I'm exploring ways to demonstrate graphs in comparison.

More specifically a web tool I've been working on needs to visualise a "Control Group" and an "Experimental Group" in a parallel way.

One solution is to design the graphs in "drawers" so that the user can slide them accordingly.

Few of the issues that arise are:

  • How not to favor one group over the other
  • How to rearrange the layout in such way to reserve as much space as possible
  • How to present the information in a way that the user instantly understands the value of the comparison
  • How to handle large amount of variables and avoid clutter.

Any comments would very helpful.

example possible outcome

  • 1
    You would probably have to determine the number of variables that you are comparing between the control and experimental group first as it will help you work out the chart type. I would suggest looking at science journals for ideas as this is something that is commonly dealt with all the time.
    – Michael Lai
    Commented Feb 22, 2018 at 0:37
  • Thank you Michael, the numbers of variables and chart types varies . The images above are just one example. In charts with many variables there is a visual problem because it becomes extremely busy. Again thanks for the info , I will check the journals. Commented Feb 22, 2018 at 11:00

1 Answer 1


Having one before the other might be inevitable. You need to think what is more important to you. Here are a few questions that might make you choose one representation over another:

  • What separation do you want to emphasize: the experimental vs control group, or the distinct categories (male vs female in the example)? It makes sense to put the values of the emphasized comparison closer, so they are easier to see?
  • Do your values share the same metric and have related magnitudes and scales? This is one of the situations when it makes sense to plot them on the same axes.
  • Do you need to have precise comparison (in this case you need put the bars so that a comparison line is easy to draw with bare eyes) of the values or just an indication of magnitudes (can put them on the two sides of the axis like e.g. a population pyramid)?
  • Do your categories add up to 100% (it might make sense to stack them on top of each-other), or are they independent values the sum of which does not have semantics?

In general graph design depends on the narrative you want to convey. This might be quite difficult if you are developing a platform, rather than having a specific data to present in mind. If this is the case, you would need to provide some affordances for users to make the above choices and others so that graphs enhance the corresponding narrative.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.