A characterisation of the methodology of qualitative research on the nature of perceived risk: trends and omissions

The issue of how risk is 'perceived' is one of significant research interest and immense practical importance. In spite of this wide interest, however, it is probably fair to say that most emerging 'risk' crises -- whether related to natural or technological phenomena -- come as a surprise to researchers and to society as a whole. Prediction of human responses to novel potential hazards (or novel manifestations of old hazards) is neither reliable nor complete; strategies to ameliorate inappropriate concerns when they arise (or to make realistic inappropriate absences of concern) do not appear totally effective. It therefore seems apt to ask the question: just what have we learned about 'risk perception'? In this paper we conduct a structured review of qualitative research on perceived risk -- to be followed by a subsequent analysis of quantitative research in a later paper -- focusing upon methodological issues. Qualitative research often precedes quantitative research, and ideally informs it; it seeks depth and meaning from few subjects rather than identifying patterns within larger samples and populations. Without adequate qualitative research, quantitative research risks misanalysis of the target phenomenon, at the very least by the omission of relevant factors and inclusion of irrelevant ones. Our analysis here -- of qualitative studies conducted across a range of disciplines, not all of which will be familiar to the readers of this journal -- suggests that this research suffers from an incomplete coverage of the 'risk perception universe', typified by a focus on atypical hazards and study samples. We summarise the results of this research, while pointing out its limitations, and draw conclusions about future priorities for research of this type.