In previous chapters different individual aspects and criteria were mentioned that color the results of measurements of speech quality. If one considers the measurement to be an entity of quality, then each aspect can of it be seen as an element of quality of measurements*. This means:

A measurement only attains quality when the quality of its elements is designed in the most effective way to suit its goal. The more effectively the influence quantities of measurements are detected, checked and scaled, the more meaningful are the results of the measurements. It might even be possible to achieve the best possible design if, within a cost-benefit analysis, a measurement of relatively lower quality leads to a higher quality of usability. Planning, specifying and carrying out measurements as well as evaluating their results are often individual phases in the design process of speech quality measurements. This effects the usefulness of the results of the measurements: Speech quality measurement results do not generally have any absolute value, but are always to a certain extent relative and specific.

This perspective leads to a critical view on standardized speech quality measuring processes. They may have the decided advantage that they can be carried out under predetermined conditions in different places and their results compared with each other. However, they have one great disadvantage: Results of measurements can be values of a measurand which is not particularly an important feature of the corresponding measuring object (validity). This is specifically the case for the measuring object...