CAJ | 학술논문

目的：探索适用于农村地区的学校缺勤监测的最佳预警模型及其参数，为实现自动预警提供技术支持。方法应用EWMA和CUSUM（C1、C2、C3）模型对湖北省沙洋县某小学在2012年4月1日至2013年1月31日期间，因病缺勤的数据进行回顾性预警分析，比较各模型在不同参数组合下的约登指数，从而优选出最佳预警模型及其参数。结果各模型最优参数下的预警功效：EWMA（λ＝0.25，c＝2.0）的约登指数为0.82，C1（k＝0.5，c＝2.0）的约登指数为0.95，C2（k＝0.5，c＝3.0）的约登指数为0.98，C3（k＝0.5，c＝5.0）的约登指数为0.97，其中C2（k＝0.5，c＝3.0）的预警功效最佳。结论4种模型中，CUSUM （C2）模型的预警功效最优。
목적：탐색괄용우농촌지구적학교결근감측적최가예경모형급기삼수，위실현자동예경제공기술지지。방법응용EWMA화CUSUM（C1、C2、C3）모형대호북성사양현모소학재2012년4월1일지2013년1월31일기간，인병결근적수거진행회고성예경분석，비교각모형재불동삼수조합하적약등지수，종이우선출최가예경모형급기삼수。결과각모형최우삼수하적예경공효：EWMA（λ＝0.25，c＝2.0）적약등지수위0.82，C1（k＝0.5，c＝2.0）적약등지수위0.95，C2（k＝0.5，c＝3.0）적약등지수위0.98，C3（k＝0.5，c＝5.0）적약등지수위0.97，기중C2（k＝0.5，c＝3.0）적예경공효최가。결론4충모형중，CUSUM （C2）모형적예경공효최우。
Objective To explore the optimal early warning model and its related parameters for school absenteeism surveil‐lance in rural areas so as to provide technical support for developing an Auto Warning System (AWS).Methods The data of ab‐sentees due to illness in an elementary school in Shayang county of Hubei province from April 1 ,2012 to January 31 ,2013 were retrospectively analyzed by using exponentially weighted moving average (EWMA)model and cumulative sums(CUSUM ,C1 , C2 ,C3)model to identify an optimal early warning model.Youden’s index was calculated in each model with different combina‐tions of parameters and compared.Results The warning efficacy of each model with the optimal parameters was as follows :the Youden’s index of EWMA(λ=0.25 ,c=2.0)was 0.82 ,that of C1(k=0.5 ,c=2.0)was 0.95 ,that of C2(k=0.5 ,c=3.0)was 0.98 ,and that of C3(k=0.5 ,c= 5.0)was 0.97 ,which suggested that the C2(k= 0.5 ,c= 3.0)model had the highest effica‐cy.Conclusion Among the 4 models applied ,CUSUM C2 model is optimal for early warning.