医学正常值参考范围在实际运用中常遇到以下三种资料类型:(1)未分组资料;(2)分组资料;(3)有协变量的资料。本文根据一般线性模型的理论,引用广义逆矩阵的方法导出了这三种情况下计算医学正常值参考范围的通用公式,并在此基础上讨论了这三种资料类型之间的区别和联系。 The normal reference range of medical science often encounters the following three data types in practical use: (1) ungrouped data; (2) grouped data; (3) data with covariates. In this paper, according to the theory of the general linear model, the generalized inverse matrix method is used to derive the general formulas for calculating the reference range of medical normal values ​​in these three cases, and on this basis, the differences and relations between the three data types are discussed.