Hi All, I apologize for the opaqueness and I will try to make it clearer.
I am comparing two diagnostic tests G (gold standard) and N (new). Both are real tests, real experiments. G is currently the gold standard because it is the best test available, not because it is a perfect test. G is a growth based test and sometimes the test fails (the sample is contaminated with multiple species of bacteria and no results). The new test is molecular based and DNA is present you get a result however, sometimes this test fails as well (the quality control parameters have been exceeded). Both tests are one-shots so there is no opportunity for retesting. Thus what happens is that sometimes G has fails while N detects DNA in the sample and sometimes the reverse is true, G has growth and N fails. So I guess the simplest way to think of this is that both G and N have some level of measurement error that is unknown but I would like to account for in my calculations. So rather than having data for a 'traditional' 2x2 matrix for sensitivity/specificity, my data (where 1=positive and 0 = negative test results) looks like this: G N 1 0 1 1 F 1 0 F 1 1 1 1 1 0 F 0 On Thu, Aug 29, 2013 at 7:32 AM, Michael Dewey <i...@aghmed.fsnet.co.uk>wrote: > At 15:18 28/08/2013, Donald Catanzaro wrote: > >> Good Day All, >> >> I am working with a diagnostic test and comparing the new test to an old >> test. Normally I would be able to calculate sensitivity and specificity >> quite easily. >> >> However, the 'gold standard' that I am comparing my new diagnostic with is >> really 'gold-plated' in that sometimes the 'gold standard' fails >> completely >> and I have no data from the 'gold standard' but I might have data from the >> diagnostic test. Of course sometimes my new diagnostic fails but I have >> data from my 'gold standard' >> > > I am not sure I completely understand the situation, my crystal ball is > becoming rather opaque, but it sounds as though you are looking for some > form of meta-analysis of diagnostic tests when there is no reference > standard. HSROC, available from CRAN, claims to provide this although I > have never used it myself. > > > To me this really starts moving towards classification but I cannot seem >> to >> find the appropriate calculations. >> >> Can someone point me to some web resources to determine the appropriate >> method to be able to deal with the NULLs ? Resources within the medical >> realm would be better (because the rest of the folks would understand them >> better) but not required. >> >> -- >> - Don >> >> Donald Catanzaro PhD >> dgcatanz...@gmail.com >> 16144 Sigmond Lane >> Lowell, AR 72745 >> 479-751-3616 >> >> [[alternative HTML version deleted]] >> > > Michael Dewey > i...@aghmed.fsnet.co.uk > http://www.aghmed.fsnet.co.uk/**home.html<http://www.aghmed.fsnet.co.uk/home.html> > > -- - Don Donald Catanzaro PhD dgcatanz...@gmail.com 16144 Sigmond Lane Lowell, AR 72745 479-751-3616 [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
