Today was the last day of Dashboard week (woho!) and we were given a 500 Cities census dataset to play with. The 500 Cities census data is about people’s health behaviors, preventive behaviours and their health outcomes across the US. Pretty interesting stuff so I was excited to see what insights I could find. You can find the dataset here. This day had a catch though. While the other days we had the whole day and would present our dashboard the next morning, today we had to finish the dashboard and present it by 3pm. In other words, it was not time to mock around.
Data Prep
The dataset was pretty clean and well structured so there was not much to do in terms of data preparation and cleaning. I fixed a few state names that were truncated, geolocated all the census locations using the provided latitude and longitude and got rid of aggregated measures for cities and states. The data set had census data from both the 2016 and the 2017 census, but the questions and categories were not the same for the two years, so I decided to only look at the 2017 data.
Visualisation and Insights
This dataset was all about finding correlations between health behaviors, preventive behaviours, and health outcomes. Due to the time pressure, I decided early on that I was not going to spend a lot of time on finding these correlations myself, but rather create an interactive dashboard where the user can choose and compare categories up against each other. I also wanted to visualize what states had the unhealthiest/healthiest behaviours and best/worst health outcomes. You can see the final dashboard below and find it at my Tableau Public profile here. If you decide to play with my viz, you will find that there are many linear relationships between the unhealthy behaviours and health outcomes, some states like Colorado and California are much healthier than Mississippi and Ohio.
Final words and Reflection
Overall, Dashboard Week has been such a great week! We’ve learned so much over t he last months we’ve been here at the Data School and it was a lot of fun and very satisfying to spend the whole week using all the skills we’ve learned. The week has also been a great exercise in working efficiently through planning and timeboxing activities. I can safely say that I am better at planning my day, quicker at understanding the data I am given and better at finding insights after this week.
