Radon gas occurs naturally in the environment with a variable distribution. In some areas radon concentrates sufficiently within the built environment that it is considered as a public health risk. It is possible, successfully, to reduce radon levels in the built environment, and it has been shown that such remediation programmes can be justified in terms of the costs and benefits accruing. However, the estimated dose received by people in their homes depends on the time spent indoors. The research presented here uses data derived from time activity surveys in Northamptonshire, together with radon data from a representative home, to model potential exposures for different population sub-groups. Average home occupancy ranged from 14.8h (probable error 2.5h) for students to 17.7 (3.1) h for adults; schoolchildren spent an average of 14.9 (1.2) h at home. Over a quarter of adults, however, were in the home for 22 h on more. These differences in occupancy patterns lead to substantial differences in radon exposure. In a home with an average hourly ground floor radon concentration of 467 Bqm(-3), modelled hourly average exposures ranged from ca. 250 Bqm(-3) for students and school children, to over 340 Bqm(-3), for women based at home. Modelled exposures show a non-linear association with total time spent at home, suggesting that exposure estimates based on linear models may provide misleading estimates of health risks from radon and the potential benefits of radon remediation. Highest hourly exposures are likely to be experienced by people with highly occupancy, living in single-storey, ground floor accommodation (for example, the elderly the infirm and non-working young mothers). Since these may be least aware of radon risks, and least able to take up remediation measures, they should be specifically targeted for radon monitoring and for assistance in remediation schemes.