DETECTION THEORY ANALYSIS OF GROUP DATA - ESTIMATING SENSITIVITY FROM AVERAGE HIT AND FALSE-ALARM RATES
DETECTION THEORY ANALYSIS OF GROUP DATA - ESTIMATING SENSITIVITY FROM AVERAGE HIT AND FALSE-ALARM RATES
复制标题
DOI：
10.1037/0033-2909.98.1.185
复制
发表时间：
1985-01-01
影响因子：
22.4
通讯作者：
KAPLAN, HL
中科院分区：
文献类型：
--
作者：
MACMILLAN, NA;KAPLAN, HL
本文评估了从受试者平均应答率计算检测理论敏感性指标（d“）的统计结果。在大多数情况下，基于这种平均比例的收缩d“仅在很小程度上低估了真实的平均d”，并且其可变性小于平均d“。另外两种方法来计算组的敏感性被认为是：平均和使用非参数措施之前，转换的个人响应率。在无法计算个体灵敏度的情况下，仔细应用检测理论往往比这些替代方法更好。
This article assesses the statistical consequences of computing detection-theoretic sensitivity measures (d'') from response rates averaged across subjects. Collapsed d'', based on such averaged proportions, underestimates true average d'' to only a small degree in most cases, and its variability is less than that of averaged d''. Two other approaches to computing group sensitivity are considered: transforming the individual response rates before averaging and using nonparametric measures. Careful application of detection theory in situations in which individual sensitivities cannot be computed is often better than these alternatives.