The psychometric objections to using essays and other subjective tests for measurement can be overcome by a many-facet Rasch model. This model enables judge-awarded grades to be transformed from their arbitrary, local, non-linear rating scale form into linear measures with explicit generalizable meaning of specifiable reliability (standard error) and validity (fit). Flexible judging plans and accommodation for missing data simplify application. Quality control over the judging processing is readily available through meaningful fit statistics. The following is a sample many-facet Rasch model for an essay examination, using a rating scale of M + 1 ordered response categories with "0" labeling the lowest and "M" labeling the highest: B(sub n) is the ability of examinee "n"; D(sub i) is the difficulty of item "i"; C(sub j) is the severity of judge "j"; and F(sub k) is the difficulty of the step up from category k-1 to category k, and k=1,M. A maximum likelihood estimate is obtained when the expected marginal score for observations in which the parameter participates is set equal to the observed score. It is concluded that the many-facet model can be applied to any testing situation in which the constituent components can be conceptualized to combine additively to govern the raw rating category responses. Six figures are included. (Author/SLD)