Innovation and Inspiration: The Campaign for Kansas University

Data Visualization

Free Tools to help visualize your data

Taking a Critical Look

What makes a good visualization? Learn more about creating quality visualizations from these sources:

Types of Data Visualizations

There are 7 common types of visualization:

  ♦  Linear (1-D)
  ♦  Planar (2-D)
  ♦  Volumetric (3-D)
  ♦  Multidimensional (n-D)
  ♦  Temporal
  ♦  Heirarchial or Tree
  ♦  Network

What is Data Visualization?

Data visualization is a way to present data in a graphical or illustrative format to allow for greater comprehension of the concepts you're trying to demonstrate.  Visualizations can be simple like pie charts and timelines, or complex, multivariate, and interactive.

A multitude of examples of data visualization can be found on A Periodic Table of Visualization Methods.

What can Data Visualization do?

Data Visualization has two primary purposes: Analysis and Communication

♦  Summary statistics may miss important trends that visualization will reveal, thus hightlighting areas ripe for analysis.

♦  Visualizations lower the barrier of entry to data analysis, both for researchers and audiences.  It can be easier to understand when you are able to see the data.  Communication between experts and non-experts becomes easier.

♦  Visualizing data can operate as an important first stage of research into a new area of study.

♦  Visualization can preserve the complexity or present multiple views of a single data set.


For an example of how different four data sets with the same summary statistics can look when graphed, check out Anscombe's quartet

When you are designing a research project, your choice of methods is guided by your research questions and hypotheses, combined with your knowledge of the limits of the tools that can be developed. These same research questions and hypotheses can guide the development of visualizations.

Questions to Ask When Deciding on a Visualization Method

When constructing your data set:

♦  How can the phenomena of interest be measured? - the relevent part of the phenomena, that is

Or, If you are working from an existing data source:

♦  Does the available data appropriately measure the phenomena of interest?  What does it measure?

When selecting a visualization type, consider:

♦  Do you intend the visualization to be exploratory, explanatory, or both? 

♦  What is the scope and/or scale of the visualization, in terms of quantity of data and/or the number of variables?

♦  What do you (or other members of your field) expect to be the most interesting?

♦  Do you expect the audience to have expertise in a certain type of specialized visualization?

♦  What can you accomplish with the data, the expertise, and the other resources you have to help represent your phenomena of interest?