청소년의 자살생각 예측모형 구축

다국어 초록 (Multilingual Abstract)

The Purpose of this study is to identify the effect of predictive factors on suicidal ideation and develop a prediction model that explains suicidal ideation among adolescents.
The hypothetical model was constructed on the basis of Bronfenbrenner's ecological system theory, the literature review and the pilot study.
This model was built of eight constructs. Four exogenous variables included in this model were parenting attitudes, family cohesion, self-esteem, problem-focused coping. Four endogenous variables were self-control, social attachment, depression, suicidal ideation. To test the effect influencing on suicidal ideation of parenting attitude type, two model was built. Affectional parenting attitude in model(1) and autonomic parenting attitude in model(2) is included.
Empirical data for testing the hypothetical model were collected by self- reported questionnaires from 1,156 high school students in Kyungsangbook-Do in March, 2008. The questionnaire items were reliable with Cronbach's Alpha ranging from .70-.91.
The analysis of the data was done with SPSS(V12.0) PC Program for descriptive statistics and correlation analysis. And PC LISREL Program(V8.12) was used to find the best fit model which assumes causal relationship among variables.

The results were as follows;
1. The results of covariance structure analysis
1) The overall measure of fit of the hypothetical model(1)(2) to the data was shown to be good(GFI=.98, AGFI=.91～.92, RMR-.01, NFI=.92～.91, CN= 492.06～536.34). But when we look at other type of measure like χ2/df(4.47～4.87), NNFI(.80～.82), and stem leaf plot, the hypothetical model(1)(2) needed modification.
2) Paths of the model(1)(2) were modified by considering both its theoretical implication and statistical significance of the parameter estimates. A revised model(1)(2) was developed by deleting one path each model, adding three paths from hypothetical model(1), adding five paths from hypothetical model(2).
3) Compared to the hypothetical model, the revised model has become parsimonious and had a much better fit to the data(χ2/df=2.27～2.18, GFI=.99, AGFI=.96, RMR=.01, NFI=.97, CN=1108.70～1126.05).
2. The results of hypothetical testing
1) Depression(β43=.24, t=7.55/β43=.24, t=7.68), self-esteem(γ43=-.14, t=-2.81/γ43=-.11, =-2.02) and social attachment(β42=-.06, t=-2.41/β42=-.07, t=-2.50) had significant direct effects on suicidal ideation. Probem-focused coping(β44=-.10, t=-3.40) had significant indirect effect on suicidal thoughts. Family cohesion(γ41=-.23, t=-2.61) and affetional parenting attitude(γ42=-.17, =-2.06) had significant direct effect on suicidal ideation in model(1), while self-control(β41=.05, t=2.26) had significant direct effect on suicidal ideation in model(2). Of these variables, depression(β43=.24, t=7.55/β43=.24, t=.68) had most significant direct effects on suicidal ideation. These predictive variables of suicidal ideation explained 48% of variance in model(1) and 50% of variance in model(2).
2) Self-esteem(γ33=-.77, t=-8.51/γ33=-.85, t=-8.26), problem-focused coping(γ34=-.47, t=-5.22/γ34=-.48, t=-5.09) and self-control(β31=-.16, t=-3.01/β31=-.14, t=-2.17) had significant direct effects on depression. Family cohesion(γ31=-.14, t=-0.91/γ31=-.02, t=-.40), parenting attitude(γ32=-.19, t=-1.30/γ32=-.06, t=-.67) and social attachment(β33=-.01, t=-0.13/β33=-.04, t=-0.66) had insignificant direct effects on suicidal ideation. Of these variables, Self-esteem(γ33=-.77, t=-8.51/γ33=-.85, t=-8.26) had most significant direct effects on suicidal ideation. These predictive variables of depression explained 40% of variance in two models.
3) Problem-focused coping(γ14=.70, t=9.82/γ14=.72, t=10.39), Family cohesion(γ11=.35, t=2.52/γ11=.13, t=2.39) had significant direct effects on self-control. Self-esteem(γ13=.10, t=1.44/γ13=.01, t=.19) had insignificant direct effects on self-control. Autonomic parenting attitude(γ12=.19, t=2.16) had significant direct effects on self-control in model(2), but had insignificant direct effects on self-control in model(1). Of these variables, Problem-focused coping(γ14=.70, t=9.82/γ14=.72, t=10.39) had most significant direct effects on suicidal ideation. These predictive variables of self-control explained 51% of variance in two models..
4) Family cohesion(γ21=.90, t=3.65/γ21=.40, t=3.34) and self-esteem(γ23=.70, t=5.10/γ23.=76, t=3.60) had significant direct effects on social attachment. affectional parenting attitude (γ22=.28, t=1.16/γ22=.22, t=1.24) had insignificant direct effects on social attachment. Self-control(β21=.45, t=4.44) had significant direct effects and problem-focused coping(β21=.45, t=4.44) had significant indirect and total effects on social attachment in model(2). But in model(2), problem-focused coping(β34=.34, t=2.60) had only significant total effects on social attachment. Of these variables, self-esteem(γ23=.70, t=5.10) in model(1) and Self-control(β21=.45, t=4.44) in model(2) had most significant direct effects on social attachment. These predictive variables of social attachment explained 59% of variance in model(1) and 63% of variance in model(2).
As a result of testing hypotheses, depression had the greatest direct effect on suicidal ideation and self-esteem had the greatest indirect and total effects. So, it should be emphasized that the level of suicidal ideation can be explained by means of personal emotional factors and personality traits. Through comparing effect of affectional parenting attitude on suicidal ideation with effect of autonomic parenting attitude, affectional parenting attitude only had a significant direct effect on suicidal ideation in model(1), and it seems that affectional parenting attitude is more important to prevent suicidal thoughts in adolescents.
On the result of testing hypotheses, it should be suggested that integrated and comprehensive intervention strategy including student, family, school and community be established to prevent suicidal ideation. Also it seems important to develop personal counseling and education program, family-focused program to strengthen family cohesion and affective parenting attitude, peer attachment program, and to make more protective social environment. And for preventing suicide and promoting adolescents' mental health, the roles of school nurses should be enlarged and strengthened through collaborating with parents, teacher, educational administrator, other mental health related personnel.

목차 (Table of Contents)

• Ⅰ. 서론 1
• 1. 연구의 필요성 1
• 2. 연구의 목적 4
• 3. 용어의 정의 5
• 4. 연구의 제한점 8
• Ⅱ. 문헌고찰 9
• 1. 청소년의 자살생각 9
• 2. 자살생각 관련변인 11
• Ⅲ. 연구의 개념적 기틀 및 가설적 모형 32
• 1. 개념적 기틀 32
• 2. 가설적 모형 37
• 3. 연구가설 40
• Ⅳ. 연구방법 43
• 1. 연구설계 43
• 2. 연구대상 및 표집 43
• 3. 연구도구 44
• 4. 자료수집 방법 및 절차 50
• 5. 자료분석 방법 51
• Ⅴ. 연구결과 52
• 1. 대상자의 일반적 특성 52
• 2. 연구변수의 서술적 통계 56
• 3. 연구변수의 상관관계 58
• 4. 가설적 모형의 검증 62
• 5. 가설적 모형의 수정 80
• 6. 연구가설의 검증 97
• Ⅵ. 논의 103
• 1. 모형의 검증결과 103
• 2. 간호학적 의의 115
• Ⅵ. 결론 및 제언 117
• 1. 요약 및 결론 117
• 2. 제언 120
• 참고문헌 122
• 부 록 144
• Abstract 155