Accessibility

Day two of dashboard week started with a presentation on what accessibility is and how it is (or should be) incorporated into Tableau dashboards. Collin gave this presentation which was a big eye opener to some of the difficulties people face when digesting visualisations. Accessibility in a dashboard context is purposefully including specific colours, text and mark sizes, labelling & captions and instructions to maximise the range of audiences that can interact with your dashboard.

One key area of the presentation which stood out and gave a lot of context was the question: Why make visualisations more accessible? There are definitely more reasons than listed below, but a good summary of why it is important to make dashboards more accessible is four-fold:

● Economics: 18% of working British people have a disability. Increasing the accessibility of a dashboard directly increases the number of people who can use your dashboard.

● Goodwill: advocate for accessibility or universal design gets your work noticed more.

● Legal: Some companies may have legal obligations to make dashboards more accessible.

● Moral: In many ways, people who have a disability can struggle with digesting forms of information, so making dashboards more accessible improves their ability to take in visualisations.

The task

After our presentation, and being given access to different resources which can help with the process of building and checking the accessibility of a dashboard, we were given the task of creating an accessible dashboard using data from the Fix My Street Glasgow data set (credit: FixMyStreet via the Urban Big Data Centre). Helpfully, Louisa had already done some data prep for us - reverse geocoding location coordinates to give us street names and addresses and inputting those into a csv file for us to use.

Step 1

First of all I downloaded the csv and loaded it up into Alteryx. On first glance, it didn't look like there was too much prep to do, however there were a couple of string fields which included the whole address of a pothole or hazard location (street name plus postcode, city and country). After some filtering and using the RegEx tool to extract the relevant sections of these string fields, I outputted the data into a csv file to upload into Tableau.

Step 2

Next I loaded up the data in Tableau. Because we had limited time today to build a dashboard, I wanted to keep it very simple whilst ensuring I had encompassed as many points from the presentation earlier today on accessibility.

The first thing I did was to visualise the locations on a map to make sure all the location coordinates had come through ok (see below).

Step 3

After investigating the data a little bit in Tableau, I decided I wanted to give the user a very simple experience to find any hazards close to them based on their postcode. I created a couple of calculated fields and used a parameter to allow the user to input their postcode. The rest of the dashboard would then filter to give them a number of how many hazards are nearby, with the option to download an excel file of the full data, with street names of the hazards nearby.

Challenges

The main challenges I faced today were trying to incorporate best practices for universal design. Although my final dashboard (below) is very simple, it took me a long time to drag and drop the specific containers in the right sequence so that a text reader would read them out in the right order. Apart from that, using colours which didn't clash and good font sizes took some time as well.