Tag Archives: social network analysis

Original image by Dean Terry

Improving the Social Network Analysis Methodology

We have been collecting data for just shy of a year now and have been developing the Twitter social network analysis methodology for a little longer than that. As you might recall, we have been following the Mobile Health conversation via the #mHealth conversation and have finalised the collection of those data. The processing is almost finished and we can now progress to the next stage of ethnography to further understand what we have collected.

We have been improving the methodology as we go and the last time we received some assistance was to write some code for the Gephi program. Recently, we have been talking with colleagues from the University of Wollongong’s SMART Infrastructure faculty, who have been developing the collection process of Twitter data. Their project is related to flooding information in Indonesia (CogniCity), however Tom Holderness has been kind enough to share his work on GitHub.

When we install this JavaScript, which uses NodeJS, we will have an automated version of the manual process we have been struggling with for the past year. Further, the code is customisable so the researcher can  query the Twitter Stream API for the specific data they require. You can read more about the CogniCity NodeJS application on GitHub.

If we can improve the processing speed further, we will have a research prototype that can be shared with other researchers who are interested in Twitter social network analysis – hopefully a post soonish will reveal this!

aaDH

Accepted paper for #aaDH2014!

We’re in! We just received notice that our social network analysis paper exploring the informal policy actors of mHealth across the the Twitter platform has been accepted for the Australasian Association of Digital Humanities (aaDH) conference 2014.

Here’s the abstract we will work from:

Scholarly interest in data privacy and the regulation of mobile Internet has intensified in recent years, particularly following Edward Snowden’s 2013 revelations about Prism, the US government’s secret communications surveillance and data mining project. Much analysis has focused on the politics and architectures of data privacy regulation and network access. However the surveillance moment also invites scrutiny of academic data gathering and mining online. In open governance movements such as Occupy there has already been considerable debate about the ethics of big data research, particularly where the aim is to track individuals’ online agency around political processes and policy activism. With that context in mind, this paper examines the methodological implications of conducting large-scale social network analysis using Twitter for mobile Internet policy research.

Mobile internet is emerging at the intersection of broadband internet, mobile telephony, digital television and new media locative and sensing technologies. The policy issues around the development of this complex ecology include debates about spectrum allocation and network development, content production and code generation, and the design and the operation of media and telecommunications technologies. However not all of these discussions occur in formal regulatory settings such as International Telecommunications Union or World information Summit meetings, and not all are between traditional policy actors. Increasingly social media platforms such as Twitter and Linked-in host new networks of expertise, informal multi-actor conversations about the future of mobile Internet that have the potential to influence formal policy processes, as occurred during the January 2012 SOPA/PIPA campaigns in the US.

As part of the three year Australian Research Council Discovery project Moving Media: Mobile internet and new policy modes, this research team is mapping and interpreting the interplay between these diverse policy actors in three areas of accelerating media development: digital news, mobile health and locative media. However research into informal policy networks and processes online presents interesting problems of scale, focus and interpretation, given the increased affordances for citizen participation within the international political arenas of social media.

To better understand who these online stakeholders might be in the mobile health field, and how they operate in relation to the normative policy and regulatory circuits, we have adopted a social network analysis methodology, in order to track Twitter-based social relationships and debates. Using a series of hashtags, including #mhealth, #mobilehealth and #healthapps to track ongoing policy-related exchanges, we have begun to identify who is influential in these spaces, what they are talking about and how their input to debate may impact on mobile internet regulation.

This paper will outline that SNA approach and highlight some of the procedural and ethical concerns surrounding big data collection and analysis, which are consistent across contemporary digital humanities research. These concerns include how we can use big data harvesting and analysis tools to align quantitative with qualitative methods, how we can justify our research claims via these tools and how we might better understand and implement these innovative research methods within the academy. In particular the paper will interrogate the methodological suggestion that qualitative methods lead quantitative research, considering instead whether a more rigorous approach is to invert the quantitative/qualitative relationship.

Ingrid Mason and Luc Small delivering the closing address of THATcamp Sydney 2013

#THATcamp Sydney, October 2013 Review

We recently presented some of our social network analysis research of the informal policy actors within the mobile health regulatory space at the THATcamp Sydney unconference. Unfortunately I missed the first day and Fiona could only attend the first session of that day, but we managed to see a full (half) day on the second.

Amongst other great projects I observed, one that certainly stood out for me was the Australian’s Women’s Register. In conjunction with the University of Melbourne, they are doing some outstanding work on collecting and analysing data about Australian women – highly recommend checking this out.

Fiona and I then conducted our session which we took as an opportunity to talk about our work so far and use expertise in the room to interrogate and develop our methodology. It was an amazing experience and you can hear the audio from our talks here.

Some of the points that emerged from the discussion include:

  • Who are the official organisations that are interacting in these conversations?
  • There is existing research to suggest that participation is driving policy
  • Debra Lupton at the University of Sydney to explore data and politics
  • Nick  Thurburger University of Melbourne, experience and developing methodology in converting Excel sheets to data clean http://languages-linguistics.unimelb.edu.au/thieberger/
  • Thresholds of data – what to use, who are they, where are they, rural voices etc – we need to think about our own thresholds which shape the data analysis
  • Demographic profiling in the process – this is critical to understand the ‘why’ of the interactions between the actors
  • Content coding around the data collection, who do we want to hear from?
  • Jake Wallace Charles Stuart university, experience in political party process
  • Policy analysis portal – I am thinking we need to develop a tool similar to this to embed in our site – that is users can bring their data to it and run their own analysis
  • Positive and negative sentiment in tweets – UWS are working in this space
  • Government institutions are legally required to collect social media conversations – interesting!
  • Internal organisational cultural behaviours influencing what is said  and what is not said via freedom of information act
  • If so, could we access Yammer data?
  • Digital engagement as opposed to social media
  • Atlas of living Australia dashboard.ala.org.au - great site to play around with in representing data
  • Internet archive to send or harvest bit.ly links – credibility of user data in government policy
  • Suirveillance in terms of peaks of use, who is using soc med and when
  • Gnip.co twitter api – this is the ‘fat tube’ of Twitter data and Fiona is working on a collaborative approach with UTS
  • Steve Cassidy at Uni of Macquarie – workflow people http://web.science.mq.edu.au/~cassidy/

And if that’s not enough, the wonderful Yvonne Perkins kept a Google doc of the session which can be viewed online.

Great session and i think it is a fantastic primer for our paper which we will present at the aaDH (Australasian Association of the Digital Humanities) conference in 2014 – thanks for the input THATcamp!