A radar chart is a graphical representation of multivariate data in the form of variables plotted on axes that originate at the same point. Each variable has an axis, and line segments are drawn to connect the variables of each axis. When several points are plotted, the chart resembles a polygon. This has earned it the names of a “web chart,” “star chart,” “irregular polygon,” and other similar names. The axes of the chart are spokes referred to as radii., and the data length of each spoke is proportional to its variable's magnitude.
The main functions of radar charts are to determine which observations are the most similar and to isolate the outliers. Radar charts reveal relative values of a single point. They also offer a simple way to use a random number of variables to observe multivariate data because every “star” is a single observation. In use, radar charts are generated as groups of stars.
Radar charts are utilized across virtually all disciplines and industries given their versatility. They can be used to examine the strengths and weaknesses of an athlete or to analyze the performance of a program or process.
The design of radar charts is very liberal. This leaves it open to certain functional and accuracy issues. One criticism of star charts is that axes can represent scales that are dramatically different, which makes variable comparisons pointless; but by design the user is encouraged to use the chart this way. Another criticism is perception of the chart in analysis. Its axes are not straight lines; thus, the user must mentally create an arc of rotation to map values, which is something that is error-prone. Grid lines are a potential solution to this problem, but they do not appear to be very effective. Some suggest that users avoid radar charts, and opt for stacked or combined charts; which allow for more data without overload or confusion.