Various rating scales have been derived to assist in the quantification flight test data. Several of these scales currently in use are examined and compared with emphasis on transference from scale to scale. The problem of nonlinearity of scales that depend upon adjective descriptions is considered, and a hypothesis is advanced that a rater may transpose his impressions of performance directly to any scale without recourse to adjective descriptions and thereby relate his mentally derived, divisionless scale directly to a numerical index. Experimental data, though limited, tend to support this hypothesis.