I will take you through common mistakes I made during my application process to the data school and what I learned from making them.
My name is Kaitlin and I am part of the newest cohort at the data school, DS37, which has become somewhat of a personality trait for me now! I have been at the data school for just over a week now, and I gained some valuable insights on best practices you can follow when creating your visualisation.
When creating visualizations using Tableau, it is important to keep in mind best practices to ensure that your data is accurately and effectively communicated. I am going to outline some common errors I made throughout the process, and what I did to correct for them!
1: Not selecting the right chart type
One common mistake people make when creating visualizations is not selecting the appropriate chart type for their data. I was working with a dataset with multiple variables, and I initially used a pie chart to visualise my findings. This was not the best choice, as it became difficult to compare the relative sizes of multiple slices. Instead I highlighted my key findings and made individual doughnut charts to visualise this. Other better alternatives to consider would be bar charts – (when in doubt, use a bar chart!).
2: Not using clear labels and titles.
It is important to use clear labels and titles in your visualizations to ensure that your data is easily understood by your audience. Initially I skipped over labelling some of my axis, but after getting feedback, I was reminded to include them, and further explain certain technical terms that may not be familiar to your audience. Always remember to double check at the end, because it’s most likely you’ve missed something.
3: Not using appropriate axes scales.
Another common mistake is not using appropriate scales on your axes. For example, I used a scatter graph with a dual axis for one of my visualisations. Originally I had not synchronised the axes, and this led to distorted results. Once I received feedback, I corrected for my mistakes and my results were a lot more robust. I drew different conclusions to the ones I originally had using my incorrect axes, however this was an interesting insight in itself.
4: Not using appropriate colours.
The use of colour can greatly enhance the effectiveness of a visualization, but it can also be a source of error if not used carefully. To visualise the proportion of votes US candidates got in the 2016 elections I was careful to use a colour scheme that matched the overall theme of my visualisation – red and blue. I made sure however I was careful using the red, as this is one colour that is difficult to distinguish for people with colour blindness. After consulting the best practices, I made sure to avoid other colours such as green and purple in my visualisation, and chose colours that provide good contrast and are easily distinguishable.
These were just a few of many best practices you can follow in your application process. By following these, and taking on feedback, I was able to avoid common errors when creating visualizations using Tableau. Best of luck in your data school journey!