Updated Snow GIS data
A while back I released a GIS dataset containing Snow’s Cholera analysis data in modern GIS formats, and georeferenced to the British National Grid (see my previous post). Unfortunately, there was an error in some of the attributes of the Cholera Deaths shapefile which caused issues when using the data.
This error has now been fixed, and the updated data are available to download here: SnowGIS_v2.zip. The link on the previous post has also been updated.
Please accept my apologies for this error.
joke theoretical question: how would the errors have impacted the epidemic?
Interesting question there Jacob! Somehow I had managed to set some of the death locations as having zero deaths (no idea how). In terms of any analysis performed on the data, there would be a ‘hole’ in the data with no deaths, which would definitely look unusual and probably cause significant differences in any statistics plotted. At the very least, seeing a big area with no deaths should lead to some investigation as to why there were no deaths – which would hopefully have turned up my error…
As for the effect on the epidemic itself – if there were less deaths then I guess a lot of people would have been a lot happier!
Hey Robin! This is such a great dataset, and I’ve used it to give a few mapping demos etc. I was just building a tutorial from your new data and ran into the death count update you made. I had always assumed that the 0s were from the Brewery workers that weren’t taking water from the community well? Is that wrong?
On looking at the original maps again, I must have just been wrong. Oh well!
[…] because one dataset is never enough, I downloaded a version of John Snow’s cholera data that Robin Wilson digitized from the original maps. Same procedure here, except color indicates the number of deaths at each […]
[…] save you reading my previous blog posts on the subject, I’ll give a brief overview of my data. John Snow produced a famous map in […]
[…] suggest you read theÂ blog post about the data. Next, download the updated datasetÂ over hereÂ and unzip the […]
[…] in a weird coordinate representation system. So let us use a different dataset. For instance, on Robin Wilson’s blog, one can get datasets in a more traditional representation (here the epsg 27700). We can extract […]
[…] OpenStreetMap ä¸Šå‘ˆç¾, åŽ» Robin’s Blog å’Œ Epsg 27700 […]
[…] Again, the tricky part comes from the fact that the coordinate representation system, here, is not the same as the one used on Robin Wilsonâ€™s blog. […]
Hi. I’m a professor at Harvard, and I’ve been using your GIS files in conjunction with preparing a “mini-course” on the John Snow Story as part of a course my team and I are preparing for HarvardX. I’m grateful for your GIS files, which I’ve imported into ArcGIS. I have a few questions, but the first one I’ll post here… If you look at
http://tinyurl.com/snow-screenshot, you’ll notice some discrepancies where Snow has bars for deaths near the corner of Broad & Cambridge Streets, and you have zero deaths. Can you tell me why? Also, I have some other questions about data, which I’d love to ask you & give you full credit for answers. Shall I ask here, or in a more private forum? Thanks! Alyssa
I’m really glad to hear that you’re planning to use this in a HarvardX course 🙂
You’re right: the discrepancy in that location (and in a few other places) is an error in my digitisation of the data. I will correct that shortly, but I’ll wait before uploading a new version in case any of your other questions also refer to other mistakes that I’ve made! As for those other questions: feel free to email me on email@example.com if you’d like to discuss in more depth via email, and then I can post a summary here (or in the update post when I release a new version).