Finally, it is day 5 of dashboard week! I feel very productive this week – one dashboard and one blog every day. The biggest challenge for me is always in the morning when preparing the data. Throughout these days, the efficient way for me is to explore data form Tableau Prep and Tableau first, because from this part I can find out which part of the data I need to fix and the direction I want to do the viz.
The data for today is from CDC website (Center for Disease Control and Prevention). The challenging part is to clarify which part of data I need because it contains too many relationships. For example, there are city and track level in geographic level, crude and age-adjusted in data value type and so on. I can only choose one each to analyze, otherwise the population will be duplicated. Here is the workflow for data preparation. I use it to clean the latitude, longitude and filter out the duplicated data.
Of the three main categories which are unhealthy behavior, health outcome and use of preventive service, I chose the first two to analyze if stroke distribution in the US and the relationship between stroke and all other unhealthy behaviors. Here is my visualization.
From the dashboard, I show the proportion of stroke patients in both state and city level listing out the top 5 population locations. For the relationship line chart, I provide a parameter including all unhealthy behaviors. It shows to us that no matter whether it is a big city or small city, the proportion of stroke population to population exhibiting unhealthy behavior are very similar.