For the past four years, the Student Learning Centre at the University of Leicester has run an event called the Learning and Teaching in the Sciences Conference, aimed at academic staff from science-based disciplines who are involved in teaching. Although this is primarily an internal staff development event, it has always relied on the impact of invited external speakers to ensure good attendance and active discussions around themes in learning and teaching. In 2009, it proved difficult to find suitable dates for external speakers which suited the university calendar, so the idea of dropping the event was discussed. As an alternative, it was decided to hold a purely internal event organized in the form of a participant-driven ‘unconference‘, focused on the theme of assessment. An unconference is a facilitated, participant-driven event centered around a theme. The theme of assessment was chosen in order to promote widespread interest among academic staff and encourage a lively discussion. In addition to the conventional staff development objectives from previous events, we attempted to explore affordances of social technologies to increase participation in the event and expand it beyond the confines of a single institution.
Prior to the event, academic staff were informed through conventional channels such as email notifications and the Student Learning Centre website. In addition to these routes, the authors used their blog networks and tools such as Twitter, Friendfeed and Facebook to promote the event to external online participants. These tools are being used with increasing frequency by conference organizers and attendees (Saunders et al, 2009). Our intention was to create interest in the event and the unconference theme from both internal and external participants. To facilitate online discussions and enable remote participants to find and follow discussion threads, we created the distinctive Twitter hashtag #uollts (University of Leicester Learning and Teaching in the Sciences). On Twitter, hashtags are a community-driven convention for adding additional context and metadata to messages (or 'tweets'). As with tags used on other services, they are user-generated and added inline to all messages destined in order to form part of the stream. By including a hashtags in a message all the messages with that hashtag can be aggregated. This process started a month in advance of the event and continued to provide regular updates of planned proceedings right up to the day of the conference.
Twenty people University of Leicester staff attended the 2009 event in person. These were mostly academic staff, but also included staff from the Library and the Student Learning Centre. A similar number of external online contributors, mostly from UK higher education institutions, also participated in the event.
The meeting started with voting on the topics the attendees wished to discuss. The conference organizers populated the initial list of topics within the assessment theme, but participants were free to add any topics related to the theme that they particularly wanted to discuss. Topics which received the most votes were then discussed in groups of 6-8 for approximately 30 minutes before a short plenary session gathering views from each group. This was followed by coffee, then a second round of group sessions in which additional topics which had not made it into the first round were discussed. Participants moved between groups to select the topics which most interested them. The entire meeting lasted just over two hours and was followed by further informal discussions over lunch, much of which centred on consideration of the unconference format and plans for future sessions in a similar vein. Following the meeting, more formal feedback was gathered from participants via an online questionnaire.
During the entire conference, Twitter messages containing the designated hashtag were projected on screen by a data projector via Twitterfall (www.twitterfall.com). The only time the Twitter stream was not visible was when a brief live technology demonstration was shown on screen in response to a query in one of the plenary sessions. Displaying the Twitter messages allowed the contributions of the participants in the room and the remote participants to be merged. A number of people in the room kept up an active commentary on Twitter via the hashtag to brief remote participants on the discussions taking place in the groups. Twitterfall also allowed participants to see commentary from groups other than the one they were in, and to participate in multiple groups if they wished to. Enough Twitter users were present to ensure that there was at least one per group. This allowed discussions from the groups to be relayed not only to external participants, but also to other groups. External participants contributed to the discussion to varying extents. A few just sent best wishes and then monitored the event as their commitments allowed. Others engaged in active discussion with the physical participants and with each other, including some extensive breakout discussions still utilizing the hashtag.
Analysing the evidence
Data from the Twitter hashtag was downloaded via the free service Twitter tag downloader (http://edumoodle.veloxserv.co.uk/twittertag) immediately after the event to prevent data loss, and analysed. A total of 244 tagged messages were posted, and 23 participants used Twitter to discuss the event online. Figure 1 shows a word cloud representing the tagged content from the Twitter messages. Unsurprisingly, assessment, the theme of the event, featured most prominently in the word cloud. A wide rang of other sentiments, services and approaches to the topic were also expressed.
