Victoria has Mapshare and DigitalTwin for looking at zoning and overlays for individual properties, but doing policy analysis on these website can be challenging. Useful information is dispersed across a multitude of files and a complex set of joins is required to understand how planning laws are applied in Victoria.

Melbourne Dwellings data solves this problem by creating a single dataset with one row for each piece of land in Melbourne. For each row there is information about how many dwellings are currently on the land, it’s zoning, overlays, heritage (including heritage register data), distance to public transport, and proximity to polluting roads.

What you can do with this data

The data makes it easy to understand how properties are zoned. For instance, if we want to see how land in City of Yarra is zoned we can download the CSV file and make a summary in just a few lines of code:

Get the data

If you just want the data you can download it here.

Your first step should be to download any of these files:

• “Melbourne dwelling data.gpkg” - A GeoPackage file, containing one row for each piece of land in Melbourne. This is the only file you need for most use cases (e.g. using R with SF)
• “Melbourne dwelling data.csv” - A CSV file, containing one row for each piece of land in Melbourne. This is handy if you don't want to deal with maps.
• “Melbourne Dwelling Data geometry_only.shp” A shape file that contains the property maps that link to the csv file through id variables lat and lon. This file is a backup for some legacy software that doesn't have good support for GeoPackage files

The files can be analysed in any GIS software such as ARCGIS, FELT, QGIS etc. If you aren’t interested in maps, you can analyse the CSV files in R, Python or any other statistical package.

Understand the data

There is a codebook available here which explains the meaning of every variable, including data quality notes.

Recreate the data yourself

Start by reading the file R/01how_to_run_code.R

Add more information about each property

The code is designed to be modular, so you don’t need to re-run everything in order to add your own useful information about a property such as it’s vicplan PFI or property price data. Simply add a folder in the /R folder and edit the 02.compile.R file to add it in.

Other data sources

• The Urban Development Program has data on each development since 2006. This repo has the most recent flow data from any project in the UDP from 2016 onwards, but a separate file is available from the UDP that has pre-2016 data.
• The VBA releases building permit approvals at a street level.
• The Valuer General has statistics about land values which includes dwelling type, but this data is sold to private providers and is only made available on request to academics or in a very summarised form.
• The state releases data on advertised rents and house prices on a suburb level here
• The ABS releases dwelling changes at a SA2 level.
• Each census tablebuilder product has dwelling counts that can be divided by the usual tablebuilder variables.