Vehicle-based studies of driving in the real world: the hard truth?

Real-world studies of driving behaviour and safety have face validity and have the distinct advantage of focussing on driving in its natural habitat. But their very naturalism can lead to problems with confounds and with noise in the data. This paper reviews the three major categories of on-road studies - controlled observation, field operational tests and naturalistic driving studies - and discusses the major applications of each study type. It also assesses some of the methodological issues that arise in one or more category of study.