病虫发生程度预报质量评估方法  

Forecasting quality assessment methods of the occurrence degree of plant diseases and insect pests

作　　者：唐洁[1] 唐启义[1] 程家安[1] 

机构地区：[1]浙江大学昆虫科学研究所,杭州310029

出　　处：《昆虫知识》2006年第3期410-413,共4页Entomological Knowledge

基　　金：国家自然科学基金(30370914);国家863计划(2002AA243041)资助。

摘　　要：介绍了对病虫发生程度预报质量进行评估的统计检验方法———McNemar检验和Kappa检验,以及这2种检验方法的原理、假设检验和统计学意义。结合2个实例来说明2种统计检验方法在病虫发生程度预报质量评估中的应用,分析结果表明McNemar检验、Kappa检验能对病虫发生程度预测质量给出更加有效、准确的评估。Two statistical test methods, McNemar test and Kappa test, were introduced in this paper to assess the quality of forecasting occurrence degree of plant diseases and insect pests. And the principle, hypothesis testing and statistical meaning of the two methods were presented. Finally, we combined two practical examples of the proposed methods to explain their application in assessing the forecasting quality of the occurrence degree of plant diseases and insect pests. The result shows that the Kappa test and McNemar test are more effective and precise assessing methods.

关 键 词：农业病虫害 MCNEMAR检验 Kappa检验 统计检验 发生程度 

分 类 号：S431[农业科学—农业昆虫与害虫防治]