An operational method is presented for deriving a linear ranking of al
ternatives from repeated paired comparisons of the alternatives. Intra
nsitivities in the observed preferences are cleared away by the introd
uction of decision errors of varying importance. An observed preferenc
e between two alternatives that causes an intransitivity in the course
of the procedure will be reversed if it is of lesser importance. The
method is applicable in case one wants to take account of intensities
of preference and assume these to be monotone with the probability tha
t an observed choice coincides with a fixed underlying true choice.