Statistical Architecture

I liked this passage from a 2008 article by Gary King and Eleanor Neff Powell:

There are fields of study that have not yet been revolutionized by increasing quantification and modern statistics, but its an easy prediction that this will eventually happen given enough enterprising scholars, wherever it would be useful (and unfortunately, at other times too!). Certainly the opportunities for intellectual arbitrage are enormous.

To take one example, for clarity outside our field, consider architecture. By far, the most expensive decisions universities make are about buildings and their physical plant. Yet architecture as a field is composed primarily of engineers who keep buildings up and qualitative creative types who invent new designs: quantitative social scientists do not frequently get jobs in schools of design.

Imagine instead how much progress could be made by even simple data collection and straightforward statistical analysis. Some relevant questions, with associated explanatory variables might be: Do corridors or suites make the faculty and students produce and learn more? Does vertical circulation work as well as horizontal? Should we put faculty in close proximity to others working on the same projects or should we maximize interdisciplinary adjacencies? (Do graduate students learn more when they are prevented for lack of windows from seeing the outside world during the day?) And if the purpose of a university is roughly to maximize the number of units of knowledge created, disseminated, and preserved, then collecting measures would not be difficult, such as citation counts, the number of new faculty hired or degrees conferred, the quality of student placements upon graduation, etc. A little quantitative social analysis in architecture could go a long way in putting these most expensive decisions on a sound scientific footing.