Assessment of 2 × 2 Associations: Generalization of Signal-Detection Methodology
- 1 February 1988
- journal article
- research article
- Published by Taylor & Francis in The American Statistician
- Vol. 42 (1) , 37-49
- https://doi.org/10.1080/00031305.1988.10475520
Abstract
The signal-detection and receiver-operating-characteristic (ROC) methods, as evolved from an engineering context, are here adapted for use in general biobehavioral contexts. It is shown why, without such adaptation, the results of using ROC methods are not robust in such biobehavioral contexts. With adaptation, the ROC methods prove a valuable addition to the established body of statistical methods of assessing 2 × 2 association, serving as a basis to organize and to interrelate otherwise seemingly disparate approaches.Keywords
This publication has 22 references indexed in Scilit:
- Bias in analytic researchPublished by Elsevier ,2004
- Reverend Bayes' silent majorityThe American Journal of Cardiology, 1986
- Indices of discrimination or diagnostic accuracy: Their ROCs and implied models.Psychological Bulletin, 1986
- The Robustness of Common Measures of 2 × 2 Association to Bias Due to MisclassificationsThe American Statistician, 1985
- Limitations of Electrocardiographic Exercise TestingNew England Journal of Medicine, 1979
- Exercise Stress TestingNew England Journal of Medicine, 1979
- Problems of Spectrum and Bias in Evaluating the Efficacy of Diagnostic TestsNew England Journal of Medicine, 1978
- Observer agreement in evaluating coronary angiograms.Circulation, 1975
- Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit.Psychological Bulletin, 1968
- Measures of Association for Cross Classifications*Journal of the American Statistical Association, 1954