目的 构建医学研究生满意度量表并对其信度和效度进行评估分析。方法 2017年9月,对广西医科大学的2 961名在读全日制研究生进行满意度问卷调查。利用探索性因子分析方法构建了满意度量表的维度并评价效度,利用验证性因子分析方法验证量表的适配性。结果 满意度量表共包含31个条目,总体克朗巴赫系数为0.972;经主成分法提取出5个公因子,累计贡献率达到81.09%。验证性因子分析结果显示,调查量表修正后,卡方自由度比χ2/df为4.451 (95%CI=4.321~4.558),近似误差均方根(root mean square error of approximation,RMSEA)为0.062 (95%CI=0.058~0.069),其他增值适配度指标均大于0.9;各条目的标准化载荷均超过0.71,且均具有统计学意义 (t>1.96, P<0.05)。结论 医学研究生满意度量表具有良好的信度和效度,且适用性良好,可以作为医学研究生满意度的调查工具并推广使用。
Objective To construct the satisfaction scale for medical graduates and evaluate its reliability and validity. Methods Totally 2 961 full-time postgraduates from Guangxi Medical University were investigated in September 2017. Exploratory factor analysis was used to construct dimensions and evaluate the validity. Confirmatory factor analysis was used to verify the scale's fitness. Results A total of 31 items were included in the scale. The overall Cronbach's α coefficient was 0.972. Five common factors were extracted by the principal components method and the cumulative contribution rate reached 81.09%. The results of confirmatory factor analysis showed that χ2/df was 4.451 (95%CI=4.321~4.558), root mean square error of approximation (RMSEA) was 0.062 (95%CI=0.058~0.069),other value-added fitness indicators were all over 0.9;the standardized load of each item was over 0.71 and had statistical significance(t>1.96, P<0.05). Conclusions The scale has good reliability,validity and applicability. It can be used as a tool to evaluate satisfaction for medical graduates.
[1] 周文辉, 黄欢, 付鸿飞, 等. 2017年我国研究生满意度调查[J].学位与研究生教育,2017(9):41-47. DOI: 10.16750/j.adge.2017.09.008.
[2]Raykov T, Dimitrov DM, Marcoulides GA, et al. On the connections between item response theory and classical test theory: a note on true score evaluation for polytomous items via item response modeling[J]. Educ Psychol Meas, 2019,79(6):1198-1209. DOI: 10.1177/0013164417745949.
[3]李跃平, 黄子杰. 验证性因子分析在量表结构效度考核中作用[J].中国公共卫生,2007,23(10):1198-1199. DOI: 10.3321/j.issn:1001-0580.2007.10.024.
[4]Barrett P. Structural equation modelling: adjudging model fit[J]. Personality & Individual Differences, 2007, 42(5):815-824.DOI:10.1016/j.paid.2006.09.018.
[5]吴明隆. 结构方程模型:AMOS的操作与应用[M]. 重庆:重庆大学出版社, 2010:24-27.
[6]戴明叶.基于结构方程模型的顾客满意度研究——以湖南省博物馆为例[D].长沙:湖南师范大学, 2014.
[7]Kline RB. Principles and practice of structural equation Modeling[M]. New York: Guilford Press, 1998:133-145.
[8]Ecklund-Johnson E, Miller SA, Sweet JJ. Confirmatory factor analysis of the behavioral dyscontrol scale in a mixed clinical sample[J]. Clin Neuropsychol, 2004,18(3):395-410. DOI: 10.1080/1385404049052415.
[9]Chibnall JT, Tait RC. Confirmatory factor analysis of the Pain Catastrophizing Scale in African American and Caucasian Workers' Compensation claimants with low back injuries[J]. Pain, 2005,113(3):369-375. DOI: 10.1016/j.pain.2004.11.016.
[10]薛二勇, 刘淼, 李健. 我国教育公平发展政策变迁的历程、特征与趋势[J].教育研究,2019(5):142-150.