For today’s dashboard week challenge we were tasked with downloading data from the Design census 2019 and creating a visualisation based on it. Generally, the task is divided into two steps; data prep and visualisation, so in this post I’ll describe what I did for both and show what my result looked like.
The as mentioned the source today was survey data from the design census, and the challenge that this presented was that on the whole survey data can be tricky to work with because of difficulty attributing responses to a single person and preventing duplication of data. The link to the current design census can be found here: https://designcensus.org
With the what I knew about the challenges posed by working with survey data, I tried to mitigate this as early as possible within my Alteryx workflow by adding on a record ID for each individual respondent and then identifying the field where a respondent would only have provided a single answer. I split these out and pushed them into Tableau, all of the columns where a respondent could provide more than one answer I split with a select statement and then parsed using the text-to-columns tool with the intention of joining them in later as I needed.
This was a bit of a misstep as I ended up not using any of them and my Alteryx flow looks like an angry octopus.
For my charts I kept everything quite simple(time was definitely not a factor in this…) and created a histogram to show the distribution of different salaries and density map to show where in the U.S the highest concentration of designers were coming from. My final chart was a Likert Scale with a parameter that allowed users to select from a list of parameters and see how this factor affected a respondent’s job satisfaction.
I made a very snazzy title in Powerpoint that I’m very pleased with, here is the final result:
Overall for lessons learned I’d say better time management, more analysis and less molluscs!