In and around: Identifying predictors of theft within and near to major mass underground transit systems

This paper identifies factors that encourage or reduce pick-pocketing at underground rail stations through a case study analysis of the London Underground. Negative binomial Poisson regression models found possible predictor variables of pick-pocketing selected from the internal characteristics of stations, and features of their nearby surroundings. Factors that increased risk were those associated with; greater congestion inside stations including lifts, waiting rooms and fewer platforms; and, increased levels of accessibility near stations, more paths and roads. Features that reduced risk were those likely to encourage detection and guardianship; stations with more personal validators, staffing levels, and shop rentals; and, the presence of more domestic buildings nearby. Station type was also influential; those that were ‘attractors’ of crime and those frequently used by tourists were at greater risk. The findings suggest a transmission of theft risk between the internal settings of underground stations and their nearby surroundings.