Wiki

Clone wiki

Power BI Visuals - Violin Plot / Home

Home | Usage and Visual Properties | Examples | Release Notes and Known Issues | Roadmap | Privacy Policy | Support


Vioin Plot for Power BI

By Daniel Marsh-Patrick

Twitter_Social_Icon_Circle_Color.png @the_d_mp bitbucket.png @dm-p medium.png @run_dmp gmail.png d.s.marsh@gmail.com linkedin.png daniel-m-p

I'm a full-stack developer and BI afficianado, based in Auckland NZ. I'm a big fan of Power BI and promoting its adoption within my organisation.

I hope that you enjoy using this visual! A short intro is here, but this wiki contains all you need to get started. Please refer to the contents at the top of the page for more details, specifically for usage, support and features. A small overview follows below.


What the Visual Does

A violin plot is a visual that traditionally combines a box plot and a kernel density plot.

A box plot lets you see basic distribution information about your data, such as median, mean, range and quartiles but doesn't show you how your data looks throughout its range. If you have a multimodal distribution (multiple peaks) or some confusion as to where things are clustered then it's not easy to figure this out.

A kernel density plot helps with this challenge by showing the variations in your data across its distribution. It works like a histogram, but uses kernel smoothing to provide a smoother curve where noise might otherwise be present.

This visual provides you with the ability to make these violin plots, with the option of displaying an accompanying combo plot (either a box plot or a barcode plot).

high-level.png

The box plot illustrates:

  • The range betwen the first and third quartiles (box)
  • Median (white line)
  • Mean (white circle)
  • 5% and 95% confidence ranges (whiskers)

The barcode plot shows the individual data points rather than a summary of their distribution. You can apply additional annotations for median & quartiles, should you wish to see these as well (more on this on the Usage and Visual Properties page).

The shape around the combo plot represents the kernel density plot and shows that the majority of our data is clustered around or below the median.

Obtaining the Visual

I Manually Imported A Specific Version and It Didn't Work!

If a custom visual is published to the Marketplace, it will always be loaded from there, even if you manually upload it; this is to ensure that reports are always using the latest version and the user doesn't have to manually upgrade all reports containing a particular visual.

The exception to this if you're using organizational custom visuals, where this is managed by an administrator in your tenant and the version applied by them will be the one that's loaded.

If you want a specific version of the visual, the preferred approach is to use organizational custom visuals functionality to apply this.

Updated