 효과크기 갖는 것으로 나타났고 가족환경 억제변인군의 하위변인은 모두 중간수준의 효과크기를 가진 것으로 나타났다. 셋째, 조사대상, 조사년도, 표집 수에 대한 메타회귀분석 결과는 모두 통계적으로 유의미한 차이가 없는 것으로 나타났다. 그러나 변인들에 대한 상관분석 결과, 자아탄력성과 교사지지, 친구와의 관계에서 변화추이가 발견되었다.

The purpose of this study was to conduct a systematic review of the previous study on adolescents` suicidal ideation. From 2000 to 2010, ninety-one articles published on academic journals and 107 masters` thesis and doctoral dissertation were collected. Of theses studies, 69 articles which had a report primary empirical data met the criteria for inclusion in the meta-analysis. The results of this study were as follows: First, the three sets of protective factors related to adolescents` suicidal ideation were included with psychological, family environmental, and school environment. The most influential factors among three sets was psychological set followed by family environment and the effect size of these factors was moderate. However, the effect size of school environment was non significant. Second, the most influential protective variables related to adolescents` suicidal ideation among the set of psychological factor was social support followed by self-esteem and resiliency. The effect size of these variables was moderate. Third, the most influential protective variables related to adolescents` suicidal ideation among the set of family environment factor was family resiliency followed by communication with parents, family harmony, and parenting behaviors. The effect size of these variables was moderate. Fourth,, although the effect size of school environment was non significant, the effect size of sub variables such as academic resiliency, academic achievement and teacher support was significant. Fifth, the results of meta-regression analysis revealed that the moderators of the sample subjects, years(the former period and the latter period of the 2000 year), and the sample size had non-significant effect on overall effect size. In conclusion, the results of the current study can be applied to educational and counseling program for the prevention of adolescents` suicide. More especially, the future of research is needed to investigate how the different types of measurement impact on the variables related to adolescents` suicidal ideation.