Data for Quantifying the Link Between Art and Property Prices in Urban Neighbourhoods

The variable "Rank.Mean.Change" is calculated using data from residential property sales registered at the Land Registry from 2004 to 2013 (https://www.gov.uk/government/collections/price-paid-data). We calculate the mean residential property price per London postcode district. We rank the postcodes according to mean residential property price, where a rank of 1 indicates the highest mean residential property price. We calculate the variable "Rank.Mean.Change", the relative changes in property price, by determining the difference between the 2013 rank and the 2004 rank for each London postcode district.

"Proportion.Art.Photos" is calculated using data from the popular image sharing website Flickr. For photographs in each London postcode district, we deem a photo as "art" related if there is a mention of "art" in its associated textual metadata. We search all the textual elements of each image, including the title, description and tags, using the following regular expression " \<art\> ". We divide the number of "art" images by the total number of images for each postcode for the whole time period (2004 to 2013), in order to create a normalised indicator of "art" per area: "Proportion.Art.Photos".