斯尔文绘画测验中文修订版的信效度分析

Reliability and Validity of Revised Chinese Version of Silver Drawing Test

目的:修订中文版斯尔文绘画测验(Silver Drawing Test,SDT),并分析其认知与情绪分测验的信度与效度。方法:本研究为横断面研究。在西安市一所公立小学与一所幼儿园的各年级(幼儿园中班-小学六年级)中,以分层随机抽样法共选取122名年龄为4岁6个月至12岁11个月的儿童,施测SDT中文修订版。其中,47名儿童间隔45天后参加SDT重测,计算2次测验结果的Pearson积差相关系数,以检验SDT的重测信度;三名评分者随机选取19名儿童完成的SDT进行计分,分析三名评分者所计分数间的Kendall和谐系数,以检验评分者信度;随机选取五年级一个班39名儿童参加瑞文标准推理测验青少版(RSPM),分析SDT认知分测验与RSPM各分测验成绩、总分的标准分数之间的相关系数,以检验SDT与RSPM间的效标关联效度;83名儿童分别根据学业与行为表现分为两组,采用方差分析法分析高、低学业成绩组及有、无行为问题组在SDT测验成绩上的差异,以检验SDT的学业与行为效标效度。结果:SDT中文修订版各分测验的重测信度在0.64～0.85之间,评分者信度在0.80～0.98之间;SDT认知分测验与RSPM中知觉辨别分测验的关联效度为0.68(P<0.01);低学业成绩组SDT想象画分测验与认知总分均低于高学业成绩组[(5.75±2.8)vs.(8.31±3.73)、(19.13±9.93)vs.(24.73±8.6),P=0.003、0.012];无行为问题组情绪投射测验成绩低于有行为问题组[(2.90±0.80)vs.(3.46±0.78),P=0.005]。结论:斯尔文绘画测验修订为中文版后,其认知与情绪分测验仍具有较好的信度与效度,达到心理测量学标准,可应用于国内儿童心理评估及其他相关领域的研究。

Objective： To assess the reliability and validity of the revised Chinese version of Silver Drawing Test （ SDT） . Methods： In this cross-sectional study, 122 children aged 4 years and 6 months to 12 years and 11 months were selected by stratified random sampling and tested with the SDT. Among them, 47 children were randomly selected to examine test-retest reliability, 19 children were randomly selected to examine scorer reliability, 39 children in one class of 5 grade were randomly selected for the criterion-related validity with the perception identify subtest of Raven＇s Standard Progressive Matrices （ RSPM ） , and 83 children were selected to examine validity by analyzing differences between high-and-low academic performance groups, and differences between with-and-without behavior problems groups. Results： The test-retest reliability was 0. 64 - 0. 85. The scorer reliability was 0. 80 - 0. 98. The criterion-related validity with the perception identify subtest of RSPM was 0. 68 （ P 〈 0. 01 ） . Imagination sub-score and cognitive subscore in low academic performance group were significantly lower than that in high academic group [ （ 5.75 ± 2. 8 ） vs （ 8. 31 ± 3.73 ）, （ 19. 13 ± 9. 93 ） vs. （24. 73 ± 8. 6 ） ; P = 0. 003, 0. 012 ] ; Emotional projective sub-score in group with behavior problems was significantly lower than that in group without behavior problems [ （ 2.90 ± 0. 80 ） vs. 13.46 ± 0. 78 ）, P = 0. 005 ] . Conclusion： The cognition and emotion subtest of the revised Chinese version of SDT has good reliability and validity. It can be applied in domestic children＇s psychological assessment, and other relevant research fields.