Statistical techniques for comparing measurers and methods of measurement: A critical review
Statistical techniques for comparing measurers and methods of measurement: A critical review
复制标题
DOI：
10.1046/j.1440-1681.2002.03686.x
复制
发表时间：
2002-07-01
影响因子：
2.9
通讯作者：
Ludbrook, J
中科院分区：
文献类型：
--
作者：
Ludbrook, J
1. 临床和实验药理学家和生理学家经常希望比较两种测量方法或两个测量器。生物统计学家坚持认为，应该寻求的不是方法或测量方法之间的一致，而是分歧或偏差。如果测量是在连续尺度上进行的，主要选择是Altman-Bland差分法和最小积回归分析。有人认为，虽然前者执行起来相对简单，但它没有充分区分固定偏差和比例偏差。最小产品回归分析虽然执行起来比较困难，但确实实现了这一目标。生物统计学家几乎普遍认为，皮尔逊积差相关系数(r)作为偏倚检验是没有价值的。如果测量是在分类尺度上进行的，无论是无序的还是有序的，最流行的分析方法是使用kappa统计量。如果类别是无序的，则未加权kappa统计量(K)是合适的。如果分类是有序的，就像大多数临床、心理和流行病学研究的评定量表一样，加权kappa统计量（K-w）更可取。但K-w对应的是类内相关系数，与连续变量的r一样，它无法检测到偏差。在有序分类变量的情况下，检测偏差的简单技术被描述和推荐给调查人员。
1. Clinical and experimental pharmacologists and physiologists often wish to compare two methods of measurement, or two measurers.2. Biostatisticians insist that what should be sought is not agreement between methods or measurers, but disagreement or bias.3. If measurements have been made on a continuous scale, the main choice is between the Altman-Bland method of differences and least products regression analysis. It is argued that although the former is relatively simple to execute, it does not distinguish adequately between fixed and proportional bias. Least products regression analysis, although more difficult to execute, does achieve this goal. There is almost universal agreement among biostatisticians that the Pearson product-moment correlation coefficient (r ) is valueless as a test for bias.4. If measurements have been made on a categorical scale, unordered or ordered, the most popular method of analysis is to use the kappa statistic. If the categories are unordered, the unweighted kappa statistic (K) is appropriate. If the categories are ordered, as they are in most rating scales in clinical, psychological and epidemiological research, the weighted kappa statistic (K-w ) is preferable. But K-w corresponds to the intraclass correlation coefficient, which, like r for continuous variables, is incapable of detecting bias. Simple techniques for detecting bias in the case of ordered categorical variables are described and commended to investigators.