The University of Bath, who graciously stick a few quid into my bank account each month, have just completed a large survey of how people get to the campus. The headline finding is that between 2002 and 2008 the proportion of staff driving to the university has dropped from 69% to 58% and the proportion of students driving has dropped from 29% to 12%. Proportions of people catching buses, walking and cycling have gone up.
So does this represent a dramatic shift in behaviour? I hope it does, and that's certainly how the university will perceive it. But this is a campus whose many and massive car parks are groaning, despite being greatly expanded in area since 2002, and where the buses do not appear any more full or numerous than they did 6 years ago. This all leads me to suggest an alternative interpretation for these findings: people who drive are no longer as willing to discuss it as they were in 2002.
If you regularly do something 'wrong', would you take the time to complete an optional survey on that behaviour? How about if you do something perceived as virtuous? It just seems quite likely that these days, a cyclist or pedestrian is more likely to fill in the survey - and thus receive a little glow of satisfaction from talking about their sustainable travel habits - than a driver who will more likely receive a little twinge of annoyance at yet another attack on their habits. So the survey captures a greater proportion of the cyclists and pedestrians than the motorists, which creates an illusory shift in travel behaviour.
Unfortunately, I can't see a solution to this which doesn't involve obliging people to take part in surveys. But as short-distance car-use becomes less and less socially acceptable, we'll have to expect to see more bias of this sort creeping into transport surveys. And we need to be very careful we don't interpret this as people driving less and cycling more!. Mark my words, this mistake will happen.
Of course, if I'm correct in my interpretation, we might try to read these figures as showing that driving to the university is 100 * (1-(58/69)) = 16% less socially acceptable amongst the staff than it was 6 years ago, which is at least a small victory!