Objective To develop a School-age Children's Eating Behavior Scale and to provide a tool to assess eating behavior of school-age children. Methods In January 2018, the research team formed a pool of 168 items through the study of relevant literature at home and abroad. Based on the results of face-to-face interviews with 50 parents of children aged 6-12 years and combining the opinions of 5 pediatric clinical experts, the researcher deleted, merged and revised the items to form the first draft of the scale with 84 items. In March, 2018, 363 school-age children were surveyed with the first draft of the scale. Items of the scale were screened and structure of the scale was analyzed comprehensively with variation coefficient method, exploratory factor analysis method, etc., and finally a scale with 48 items was formed. In May 2018, a trial scale was used to investigate 700 parents of children aged 6-12 randomly selected from 2 primary schools in Xi'an. The reliability, validity and differentiation of the scale were verified by factor analysis and correlation analysis. Results Exploratory factor analysis indicated that the scale had 8 dimensions. They included ill-balanced eating, food responsiveness, satiety responsiveness, bad eating habit, external factor eating, eating restriction, food enjoyment, craving for junk food. The cumulative variance contribution rate was 52.16％. All fitting indexes of confirmatory factor analysis met the requirements of measurement [GFI=0.90, AGFI =0.89, NNFI=0.92, CFI=0.93]. The results of principal component analysis of each dimension showed that there was only one factor with the characteristic value greater than 1 in each dimension, the variance contribution rate ranged 45％-71％, and the load range of each item was greater than 0.40. The correlation coefficient between the scores of each dimension was 0.08-0.48, the correlation coefficient between the scores of each dimension and the total score of the scale was 0.51-0.72, the Cronbach's α coefficient of the scale was 0.89, Guttman's split-half reliability was 0.71, and the test-retest reliability coefficient was 0.73. There were differences in scores of different dimensions among different groups in terms of education, family structure and BMI (all P<0.05). Conclusion The School-age Children's Eating Behavior Scale with 8 dimensions and 48 items has good reliability, validity and differentiation, which can reflect the situation and characteristics of the eating behavior of school-age children in China comprehensively. It can be used for measuring and evaluating the eating behavior of school-age children in the future.%目的 编制可用于我国学龄期儿童饮食行为测量的量表, 为学龄期儿童饮食行为评估提供工具.方法 2018年1月, 研究小组在查阅国内外相关文献的基础上, 形成了包含168个问题的条目池.由研究人员对50名6～12岁儿童家长进行面对面的逐条访谈, 依据访谈结果, 并结合5名儿科临床专家意见, 对条目进行删除、合并、修订后形成包含84个条目的量表初稿;2018年3月, 采用量表初稿对363名学龄期儿童进行调查, 综合运用高低分法、变异系数法、探索性因子分析法等进行条目筛选和量表结构分析, 形成包含48个条目的"儿童饮食行为试用量表";2018年5月, 采用试用量表在西安市2所小学随机抽取700名6～12岁儿童家长进行调查, 运用因子分析和相关分析等方法对量表的信效度及区分度进行验证.结果 探索性因子分析可提取出挑食、食物响应、过饱响应、不良进食习惯、外因性进食、限制饮食、食物喜好、垃圾食品渴望8个维度, 累计方差贡献率达52.16％;验证性因子分析各拟合指标[拟合优度指数 (GFI) =0.90, 调整的拟合优度指数 (AGFI) =0.89, 非规范拟合指数 (NNFI) =0.92, 比较拟合指数 (CFI) =0.93]等均达到测量学要求.各个维度主成分分析结果显示:每个维度特征值大于1的因子数都仅有1个, 方差贡献率在45％～71％, 各条目载荷范围都大于0.40.各维度得分间的相关系数为0.08～0.48, 各维度得分与问卷总分的相关系数为0.51～0.72, 量表的Cronbach'sα系数为0.89、Guttman分半信度为0.71、重测信度系数为0.73.不同学历照护人、不同家庭结构、不同身体质量指数 (BMI) 儿童间各维度得分比较差异均有统计学意义 (均P<0.05) .结论 8个维度48个条目的学龄期儿童饮食行为量表有良好的信效度和区分度, 均能比较全面地反映出我国学龄期儿童饮食行为的情况和特点, 可用于将来学龄期儿童饮食行为的测量及评估.
展开▼
