HIGH AGREEMENT BUT LOW KAPPA .2. RESOLVING THE PARADOXES

被引:1341
作者
CICCHETTI, DV
FEINSTEIN, AR
机构
[1] YALE UNIV, SCH MED, DEPT PSYCHIAT, NEW HAVEN, CT 06510 USA
[2] YALE UNIV, SCH MED, CTR CHILD STUDY, NEW HAVEN, CT 06510 USA
基金
美国安德鲁·梅隆基金会;
关键词
Agreement; Concordance; Kappa; Paradox;
D O I
10.1016/0895-4356(90)90159-M
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
An omnibus index offers a single summary expression for a fourfold table of binary concordance among two observers. Among the available other omnibus indexes, none offers a satisfactory solution for the paradoxes that occur with p0 and κ. The problem can be avoided only by using ppos and pneg as two separate indexes of proportionate agreement in the observers' positive and negative decisions. These two indexes, which are analogous to sensitivity and specificity for concordance in a diagnostic marker test, create the paradoxes formed when the chance correction in κ is calculated as a product of the increment in the two indexes and the increment in marginal totals. If only a single omnibus index is used to compared different performances in observer variability, the paradoxes of κ are desirable since they appropriately "penalize" inequalities in ppos and ppeg. For better understanding of results and for planning improvements in the observers' performance, however, the omnibus value of κ should always be accompanied by separate individual values of ppos and pneg. © 1990.
引用
收藏
页码:551 / 558
页数:8
相关论文
共 14 条
[1]   MEASUREMENT OF OBSERVER DISAGREEMENT IN RECORDING OF SIGNS [J].
ARMITAGE, P ;
BLENDIS, LM ;
SMYLLIE, HC .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-GENERAL, 1966, 129 :98-109
[2]  
CHAMBERLAIN J, 1975, LANCET, V2, P1026
[3]   THE VALUE AND HAZARDS OF STANDARDIZATION IN CLINICAL EPIDEMIOLOGIC RESEARCH [J].
CHAN, CK ;
FEINSTEIN, AR ;
JEKEL, JF ;
WELLS, CK .
JOURNAL OF CLINICAL EPIDEMIOLOGY, 1988, 41 (11) :1125-1134
[4]   HIGH AGREEMENT BUT LOW KAPPA .1. THE PROBLEMS OF 2 PARADOXES [J].
FEINSTEIN, AR ;
CICCHETTI, DV .
JOURNAL OF CLINICAL EPIDEMIOLOGY, 1990, 43 (06) :543-549
[5]   MEASURING AGREEMENT BETWEEN 2 JUDGES ON PRESENCE OR ABSENCE OF A TRAIT [J].
FLEISS, JL .
BIOMETRICS, 1975, 31 (03) :651-659
[6]   MEASURES OF ASSOCIATION FOR CROSS CLASSIFICATIONS [J].
GOODMAN, LA ;
KRUSKAL, WH .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1954, 49 (268) :732-764
[7]   ESTIMATING FALSE ALARMS AND MISSED EVENTS FROM INTEROBSERVER AGREEMENT - COMMENT [J].
KRAEMER, HC .
PSYCHOLOGICAL BULLETIN, 1982, 92 (03) :749-754
[8]   COEFFICIENTS OF AGREEMENT BETWEEN OBSERVERS AND THEIR INTERPRETATION [J].
MAXWELL, AE .
BRITISH JOURNAL OF PSYCHIATRY, 1977, 130 (JAN) :79-83
[9]   A PROPOSED INDEX FOR MEASURING AGREEMENT IN TEST-RETEST STUDIES [J].
ROGOT, E ;
GOLDBERG, ID .
JOURNAL OF CHRONIC DISEASES, 1966, 19 (09) :991-&
[10]  
SHROUT PE, 1987, ARCH GEN PSYCHIAT, V44, P172