Comparison of Preference Scales
W. Pachl, E. H. Rothauser, Günther E. Urbanek · The Journal of the Acoustical Society of America · 1968
For many practical applications, the well-established parameter preference may be said to describe speech quality sufficiently. For the evaluation of preference, several relative and absolute-measurement methods have been proposed that are different—not only with regard to the test procedure but also with regard to the presentation of the test results. Given a particular practical problem it is difficult to select one of these utilitarian methods, because all methods have some advantages and their comparison is difficult. The present paper proposes a unified preference rating scale similar to the one described by Munson and Karlin in 1962. It can be shown that by projection of the individual rating scales onto the preference unit scale, a meaningful comparison of the different measurement results become possible. Based on this unidimensional rating scale, the performance of the different methods is discussed, especially with regard to preference discriminability within a set of continuously degradable speech signals.