Abstract
Inter-disciplinary research has been published on the adverse effects of noise in health, education and other areas. These studies often produce findings that on close examination are ambiguous; display flawed methodology and conclusions not supported by the measurements. This paper highlights common anomalies and errors in noise assessment which have passed through the peer-review process to publication in journals. Often ambiguity in the findings is the result of using incorrect notation and terminology, combined with a poor understanding of appropriate noise descriptors. It is common practice to perform the statistical analysis on the sound levels in decibels. But this leads to an underestimation of the effect because adverse health effects of noise typically display a dose response to the product of the sound pressure squared and time. Methodologies are often described without reference to best practice or existing standards that typically describe in detail robust measurement procedures. There is a need to educate researchers involved in noise studies about the need for competency in noise measurement and assessment. It also highlights to journal editorial staff about the need to include in their peer review process someone with appropriate expertise in noise.