Anne Harding

Yiyang & Lingjie:

Yiyang and Lingjie, I really enjoyed the variety of ways that you approached the visualization of US crime data - it's likely that both analysts and the general public would want to understand this information from a number of different perspectives, and you provide several that are particularly useful. The dashboard you put together with the crime rate map and bar chart side-by-side is particularly effective; the two visualizations support each other nicely, with the bar chart showing clear rankings and the map putting these rankings into geospatial perspective. Your population vs. crime visualization, with the option to select individual types of crime from the drop-down box, supports analysis particularly well with its interactivity, which allows the user to hover over data points to see state/population data information. In your total crime ranking bar chart, it was helpful to include the national average as a distinct bar, which allows for immediate understanding of the states with better-than-average vs. worse-than-average crime rates.

If I had any suggestions, I would say that for the map-based visualizations, while the divergent color scheme is definitely appropriate in general and intuitive (red is easily recognizable as indicating a higher crime rate while blue is recognizable as a lower crime rate), the gray values indicating an average crime rate are too similar to the color of the Tableau base map - I don't think you would necessarily have to change the visualization color, but perhaps the color of the base map could be adjusted to better differentiate the two. In addition, I think the scatter plot data with fit lines is a really excellent tool to see how the prevalence of different types of crimes might be related, but because they are so small, it can be difficult to see. If we had a way to have a zoomed-in tooltip view on hover (I don't even think you can do this in Tableau without extensive customization, so this is a "dream" version of your visualization), or perhaps if you used color to indicate squares in the grid with higher levels of correlation, that part of your visualization might be readable. Overall, you clearly put a lot of thought into this project, and I think you've been very effective in communicating the data.

Lance:

Lance, I thought this was a great project - I was particularly interested in reviewing yours because I'm actually not all that familiar with UNC basketball, but your visualization was accessible and understandable for me despite the fact that I might not have been its target audience. This was an especially good example of using annotations to improve clarity and add valuable insight; annotations are often implemented poorly and add visual confusion, but your notes were thoughtfully considered and well-placed. You also made good use of the Tableau story format, incorporating images and using it to highlight changes between "slides" (as in the "Who Was the Coach?"/"Why the Bad Shooting?" pairing). Including the 50% line across your separate visualizations was a useful guideline for easily seeing divergence from the overall averages.

If I were to change anything, I would perhaps reconsider the light blue/dark blue color choice. While they look really great together, and in some ways it makes sense to use a light blue color scheme for a visualization about Carolina basketball, within the user context (Carolina basketball fans), there might be a natural inclination to assume that light blue is UNC and dark blue is Duke. In addition, the dates at the bottom are a little crowded, and you might consider having dates noted only at some sort of interval (5 year, etc.), with the precise date available via tooltip on hover. Overall, you did a fantastic job - I really enjoyed your project!

I have completed the graded portion BEFORE looking at anyone else's work on this assignment. Anne Harding.