Evaluation of test results is a frequent task for users of laboratory tests
, e.g. when managing clinical cases. In this presentation, interpretation o
f test results is outlined by considering sensitivity, specificity, predict
ive values and receiver-operating characteristic (ROC) curves. These techni
ques are useful when evaluating few laboratory tests. However, multiple lab
oratory tests are increasingly being performed on single animals and in thi
s case, other techniques such as multivariate analysis may be of value. Esp
ecially, logistic regression and principal component analysis are outlined
in this presentation. To illustrate the above techniques, examples are incl
uded.