Permap and SPSS create charts that are not visually appealing/easy to study. Hence the need for bubble graph and radar. Hans Rosling made bubble charts popular through his Ted talk where he showed over a 100-year data in the simplest way possible.
A bubble chart displays three dimensions of data. Each entity with its triplet (v1, v2, v3) of associated data is plotted such that v1, v2 are on x and y axes where as v3 is the size of the bubble. These charts facilitate the understanding of social, economical, medical, and other scientific relationships.
To understand how it is done, let us look at the following example:
This table is obtained from SPSS. It compares three attributes across five service providers:
To make a bubble chart, we select the table as shown and navigate through Insert--> Other Charts-->Bubble
However the first chart obtained may not be accurate. Hence the existing bubbles are deleted and the rows are added again and the navigation is as follows:
You remove all the rows and then add each one of them. The graph is then obtained as follows:
To read the graphs in a better way and break the clutter there is format axis option where the values can be changed from origin so that the graph is more spread out.
The end graph is obtained as follows:
The radar for the satisfaction across four stores is as follows:
Thus both the graphs reduce a great deal of complexity in studying multi-dimensional data.
http://en.wikipedia.org/wiki/Bubble_chart
http://chandoo.org/wp/2009/10/05/bubble-chart-tutorial/
No comments:
Post a Comment