Open AccessA method to automate probabilistic sensitivity analyses of misclassified binary variables
Author(s) -
Richard Marshall
Publication year - 2006
Publication title -
international journal of epidemiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.406
H-Index - 208
eISSN - 1464-3685
pISSN - 0300-5771
DOI - 10.1093/ije/dyl226
Subject(s) - sensitivity (control systems) , binary number , probabilistic logic , statistics , binary data , mathematics , computer science , engineering , arithmetic , electronic engineering
where Q0, Q1 (Q 0 0, Q 0 1) are quality indices of misclassification in cases (controls). Here Q1 is sensitivity (SE) re-scaled according the measured prevalence of exposure PX i.e. Q1 5 (SE PX)/(1 PX) and Q0 is specificity (SP) re-scaled according the measured prevalence of non-exposure P X i.e. Q0 1⁄4 SP P X ð Þ= 1 P X ð Þ. The Q indices are sometimes known as chance corrected sensitivity and specificity. 4
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom