In this blog post, I actually let my roommate choose which data set I should use because I\u00a0was interested to see what she would choose from such a wide range of subjects. Of course, she chose the “Death Data” data set… and I’m slightly scared of falling asleep tonight now….<\/p>\n
For this data set, I initially tried using the Tableau Public data visualization tool, but the column that contained the states names (qualitative variables) did not even show up. So, I went back into the Excel document and added a title to the column itself in hopes of better setting up the data in Tableau… to no avail. I then tried opening up the data in the RAW website, which did not accept the data into regular code for some reason. It had changed colors even and it looked like the data was not properly being transferred over, so that obviously didn’t work. I also tried opening up Google Fusion Tables, but neither Safari nor Chrome was allowing me to open it up… there seemed to be an issue with public\/private access.<\/p>\n
FINALLY, I tried using Plot.ly, which consumed the data set beautifully! After playing around with different types of graphs, I settled on the scatterplot, mainly because it was the best way to show multiple types of deaths at the same time, especially being able to show individual states’ differences. I then played around with shape and color, something Nathan Yau mentions in his article as pertinent.<\/p>\n
My graph shows a simpler visual distinction of how each state overlaps in death rates, crossing homicide rates and suicide rates. It is interesting to see the outliers in a visual manner. I noticed a positive trend in my data (excluding the outliers) as a result of the visual representation.<\/p>\n
*Unfortunately I am unable to embed the link to my data set or visualization into this post because I need to pay for an upgrade to be able to share it, since I used “pro” features. Sorry!<\/p>\n
<\/p>\n","protected":false},"excerpt":{"rendered":"