Thing 20: Visualisation tools

Researchers produce data in a variety of forms and usually in large quantities. Visualisation tools can help you to synthesize this data and provide engaging ways for presenting it to a broader audience.  This week we take a look at a range of some popular visualisation tools that work for various different types of data. Thing 20 was written by Andy Tseng (Data Infrastructure Architect Research Services, ITS), Bernard Meade (Innovation and Outreach Research Services, ITS), Michael Jones (Senior Research Archivist, eScholarship Research Centre), Leo Konstantelos (Research Data Curator, Digital Scholarship), David Jones (Client Services & Liaison Librarian, Map Collection).

Getting started

Google Public Data Explorer

Google Public Data Explorer is a tool developed by Google Labs that makes large datasets easy to explore, visualise and understand. It offers a simple way of generating different views and graphs (e.g., bar charts, line graphs, etc.) to better understand and present data.

Gapminder

Gapminder is a visualisation software package created by a Swedish Foundation, directed by Hans Rosling, to help enliven and disseminate freely available social science data using animated, interactive graphs.

Tableau Public

Tableau Public is a free desktop tool for generating interactive data visualisation, graphs and reports onto the Internet. You can use this application to analyse any type of structured dataset, and can publish the work to Tableau Public web servers where they will be readily accessible to the general public.

Reflection and integration into practice

Google Public Data Explorer

Currently a range of public data (130 datasets as of 6 August 2014) from organisations and academic institutions—including US Bureau of Labor Statistics, Eurostat, Statistics Iceland, etc.—are available for users to explore interactively. You can also upload your own datasets, using the Dataset Publishing Language (DSPL) format, to Google Public Data Explorer for visualisation and exploration.

It is important to note that you will NOT be able to export data, only manipulate them within the Google Data Explorer environment. However, you can embed the data as part of a website or email the link to someone else. The tool produces interactive, animated graphics using the four available chart formats, i.e., Line Chart, Bar Chart, Map Chart and Bubble Chart.

Gapminder

Gapminder is powered by a software called Trendalyzer (which is owned and licensed by Google) and comes with a staggering range of data collected worldwide (519 datasets as of 6 August 2014), on subjects from national economies to AIDS.

It is also possible to use Gapminder to display data over a map so the statistical changes can be seen geographically. However, it has a limited ability to upload and visualise private datasets (possibly via the use of Google Docs) with certain functionalities (e.g., map) not supported.

Tableau Public

Tableau Public is an advanced desktop tool for people who don’t have programming skills but still want to create highly interactive data visualisations on the web. It offers a “visual data window” that allows you to connect different data sources by simply pointing and clicking. You can also apply various filters before exporting the data. Tableau Public can connect to Microsoft Excel, Microsoft Access, and multiple text file formats but has a limit of 1,000,000 rows of data in any single file.

There are a rich selection of visualisation features such as “word cloud”, “bubble map”, “tree map”, etc. After the data is published, you can browse these visualisations using Tableau Viz, a thin AJAX-based application, directly within their web browsers.

The published data saved to Tableau Public is accessible by the general public but the you can remove your content later if needed. There are also paid versions of Tableau software, namely Tableau Personal and Tableau Professional, that allow you to save your visualisation works locally.

Deep dive

Apart from the above-mentioned tools, here are several alternative options that are worth mentioning:

IBM’s Many Eyes is free web tool that allows anyone to visualise data in a variety of attractive ways, as well as present the data online. It’s a community-supported tool with just over 150,000 free datasets currently available to choose from. You can also upload your own datasets to the Many Eyes website, and use its visualisation tools to explore that data.

QlikView is an advanced visualisation tool mainly for Business Intelligence (BI) data. It can also be easily integrated with other BI platforms to provide demos and training courses.

University staff members also have access to ABS TableBuilder Pro, a web-based self-service tool that allows you to create tables, graphs and maps of Australian Census data.

Finally, AURIN Portal is one of the end products of a $20 million initiative (AURIN), led by the University of Melbourne, that offers open access to data from more than 60 institutions and data providers. Secure data and the tools for researchers to analyse, visualise, map and integrate diverse datasets are available via the portal.

As you can see, there is are a wide variety of visualisation tools out there. Some are pretty much ready-to-use and are an easy way of manipulating and presenting data if you’re not particularly experienced in programming. If you don’t mind getting your hands dirty,  so to speak, there are also some free, open-source JavaScript libraries that you can use to develop your own visualisations on your own website. For example, Timeline is a web widget that creates interactive horizontal timelines (great for visualising temporal data), Modest Maps is a small but extensive library for generating interactive maps, and Flot is a jQuery library for plotting attractive interactive graphs.

Considerations

Most of the tools discussed here use publicly available datasets for generating the visualisations and graphs. When using a tool that allows you to upload your own data collection, for instance Tableau Public, you need to consider if these are any restrictions on those data being hosted on a public server. If in doubt, always consult with the University’s Office for Research Ethics and Integrity in advance. There is also an excellent blog post from the Copyright Office that provides useful information when hosting/sharing your research data with external tools or services.

Andy Tseng (Data Infrastructure Architect Research Services, ITS), Bernard Meade (Innovation and Outreach Research Services, ITS), Michael Jones (Senior Research Archivist, eScholarship Research Centre), Leo Konstantelos (Research Data Curator, Digital Scholarship), David Jones (Client Services & Liaison Librarian, Map Collection).