Figure 1: Word cloud representing the tagged content from the Twitter messages (generated at www.wordle.net)
There were five blog entries reflecting on the event’s structure and content (three of which were from participants in attendance and two from those who were contributing purely online). There were also several photos posted on Flickr and twitpic.com, and six shared links to the blog entries which were generated from the bookmark sharing website delicious (http://delicious.com).
Further analysis of the Twitter data collected was performed using AGNA network analysis software, a free social network analysis tool (Betna, 2005). Figure 2 shows the network connectivity between Twitter users. In this network diagram, participants who were physically present in the room (P) are indicated by blue spheres, remote participants (R) by green squares. The lines between the participants indicate connections representing online conversations (Twitter @replies). Of the 20 people present in the room, six used Twitter during the session and all of these had used it previously. On this occasion, there was no attempt to publicise Twitter to non-users prior to the event. Five out of six physical participants were highly connected (to each other and to remote participants), but one (P006) was not, indicating some variability. Remote participants also showed a range of connectivity: some highly connected and bidirectional (e.g. R012), and some who used the hashtag but not for communicating with others in this network (e.g. R007, etc).
Figure 2: Network connectivity between Twitter users
Twitter data was only recorded for messages containing the hashtag, hence this analysis represents an underestimate of network connectivity since the hashtag may have been forgotten sometimes or not used in bilateral conversations relating to the meeting but not intended as 'broadcast'. All of the data collected as part of this study were available in the public domain. Secondly, the data were only collected from content that had been marked with the event tag. This could potentially allow for data to be missed and offer incorrect analysis. There needs to be an awareness of potential spelling errors and/or participants who are there but not tagging. It could be suggested that the hashtag is a representation of permission for analysis, when briefing the delegates about Twitter (and other social media tools) if they are made aware of the tag and its links to the ongoing data collection.
In addition to network analysis, an online questionnaire was made available and promoted to the local event participants. Information from the feedback questionnaire included the following responses:
Asked to describe their feelings about the event, participants responded:
Would come again:
In response to the Question ”What was the best part of the meeting” participants indicated that:
- The lack of forced structure and relaxed ethos - allowed time for some good conversations to develop and take place.
- Sharing good practice with colleagues
- Getting to know more colleagues within the University
- Really liked the "unconference" style
- The opportunity to meet new people
- Making some very useful contacts & picking up some good T&L practice
- Discussion with colleagues from other departments to find out their experiences
- Learning about what others were doing and getting feedback on some things I was doing.
- Talking with people other than 'the usual suspects'
The worst part was:
- Not enough time - and wished we'd got more support staff (SSDS/Careers/Librarians) out of the woodwork
- Coffee got taken away too early!
- The room, lack of coffee
- the fact that the topics were not decided in advance
- It would have been nice to have been there...
- At times not enough in the group - peer assessment, despite being most popular, only got 4 people to start including 2 from one department so not much of a cross section.
- Needed more opportunity for movement - maybe shortening each segment of the meeting
In the days following the meeting, several of the participants published blog posts describing the event and their impressions of the session. This again reinforced the value for remote participants by providing them with additional channels for comments and presumably increased the likelihood of them contributing to similar future events. Links to these blog posts were collected on the unconference website for future reference and further dissemination via Twitter and email to attendees and remote participants.
One of the attendees brought a prepared handout summarising their work on a novel form of assessment. The electronic file from this handout was added to the unconference website following the event. Another participant later supplied a file summarising a research project on keyboard examinations that he undertook which was also added to the website.
Two of the remote participants discovered areas of potential collaboration through the unconference discussion on Twitter. Both had used a form of pre-practical assessment for students and ideas on how this operated and was marked were shared. Another pair of remote participants discussed the use of mp3 files for the provision of audio feedback to students as a result of the unconference.
Conclusions and recommendations
Although we have described an amplified event run for training purposes, it is easy to imagine how similarly structured meetings could be used in educational contexts, for example undergraduate or postgraduate unconferences aimed at opening up course structures to external participation. Assuming the availability of online connectivity, the major barrier to such innovation is the willingness to participate in open, public, rather than closed discussions. Considerable evidence exists that Twitter is capable of use as a collaborative medium rather than the vanity-driven status updates which are often discussed (Honeycutt and Herring, 2009).
