分组对应分析在流行病学研究中的应用

The Use of Subgroup-Correspondence Analysis in Epidemiology Study

目的了解恶性肿瘤的地区分布和肿瘤类型分布特征,探讨对应分析时前2个因子累计贡献率比较低时的解决办法。方法经过因子分析或聚类分析对列变量或行变量进行合理分组,应用分组对应分析对该县恶性肿瘤死亡资料进行分析。结果2组对应分析结果表明,前2个因子的累积贡献率分别为77.4%和85.4%,可以基本表达原始资料的信息,根据第一、二因子负荷系数绘制对应分析平面负荷图,可以清楚看出恶性肿瘤死亡率的聚集性及其高发地与低发地的分布。结论应用分组对应分析可提高前2个因子累计贡献率,充分表达和利用资料的信息,正确分析行因素与列因素之关系达到研究目的。

Objective Studying the area and types distribution of malignant tumor mortality. Approaching a method to solve the question when the first 2 factors＇ cumulative percentage are lower. Methods Column variable or row variable are divided into groups by factor analysis or cluster analysis, and using subgroup correspondence analysis to study the data of cancer mortality. Results 2 analytic results indicate that the first 2 factors＇ cumulative percentage are respectively 77.4 % and 85.4 %, which can express the information of initial data. Then drawing a graph of factor loading by using the first and second factor loading. The species and area distribution of malignant tumors can be clearly obtained in the loading factor chart of COrrespondence analysis. Conclusion Subgroup-correspondence analysis can raise the first 2 factors＇ cumulative percentage, and can analyse the relationship between row and column variables for 2-dimention data, accordingly achieving the purpose.