Factors influencing the results of a social survey

Social surveys are conducted to determine how annoyed people are in a certain noise situation. The results are typically presented as exposure-response curves showing the percentage highly annoyed as a function of the noise exposure level. A re-analysis of previously conducted surveys reveal that the result is highly dependent not only on the noise level itself but also on a number of other factors like how the survey was conducted, how the annoyance questions were phrased, what kind of response scales were used, how was high annoyance defined etc. The influence of different factors will be discussed and quantified.