目的：探索综合测量儿童健康认知状况的工具，编制儿童健康认知问卷，并检验其信效度。方法以世界卫生组织（WHO）界定健康理论为依据，2013年9—11月，采用开放式访谈（n ＝13）方法收集题项资料，选取320例儿童进行预测；通过项目分析（临界比法、题总相关法、共同性和因素负荷量）、探索性因子分析（主成分分析法、极大方差正交旋转法和事先决定准则法）和信度分析筛选后得到正式问卷，对上海市10所小学1336例儿童施测以进行问卷的验证性因子分析。正式调查结束后1个月，运用正式问卷对参与调查的140例儿童重新调查，检验问卷的稳定性。结果开放式访谈共建立儿童健康认知问卷备选条目库60条，通过项目分析剩余46个条目进行探索性因子分析，适合度检验 KMO 值为0.740，Bartlett’s 球形检验结果显著（χ2＝3574.707，P ﹤0.001），说明适合进行探索性因子分析。探索性因子分析结果显示，最后保留21个题项，3个维度，分别为身体健康认知、心理健康认知和社会适应健康认知，因子载荷数为0.402~0.763，可解释儿童健康认知分数方差变异的58.47％。信度检验：问卷总条目Cronbach’s α系数为0.806，身体健康认知、心理健康认知、社会适应健康认知 Cronbach’s α系数分别为0.829、0.778、0.734。效度检验：该问卷效度检验三因素模型中χ2/ df ＝1.39，拟合优度指数（GFI）＝0.92，调整的拟合优度指数（AGFI）＝0.90、比较拟合指数（CFI）＝0.93、递增拟合指数（IFI）＝0.93、Tucker - lewis 指数（TLI）＝0.92，均﹥0.90；且简效拟合优度指数（PGFI）、简约规范拟合指数（PNFI）分别为0.74、0.70，高于0.50的标准，近似误差平方根（RMSEA）为0.04，问卷模型与数据的整体拟合效果较好。问卷再测信度：问卷总条目得分的 Pearson 相关系数为0.78，身体健康认知、心理健康认知、社会适应健康认知得分的 Pearson 相关系数为0.72、0.65、0.74。结论编制的儿童健康认知问卷具有良好的信效度，可用于测量儿童健康认知水平。但对低龄、经济或教育欠发达地区是否适用，仍需要后续研究验证。%Objective To investigate a tool for comprehensive assessment of children’s health cognition,compile a children’s health cognition questionnaire,and conduct tests on its reliability and validity. Methods Based on the health theory defined by WHO,we collected item data by open - ended interviews( n = 13) from September to November in 2013. We selected 320 children to undertake prediction and completed the formal questionnaire by item analysis( critical ratio method, subject total correlation method,intercommunity and factor loading),exploratory factor analysis(principal component analysis, varimax rotation method,predetermined criterion) and selection by reliability analysis. Questionnaire survey was conducted on 1 336 children in 10 primary schools of Shanghai to undertake confirmatory factor analysis. After formal investigation for one month,the formal questionnaire was used to conduct survey on 140 children again to test the stability of the questionnaire. Results By open - ended interviews, 60 alternative items of children’ s health cognition questionnaire were established. Exploratory factor analysis was conducted on the rest 46 items by item analysis. The KMO value of the test of goodness of fit was 0. 740,and Bartlett’s sphericity test result was significant( χ2 = 3 574. 707,P ﹤ 0. 001);the results showed that these items were suitable for exploratory factor analysis. The exploratory factor analysis showed that 21 items and 3 dimensions were retained,and the 3 dimensions were physical health recognition,mental health cognition,and social adaptability health cognition. The number of factor loading was 0. 402 - 0. 763,which could explain the 58. 47% of variance variation of children’s health cognition score. In the test of reliability,the Cronbach’s α coefficient of all the items of the questionnaire was 0. 806,and the Cronbach’s α coefficients of physical health cognition,mental health cognition,and social adaptability health cognition were 0. 829,0. 778 and 0. 734 respectively. In validity test,the models of χ2 / df = 1. 39,GFI = 0. 92,AGFI = 0. 90,CFI = 0. 93, IFI = 0. 93,TLI = 0. 92 were all ﹥ 0. 90,and PGFI and PNFI were 0. 74 and 0. 70 respectively,higher than the standard of 0. 50,and RMSEA was 0. 04. The overall fitting effect of the questionnaire model and data was good. For the test - retest reliability of the questionnaire,the Pearson correlation of all the items of the questionnaire was 0. 78,and the correlation coefficients of physical health cognition, mental health cognition and social adaptability health cognition were 0. 72,0. 65 and 0. 74 respectively. Conclusion The children’s health cognition questionnaire has good reliability and validity,and it can be used to evaluate children’s health cognition level,while further research and verification are needed to determine if it is fit for young children and regions with underdeveloped economy or education.
展开▼
