In this paper, we apply random utility modelling techniques to rock-climbin
g in Scotland. Attributes relevant to choices over rock-climbing sites were
identified from focus groups with climbers, along with a categorisation of
principal climbing areas. A survey of climber yielded 267 responses, which
were then used as the basis for modelling. We compare a standard multi-nom
inal logit model with a random parameters approach, and look at seasonal di
fferences in behaviour, and at the implications of different treatments of
travel time. The random utility models showed that most of the attributes s
elected were significant determinants of choice. Welfare estimates of chang
es in site attributes are presented, which are relevant to policy choices c
urrently facing land managers.