There's been another small flutter of dissent following the publication of a comment by Tim Harford, the Washington Post's "undercover economist" citing some old evidence that, all other factors notwithstanding, crime rates are correlated with building height. Glaeser and Sacerdote, the economists who authored the original study, suggested that the main reason for this might be that tall buildings put residents further away from the street and so make street surveillance more difficult. In the tried and true language of Jane Jacobs, such architecture takes eyes off the street. The dissenters argue, perhaps with some merit, that there might be other differences between neighbourhoods with heavy high rise densities and those without, even despite the fact that Glaeser and Sacerdote did their best to take these kinds of differences into account.
My thoughts about this take me in a slightly different direction. We've been thinking quite a lot about height in our lab recently, mainly because the architectural theories we're trying to marry to our psychological work seem mostly to deal only with the two dimensions of a plan view. But in terms of crime, safety, security, and psychological comfort, there are other types of considerations. One of the more surprising statements in Christoper Alexander's early masterwork "A Pattern Language" is his argument that people who live in high places go, to use the psychological term, "nuts".
An interesting study in environmental psychology by Kaya and Erkip showed that in students living in a dormitory where all of the rooms were identical in size and shape, those who lived on the upper floors thought that their rooms were bigger and more private than those who lived on lower floors. Presumably this has to do with things like exposure to strangers, impressions of privacy, and perhaps even the vistas seen from the windows. Though this might seem at odds with Alexander's claim, it still seems to me that people living in units stacked one on top of another have to contend with some very interesting resonances set up by their movements and interactions.
It would be interesting to see a study in which we know not only about the physical living arrangements of the victims, but of the perpetrators as well.