# Precision of Qualitative Tests

In order to assess the precision of your qualitative diagnostic test, calculate measures of spread in your outcome data, e.g., SEs (standard errors) or 95 % confidence intervals (±2SEMS). The SEs of the sensitivity and specificity are calculated as follows.

Sensitivity (if sens = a/(a+c), then its SE = д/ ac/(a+c)3)

Specificity (if spec = d/(b + d)),......

As an alternative, the underneath procedure is adequate. If 95 % ci intervals cross a prior defined boundary, then the diagnostis test will not be valid (e.g., a boundary of 0.50 or 0.55 may be used). The STARD (standards for reporting diagnostic accuracy) Working Party (Sect. 10.7) says that precision of a qualitative diagnostic test is, traditionally, rarely assessed, and that, therefore, many tests, that are routinely used today, have been erroneously been validated in the past.