We provide state-of-the-art research infrastructure and expertise for tracking, collecting and analysing dynamic digital data.
We operate and maintain a databank - the Australian Twittersphere - which is an ongoing collection of public tweets from approximately 838,000 Australian Twitter accounts.
Specialist data scientists and data engineers provide data-related services such as collection, pre-processing, and analysis. This allows researchers to focus on analysis and interpretation.
We can connect industry partners to cutting-edge QUT researchers who can provide social media intelligence and insight.
Using our services
The Digital Observatory collaborates with researchers and organisations to determine their data and analytical requirements, allowing them to focus on their analysis and interpretation.
We provide either raw data or data analysed in a way that is relevant to the research question and able to be interpreted by the researcher.
In general, the Digital Observatory tries to keep costs low for researchers by only charging on a cost-recovery basis.
We operate and maintain a databank (incorporating the Australian Twittersphere) and offer data-related services provided by data scientists and data engineers.
The Australian Twittersphere is a longitudinal, curated collection of tweets from approximately 838,000 Twitter accounts identified as ‘Australian’. The Digital Observatory has maintained reliable, ongoing data collection since early 2018, with approximately 23 million tweets being collected per month. There is also an archive of approximately 2 billion tweets from 2006 to 2016. The Digital Observatory currently collects approximately 37 million tweets per month.
Reliable data goes back to March 2018, however some data are available before this time. There are other options for obtaining historical Twitter data (e.g., using Twitter’s premium API at a cost), and we are happy to discuss this with you.
Geolocation data in the Australian Twittersphere is limited (< 1% of tweets) as most Twitter users opt to not include their geolocation information. If your research question requires geolocation information, it might be possible to infer location information from the tweet text (eg. tweet mentions a specific location) or user profile information (eg. user indicates their location in their profile). However, these methods require additional work and are not suitable for all research questions, especially if high accuracy is important or fine-grained geolocation information is required. We are happy to discuss your needs to see if there is a solution for your research question/s.
The Digital Observatory has published several other specific Twitter datasets, including tweets about Coronavirus during the first 100 days of COVID-19 in Australia, and tweets relating to the 2019 Federal Election. Datasets published by the Digital Observatory can be found on Research Data Finder.
The Digital Observatory has the capability to set up custom longitudinal Twitter collections for research. Depending on the needs of your project or research question, we can customise collection using parameters including time periods, keywords, hashtags, accounts and more.
The Digital Observatory can respond quickly to emerging events and begin a custom collection at relatively short notice. For example, we collected Twitter search results over a period of weeks for tweets about coronavirus that contained YouTube video links. We have also set up a custom collection of tweets from and directed at candidates during the weeks leading up to, and just after, the 2019 Federal Election.
Access to other platforms is currently under consideration including web archiving tools on Yelp, Weibo, Instagram, Reddit, YouTube and Flickr. Other non-social media platforms are also being investigated. If you would like to discuss these or other platforms, please contact us.
The Digital Observatory has skilled data scientists and data engineers who are ready to help you with your research data pipeline. This includes data collection, storage, tidying, pre-processing, analytics and visualisation.
These tasks can be time-consuming and often involve a steep learning curve. We can reduce or eliminate this burden so that you can focus on analysis and interpretation.
Hacky Hour is a weekly informal meetup at QUT for researchers and students wanting to receive assistance from experts in areas such as open research, programming, and data analysis. Hacky Hour is held every Thursday from 2pm to 3pm. The venue alternates weekly between The Pantry at Gardens Point and Beadles at Kelvin Grove. Please email or follow @QUTHackyHour on Twitter for more details.
Due to COVID-19, Hacky Hour is currently being held online via Zoom - registrations essential.
Network analysis using R
We offer training sessions to provide researchers with the fundamentals of network analysis in R. As part of this training, we use Twitter data extracted from the Australian Twittersphere to construct a basic network graph.
By attending the training session, researchers will walk away with:
- an understanding of fundamental principles of using R for network analysis
- confidence to try using R for their research
- additional resources to support further learning about R.
100 days of COVID-19
QUT’s Digital Observatory and Visualisation and e-Research (ViseR) teams have tapped into Twitter to create a powerful and innovative interactive visualisation that tells the story of the first 100 days of the COVID-19 pandemic in Australia.Visit the interactive visualisation
Federal election Twitter talk
The Digital Observatory collaborated with researchers who studied how candidates engaged on Twitter throughout the 2019 Federal election campaign.Find out who was sharing and what was trending
Bushfires: an #ArsonEmergency?
QUT researchers studied the origin and behaviours of Twitter accounts related to the #ArsonEmergency hashtag during the recent bushfires.Read what Twitter revealed
Sun smart awareness
Can social media be used to help improve skin cancer prevention efforts, especially by young people?Encouraging people to #SlipSlopSlap
The Digital Observatory is affiliated with the QUT Centre for Data Science, QUT Digital Media Research Centre and the ARC Centre of Excellence for Automated Decision-Making and Society.