The development of an annoyance scale for community noise assessment

Abstract

The development of an annoyance scale for use in community noise assessment is described. In previous studies in which annoyance scales have been used descriptors, intervals, and metrics have been variable and non-standard. Such variability may be a serious shortcoming in comparing studies of assessment of noise reactions. It is not clear how many semantic distinctions can be made to describe subjective reactions to noise, especially the degree of annoyance. There is also great variability in language use reflecting educational, social class, regional and sub-cultural differences. Finally, scaling annoyance responses ordinally, when the underlying intervals are skewed, may seriously distort relationships when analyzed with parametric statistics. In the research described here, an attempt has been made to produce a standardized annoyance scale with descriptors marking clear semantic distinctions, roughly equidistant from each other, and having wide acceptability. Using the Thurstone scaling method, 51 subjects from both high and low educational backgrounds were asked to rate 43 different descriptors of annoyance. The results show that six to seven semantic distinctions can be made. Seven descriptors which mark off roughly equal-sized intervals and which show the greatest agreement are chosen for an annoyance scale. It is felt that the use of such a scale may be more useful in comparing subjective response to noise. Data from a survey lend support to this.