中文版物质使用风险评估量表初步修订及信、效度评价

Primary revision of Substance Use Risk Profile Scale-Chinese Version and its reliability and validity evaluation

目的初步修订中文版物质使用风险评估量表(SURPS),并评价其信度和效度,为该量表在中国的推广使用提供科学依据。方法通过双向翻译获得中文版SURPS初版量表,结合中国小学生特点初步对该初版量表的理论结构进行修订,并采用项目分析、探索性因子分子、验证性因子分析和信效度检验方法对广西壮族自治区百色市2014年3月—2015年6月抽取的3所小学共948名小学生的物质使用风险进行分析,验证该量表理论结构的合理性。结果修订后的中文版SURPS模型与原量表因子拟合指数基本达到可接受的标准,探索性因子分析结果显示,中文版SURPS量表4个公因子的累计方差贡献率为54.16%,各条目所属因子结构与原量表基本一致;验证性因子分析结果显示,χ~2/df为5.214,近似误差均方根为0.063,拟合优度指数、调整拟合优度指数、规范拟合指数、增量拟合指数分别为0.889、0.862、0.898、0.893;修订后的中文版SURPS量表具有较好的项目区分度,各条目得分与其所属维度得分的相关系数均在0.667~0.818(均P<0.001);量表总体Cronbach'sα系数和重测信度分别为0.801和0.776,感觉寻求、绝望、焦虑敏感、冲动4个维度的Cronbach'sα系数为0.698~0.811,重测信度为0.601~0.754。结论修订的中文版SURPS量表在中国小学生中的信度和效度指标均符合心理测量学的要求,可适用于我国青少年物质使用相关个体特征因素的风险评估。

Objective To revise and evaluate the reliability and validity of Substance Use Risk Profile Scale （SURPS）-Chinese Version and to provide evidences for its application in China. Methods The original SURPS was first translated from English into Chinese and then back translated into English. The theoretical construct of SURPS was primarily revised based on the characteristics of Chinese pupils. The revised SURPS-Chinese Version was administered among 948 grade 4 to 6 pupils recruited with random cluster sampling from 3 primary schools in Balse city of Guangxi Zhuang Autonomous Region between March 2014 and June 2015 to test theoretical construct rationality and to evaluate the reliability and validity of the revised SURPS. Correlation analysis, exploratory and confrrrnatory factor analysis were adopted in the evaluation. Results The factor construction for the items of the revised SURPS-Chinese Version was ba- sically the same as the original scale. The results of exploratory factor analysis showed that 54. 16% of the total variation could be explained by 4 common factors. The confirmatory factor analyses showed following results ：X2/ df = 5. 214, goodness-of-fit index （GFI） = 0. 8893, adjusted goodness-of-fit index （AGFI） = 0. 862 4, normed fit index （NFI） = 0. 8984, incremental fit index （IFI） = 0. 8927, root mean square error of approximation （RMSEA） = 0. 063, demonstrating good fitting of the model. The Pearson correlation coefficients between each item and dimensions were from 0. 667 to 0. 818 （P 〈 0. 001 ）. The total internal consistency Cronbach＇s alpha was 0. 801 ; the total test-retest correlation coefficient was 0. 776. The internal consistency Cronbach＇s alphas for the four dimensions were between 0. 698 to 0. 811 ,and the test-retest correlation coefficients were between 0. 601 to 0. 754. Conclusion The reliability and validity of the revised SURPS-Chinese Version for Chinese pupils meet the requirement of psychological test and could be used in individual risk assessment on substance abuse among Chinese adolescents.