v4

Built with blockbuilder.org

Data is obtained from the medicare website cms.gov

(https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Physician-and-Other-Supplier2012.html)

Since the dataset is huge. I have made use of portion of the dataset to represent it in the graph.

Discriminability is achieved in the following way for each of the graphs.

1. From the first graph, individual items or scatter points can be identified just by looking at it. From this graph, one can identify position of a point on a common scale. However, it is difficult to distinguish to which category every point belongs to which is clearly addressed in the second graph.

2. From the second graph, one can easily distinguish the scatter points belonging to every category using suitable color codes. Eg: Correlation between average submitted charge amount and average medicare payment amount is calculated using scatter plot. From the graph, we can identify the points belonging to every state with a color code.

3. From the third graph, one can easily identify the data based on area and category because in the second graph even though we have distinguished it by sizes, sometimes its hard to identify as the size of certain scatter points are close to each other. Hence the bubble chart is suitable. Eg: Average submitted charge amount of every state is identified based on the size and suitable color code as shown.