Although the organisers had experience of amplified events of this sort previously, to guard against unwillingness to participate or technical failures on the day, a backup agenda was prepared to allow the event to drop back into a more conventional format if required. This was not needed and the event was considered a success by most, if not all, of the participants. One of the reasons for this success was that online conversations were promoted in advance of the event and enhanced on the day by a core of social connectors, as shown by the analysis in Figure 2.
For an event of this sort to be anything other than ephemeral, it is important that as much of the online data as possible is captured and written up in some form after the event to achieve permanency and lasting impact. For this reason, all participants need to be informed and made aware of the public nature of the online discussions and data collection via any hashtag. However, the hashtag provides a convenient means of allowing any participant who wants to attend the physical event and to have conversations online but not to have their data collated to opt out without isolation.
How to run an amplified event
Running an amplified event is not difficult but it does require some advanced planning and careful use of freely available web tools. What follows are some recommendations, they are not definitive, but they are important.
Effective amplification of an event requires momentum, so it is important to start early. This means talking publicly (and encouraging others to do the same) about the build up to the event; booking, directions, expectations, etc.
Agree a tag and/or hashtag
As explained in the background section above, tags are crucial to amplifying events as they enable the aggregation of what would otherwise be disparate information. For most web services, blogs (such as Wordpress or Blogger) or photo sharing sites (such as Flickr) a simple alpha-numeric tag is sufficient. For Twitter, it is useful to prefix the tag with the hash character (#). The important thing is that the tag is both unique to the event (so that only items related to the event concerned are aggregated) and short (so it is easy to input).
Promote a unique tag and/or hashtag
Just having a tag or hashtag is not sufficient; participants (and potential participants) need to know about it, so advertise it as widely as possible as soon as possible.
Build a network
It might be that the event concerned has a number of people involved who are already part of an existing network, that is to say they already know each other, perhaps in person, or perhaps through interacting online. This means it will be more likely that they will be comfortable with interacting with each other at the event and reduces inertia and generates momentum. To build a network, look out for who is using the tags or hashtags (set up alerts via RSS or FriendFeed) and start conversations.
If a network is already up and running it can be a bit intimidating for newcomers if they are not used to interacting in this way. Inclusivity should be encouraged by simply being welcoming and friendly to newcomers (just as would be done in person).
Getting permissions from people applies largely to photographs. If you are going to be posting photos of people online (which will be discoverable by search because the event is so well amplified) make sure you get participants' permission. The easiest way to do this is via booking form or sign in sheet with a check box for people to opt out and an instruction for them to make themselves known to the organisers at the beginning of the event so that it is clear who does not want to be photographed.
Display the amplification on screen
It is unlikely that everyone who is physically present at an event will be participating online too. Therefore, in order to help everyone see what is going on you can project onto a screen a site which provides an up to date summary of everyone's contribution (we used http://twitterfall.com but other free sites provide similar functionality). To enhance the experience for those who are participating online but are not physically present, also consider streaming video of what is going on in the room (www.ustream.tv).
Use an aggregating site
An aggregating site, such as a FriendFeed group (friendfeed.com) can be used to collect all tagged information into one stream; such as tweets, blog posts, photos, etc.
Collect and archive the data for analysis before it disappears
Information on Twitter is only retained for a finite period so it should be downloaded and recorded after an event (we used http://edumoodle.veloxserv.co.uk/twittertag/) if analysis of the sort described here is required.
School of Creative Industries, University of the West of Scotland
School of Biological Sciences, University of Leicester
Student Development, University of Leicester
Department of Biology, University of Leicester
Betna, M. (2005). Studying Communication Networks with AGNA 2.1. Cognition, Brain, Behaviour. Vol. IX(3), 567-574.
Honeycutt, C., Herring S.C. (2009). Beyond Microblogging: Conversation and Collaboration via Twitter. HICSS '09. 42nd Hawaii International Conference on In System Sciences, 2009, 1-10.
Saunders, N., Beltrão, P., Jensen, L., Jurczak, D., Krause, R., Kuhn, M., Wu, S. (2009). Microblogging the ISMB: A New Approach to Conference Reporting’. PLoS Comput Biol 5(1):e1000263.