Skip to main content

Data Collection & Analysis: Data Visualisation

A photograph of a laptop screen displaying a pie chart and a line graph.

Data visualisation is an umbrella term, usually covering both information and scientific visualisation. This is a general way of talking about anything that converts data sources into a visual representation (like charts, graphs, maps, sometimes even just tables).

Scientific visualisation is generally the visualisation of scientific data that have close ties to real-world objects with spatial properties. The goal is often to generate an image of something for which we have spatial information and combine that with data that is perhaps less directly accessible, like temperate or pressure data. The different scientific fields often have very specific conventions for doing their own types of visualisations.

Information visualisation is another broad term, covering most statistical charts and graphs but also other visual/spatial metaphors that can be used to represent data sets that don't have inherent spatial components.

Visual analytics is the practice of using visualisations to analyse data. In some research, visualisations can support more formal statistical tests by allowing researchers to interact with the data points directly without aggregating or summarising them. Even simple scatter plots, when the variables are chosen carefully, can show outliers, dense regions, bimodalities, etc.

Types of Visualisations

A candlestick chart.

A candlestick chart is a combination of a line-chart and a bar-chart: each bar represents all four important pieces of information. This style of visualisation is used to describe financial information such as price movements of a security, derivative, or currency. Being densely packed with information, they tend to represent trading patterns over short periods of time, with each candlestick showing data for a single day.

A line graph

A line graph is a visualisation that displays data in a series of data intervals connected by straight lines. This type of visualisation is usually used to describe changes in quantitative data over time, and the relationships between different sets of data. 

 

A Marimekko graph.

A Marimekko graph is a type of bar graph where each bar is of equal length, and is divided into segments. Named after the Finnish design company, Marimekko charts can be hard to read.

 

 

A Nightingale Rose chart.

A Nightingale Rose diagram is a type of histograph displayed in a circle. It is similar to a pie chart, except sectors are equal angles and instead differ in how far each sector extends from the center of the circle. This type of visualisation is usually used to describe cyclic phenomena.

 

 

A bubble chart.

A bubble chart uses circles to represent sets of data that are gathered in groups of three. The three data points determine the position of each bubble on the y and y axis, as well as the size of each bubble. This type of visualisation is suited to describing economic, social, and health data. One disadvantage of the bubble chart is clearly depicting null values or negative data.

 

A pie chart.

A pie chart is a circular visualisation used to illustrate proportions of a whole. The arc length of each slice is proportional to its quantity. Pie charts are suited to describing small sets of data with no more that five values. Pie charts with many values become hard to interpret, and can easily be replaced with a box plot, or bar chart, both of which are easier for the human eye to interpret and understand.

 

A radar chart.

A radar chart is a circular representation of more than three variables. Traditionally, radar charts are shown in a multi-plot format with many charts grouped together on the page, each one representing a single observation.This type of visualisations is ideal for illustrating outliers and commonalities in a dataset. 

 

 

A Sankey diagram.

A Sankey diagram is a type of flow diagram where the size and width of each arrow is proportionate to the quantity of flow. Sankey diagrams visually emphasise the major transfers or flows within a system and are helpful for locating dominant contributions to an overall flow. This type of visualisation is typically used in engineering to describe the transfer of a variable between processes, such as energy, material or cost.

 

A bar graph.A bar graph is a chart that uses either horizontal or vertical bars to show comparisons among categories. The lenght of each bar is used to indicate the value attributed to that category. One axis of the chart shows the specific categories being compared, and the other axis represents a discrete value. Variables in bar graphs can also be clutered in groups of more than one to display results over time. However, this type of visualisation is ideal for illustrating a quantitative value for multiple categories.

Learn more about different types of visualisations by visiting the Data Visualisation Catalogue

Tools

SPSS, SAS & NVivo can all be downloaded from the Edinburgh Napier University Software Download Service for use on your machine (PC/Laptop/Mac), whereas R is available as a free online download from The R Foundation website.

If you would rather not load software onto your device, all of these applications can be accessed via the core applications on an Edinburgh Napier networked PC.

They are also available via this route if you are off campus too, by using the Virtual Desktop Service.

Geovisualisation (Geographic or Geospatial Visualisation), refers to a set of tools and techniques supporting the analysis of geospatial data through the use of interactive visualisation.

Infographics (Information Graphics) visual representations of information, data or knowledge intended to present information quickly and clearly using a combination of text, images, graphs, charts, and tables. Infographics have become popular on the web as a way of combining various statistics and visualisations with a narrative and, sometimes, a polemic. The danger with infographic tools is the tendency to emphasise form over function. 

Loading ...

Eight Elements of Good Data Design

Repetition: Using pattern and repeated elements within a visualisation creates a sense of unity and helps to make the data appear more rhythmic and easy to understand.

Colour: Visualisations don't strictly need colour, but introducing colour helps to anchieve are more captivating design. Beware of colour-combinations that can't be interpreted by people who are colourblind, such as red with green, and blue with yellow.

Contrast: Elements within a design can contrast with one another to reveal relationships and patterns. Contrast is traditionally achieved through colour, but size, shape, and textual contrast are also useful.

Emphasis: Differentiating part of the visualisation to catch the viewer’s attention. Elements can emphasised by varying size, colour, texture, shape, etc.

Balance & Scale: Distributing the visual weight of elements using colours, space, and texture creates hamony and balance. While balance is traditionally achieved through symmetry, it can also be created using asymmetrical, radial, or scaled elements.

Movement: Designing a path for the eye to follow creates movement in a visualisation. This movement can be directed along the lines, edges, shapes, and colours within the visualisation.

Simplicity: Simple visualisations are easy to interpret. Do not expect your audience to engage in visual mathematics in order to understand your data.  

Interaction: Interactive visualisations add depth to data, and can be used to show variations. Visualisations that encourage the viewer to 'play' communicate the story behind the data with ease.

More from the Library

Test Your Knowledge

ake this test to see how well you can recognise the various type of data visualisation that exist.

With thanks to:

Duke University Library (2015) About Data Visualisation. Available from: http://guides.library.duke.edu/c.php?g=289678&p=1930713 

Infographic vectors designed by Freepik