This post was originally written during my training period at The Data School explaining a technique I used on a client project. For a recent client project I was trying to visualise salary data on a map as a means of exploring salaries in different locations. The problem I had was that I only had a city/area name for each of my data sets, as can be seen below. Of course Tableau can pick up these names and plot them as cities but I was finding that some of them were not accurate. Having no geographic coordinates also meant that I had no means of matching these locations to a shapefile to make ..read more

The penultimate day of dashboard week has had DS12 looking at drought data in the US since the year 2000. Find my viz here. As has been a common theme this week, the data had to be obtained from an API call to the United States Drought Monitor. There were a range of parameters you could call from the API, such as land coverage and % land coverage. This could also be obtained on a national level, a state level or even a county level. I decided to go for a county level, as this was the lowest level of granularity. This returned A LOT of data, as it was bringing back data for every US county, for ..read more

Today marked the midway point of dashboard week and the day of Power BI. Getting hands on with a rival software to Tableau has definitely been beneficial. I now have my own opinions of Power BI and will be able respond to questions with these when I’m inevitably challenged on why someone should make Tableau their visualisation tool of choice. More of that later. The challenge today was to obtain data about calls made to the 311 number in New Orleans. “NOLA 311 is New Orleans’ primary source of local government information and non-emergency services. Whether you are a local resident, visitor, o ..read more

The second day of dashboard week had DS12 exploring the finishers data from every year of the London Marathon. Each of us were assigned a two letter code from either our first or last name to narrow down the searches for our web scraping, for me this was ‘ja’. My final viz can be found here. I started my prep in Alteryx by building out a basic workflow that would parse out the relevant data from one page of web scraping. This involved a series of regex tools which looked for data between different HTML tags and can be seen below. At this point I had the data for one page and one year. Next ste ..read more

Today marked the first day of dashboard week for DS12 which believe it or not I was looking forward to as a break from client projects (I’ll let you know whether this was the right way to be feeling later on in the week). The challenge today was to connect to the Star Wars API (SWAPI). As there were only six of us here, we each took on one of the resources that could be called from the API, for me this was planets. The data from SWAPI was in good shape, so it was easy to parse and prep in Alteryx. There were several pages to call so I used a multi-row formula in order to update the URL and mak ..read more

This simple but useful Alteryx process is one that we worked on during our first week at The Data School. Looking back at this work now, it seems very simple, however, at the time, it seemed like a very complex process. I think that is a testament to just how much you learn even in the first few weeks at The Data School. The data we started with can be seen in the image 1, Alteryx interpreted the data as seen in image 2 and we wanted the data to look like image 3, so how did we do that…  Image 1 Image 2 Image 3 Step 1. Use the input data tool to browse for and connect to your text file ..read more

This post was originally written during my training period at The Data School explaining a technique I used on a client project. For a recent client project I was trying to visualise salary data on a map as a means of exploring salaries in different locations. The problem I had was that I only had a city/area name for each of my data sets, as can be seen below. Of course Tableau can pick up these names and plot them as cities but I was finding that some of them were not accurate. Having no geographic coordinates also meant that I had no means of matching these locations to a shapefile to ma ..read more

The penultimate day of dashboard week has had DS12 looking at drought data in the US since the year 2000. Find my viz here. As has been a common theme this week, the data had to be obtained from an API call to the United States Drought Monitor. There were a range of parameters you could call from the API, such as land coverage and % land coverage. This could also be obtained on a national level, a state level or even a county level. I decided to go for a county level, as this was the lowest level of granularity. This returned A LOT of data, as it was bringing back data for every US county, for ..read more

Today marked the midway point of dashboard week and the day of Power BI. Getting hands on with a rival software to Tableau has definitely been beneficial. I now have my own opinions of Power BI and will be able respond to questions with these when I’m inevitably challenged on why someone should make Tableau their visualisation tool of choice. More of that later. The challenge today was to obtain data about calls made to the 311 number in New Orleans. “NOLA 311 is New Orleans’ primary source of local government information and non-emergency services. Whether you are a local resident, visit ..read more

The second day of dashboard week had DS12 exploring the finishers data from every year of the London Marathon. Each of us were assigned a two letter code from either our first or last name to narrow down the searches for our web scraping, for me this was ‘ja’. My final viz can be found here. I started my prep in Alteryx by building out a basic workflow that would parse out the relevant data from one page of web scraping. This involved a series of regex tools which looked for data between different HTML tags and can be seen below. At this point I had the data for one page and one year. Nex ..read more