One of the typical use case for a radar chart is for profiling of datasets that comprise of several characteristics that can't be easily compared. The idea is that when you plot these characteristics in a single chart, it creates 'profiles' that allows you to compare between different datasets more easily.

For example, a method of visualizing NBA offenses and defenses is through the use of summary statistics presented in the format of a radar chart.

enter image description here

It seems to me that this information can be presented just as easily using separate bar charts without losing any of the meaning, yet somehow these types of charts are very popular.

Is there any research to suggest that presenting data in this way in much easier to understand when trying to understand more complex relationships?

  • Just so you know, this is Area 51 Meta. (You're a mod on one site, which makes me think that this was just a weird oversight.)
    – CDR
    Apr 4 at 0:26
  • @CDR I don't see a good place to ask this question, and was interested in starting a site related to discussions on data visualization. Apr 4 at 22:09
  • I mean, yeah, it's an interesting question, but we're apparently not supposed to post such things on Area 51 Meta. (I once made a post here asking whether there was interest in having a certain proposal, and was told that that, too, was off-topic.) You're supposed to have a pre-existing community before proposing a site, after which you submit the sample questions on the main site. I suggest you look around (perhaps in the chat of a somewhat related site), and ask if there's interest. (I've long thought this system misguided, but those are the rules.)
    – CDR
    Apr 4 at 22:39
  • @CDR there is a data science stackexchange community already but they don't really deal with the visual presentation/representation of data at the moment, hence I want to create a community for this. Apr 5 at 21:59


You must log in to answer this question.

Browse other questions tagged .