Data Visualisations 4 – The basics

A skilled author of data presentation will choose the right visualization to emphasize a message. The data, chart, and supporting descriptions will work in harmony to point out what is interesting. The reader simply goes along for the ride. Unfortunately this is the exception more than the rule. Many data products are a muddled mess of chart choices, obscure labeling, and arbitrary layout. In essence, the author has passed responsibility to their audience to find the meaning.”
Zach Gemignani
on the Juice Analytics blog.

The blog entry shown above provides some fantastic insights for us to consider when presenting data from our OSS data stores. Zach recommends that the best place to find insight in an ocean of data is in the unexpected and then provides the following approaches to finding the unexpected:

  1. Unexpected distributions – pie charts are a good example of visualising weightings of a group of measurements. This becomes powerful particularly if one or more of the weightings are far different than most readers would expect. In the world of OSS, this could take the form of a pie chart of alarm types in a given month showing that certain types of events are using up most of the network operations team’s time
  2. Unexpected patterns or relationships – X-Y scatter-plots are great at showing relationships between data sets. An example might be to show a trend between event rectification time and the number of operator hours dedicated to administrative tasks per week (as opposed to managing the network). Note that outliers may impact trends, where outliers (from trend) may be discarded in certain situations, but may actually form the basis of further research and insight in other cases
  3. Unexpected trends – Line charts are often great at identifying variations from expected trends. An example may be a sharp upward spike in events compared to a long-term trend-line showing a stable number of events per day. These spikes could potentially be correlated with other data sets, such as weather patterns, to see whether rainfall, snow or ice are affecting the reliable operation of certain device types
  4. Comparisons – Side-by-side graphs / data-sets are often helpful for showing results against benchmarks. There are other specialised visualisations such as scoreboards or traffic-light methods that are used to show variation from standard. An example might be to show how many faults are being resolved by the different operators within a team in any given day / week
Read the Passionate About OSS Blog for more or Subscribe to the Passionate About OSS Blog by Email

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.