Compare two tests

Input Values

  

This utility can be used to compare the performance of two tests evaluated in the same population. Kappa and proportions of positive and negative agreement are calculated, as well as McNemar's Chi-squared value. Corresponding p-values are also calculated for both kappa and Chi-squared values. For infected populations, sensitivity values are calculated for both tests, as well as sensitivity covariances. Similarly, if the population is specified as uninfected, specificity values and specificity covariances are calculated, including Wilson binomial confidence limits for Se and sp estimates.

See also:

  • Fleiss JL, 1981. Statistical methods for rates and proportions, 2nd Ed. John Wiley and Sons, New York. Chapter 13.
  • Cicchetti DV, Feinstein AR, 1990. High agreement but low kappa: II. Resolving the paradoxes. J. Clin. Epidemiol. 43: 551-558, for details of kappa(pos) and kappa(neg); and
  • Gardner IA, Stryhn H, Lind P, Collins MT, 2000. Conditional dependence between tests affects the diagnosis and surveillance of animal diseases. Prev. Vet. Med. 45: 107-122 for calculation and application of test covariances.

Test results:
  +ve -ve
+ve:
  -ve:
Desired level of confidence:
Population status:
Unknown/Mixed
All Infected
All Uninfected

Top

  Home  |   About this site  |   Glossary   |   References   |   Links   |  104 recent calculations


This site was created by Ausvet with funding from a range of sources.
It provides a range of epidemiological tools for the use of researchers and epidemiologists, particularly in animal health.
Please send any comments, questions or suggestions to Evan Sergeant
© 2017 Ausvet