【摘要】 目的研究中文SCL-90量表在农村五保老人使用的信度和效度。方法采用多阶段分层整群抽样法,用中文SCL-90量表对湖北省农村60岁及以上的546名集中供养五保老人(A组)、139名分散供养五保老人(B组)进行面对面访谈,并计算各分量表及总表的评分。采用信度分析和因子分析等方法对量表信度和效度进行评价。结果除敌对分量表外其余各分量表Cronbach’a系数均大于0·65;2组各分量表与总量表分的相关系数r均大于0·65;2组部分分量表之间的部分相关系数大于其分量表的α值;2组因子分析显示大部分分量表提取出两个或两个以上的公因子。结论除敌对因子分量表外,中文SCL-90其余分量表内在信度较好;各分量表结构效度较差;总量表的内容效度良好,结构效度欠佳。它基本适于评价农村五保老人的心理健康。更多还原

【Abstract】 Objective To explore the reliability and validity of SCL-90 used in the rural ′wubao′ senior citizens. Methods 546 gregarious ‘wubao’ senior citizens and 139 scattered wubao senior citizens were interviewed with multi-stage stratified cluster sampling to calculate the scores of divided measuring tables and total table. The reliability and validity of SCL-90 were evaluated by reliability analysis and factor analysis. Results In addition to hostile scale, the rest of the divided scales’ Cronbach’a coefficients were greater than 0.65; the correlation coefficients of the divided scales and the total scale were greater than 0.65 in the 2 Group; the part correlation coefficients of the part divided scales were greater than their Cronbach’a coefficients in the 2 Group; factor analysis showed that two or more public factors were extracted in the majority divided scales of the 2 Group.Conclusion In addition to hostile scale, the rest of the divided scales have better reliability; The divided scales have bad construction validity; The total scale has better content reliability and bad construction validity. It is basically suited to evaluate mental health of the rural ′wubao′ senior citizens.更多还原