Multi-Variate

When making my multivariate display, I used the chart that I had already created about Roger Williams crime statistics as my starting point. I then searched online for statistics that could be relatable to this subject, and decided to use Johnson and Wales crime statistics as another variable. The multiple variables that this new display looked at were five different crimes for two different Universities in Rhode Island over a period of three years (since the Johnson and Wales statistics I found did not include 2006). I took out the 2006 statistics from the RWU chart and then played around with ways to add the new data. I tried making a stacked bar graph, as well as layering different bars. I eventually decided on a line graph for the second set of statistics, and created this to place over the bar graph. In my first solid iteration, I saw that the numbers wouldn’t line up because the bar graph was thicker and therefore the years were off. I solved this by making the X and Y labels invisible on the second graph. Then I saw that the different charts seemed to blend too much as they were both multi-colored. I solved this by making the bar graph different shades of gray, so the line graph stood out against it. Finally, I made the different keys telling what each color or symbol was smaller and labeled each of them. My AHA moment was when I realized that one of the charts needed to not be colored, and when I fixed this it made a huge difference.