New Observations by the UGC Research Team

In all the planning we undertook for this research on User Generated Content and television news, we never expected that it would involve watching Nelson Mandela’s memorial service at speed. 32 times the normal speed in fact, which is the speed we have been watching segments where we knew it was highly unlikely there would be any UGC.

We certainly hadn’t expected that our random three weeks of data collection would cover so many interesting and valuable stories either: the death and funeral of one of the world’s great statesmen; major demonstrations in Ukraine and Thailand; a helicopter falling out of the sky on to the roof of a pub – as well as the ongoing horrors of the conflict in Syria.

It’s been exciting for us to have so much interesting material to analyse, but we have to admit that actually getting our hands on the recordings from 8 different 24 hour news channels was a huge challenge. (As a reminder, we are analysing seven-days of content from eight channels, but to ensure a range of stories and to limit the inevitable repetition built into 24 hour news coverage, we are analysing eight hours a day over 21 days, which totals 1344 hours of coverage.)

In the planning stages, we had assumed that the news channels would have the coverage themselves, and if we asked nicely, we could gain access. If we couldn’t get access, it would be easy to simply record them via their online livestreams.

Both of these assumptions were utterly wrong. It turns out the majority of news channels don’t routinely record and archive their coverage. So it wasn’t about the difficulties researchers sometimes face in terms of gaining access, it was about the recordings not existing in the first place. All broadcasters said they were willing to help – they just weren’t able to.

A wonderful story about a Philadelphia woman emerged late year. She had consistently recorded television news for 25 years on a complicated arrangement of video recorders at home.  Her son explained how she sometimes had to leave dinners with friends to make sure she had changed tapes. It is an amazing story, but when you start trying to record news consistently yourself, you realize what a stunning achievement it was, and how she really has created a television news archive of 800,000 hours single handedly. When fully digitized this will be an incredible resource.

When we realized there was no ‘easy way’ to collect data from the news organisations themselves, we tried the “we’ll simply record livestreams” approach. This process was certainly far from easy. It required one of the researchers getting up in the middle of the night to check streams hadn’t fallen over, and as with all research projects, we had to call in some favours from friends and family.

A big thanks, in particular, goes to the team in the Media Strategy and Cooperation department of Japanese public broadcaster NHK. The team there recorded all of NHK World in their office for three weeks, just for us.

With the recordings in hand, we have been spending our time since the holidays going through those 1,344 hours of news we have collected. We are about half way through the coding of the content now – hence watching heads of state giving extremely fast eulogies.

We have to say that the results are turning out to be fascinating, especially when we contrast the empirical reality of the output to the normative discussions you hear about UGC. Even in the few interviews we’ve already completed with news managers, there was an understanding and acknowledge of the importance of the main issues that surround UGC: verification, crediting, and seeking permission.

One of the most striking realisations across the board is the inconsistency of crediting User Generated Content. Sometimes it is credited to the user, sometimes it isn’t. Sometimes it is credited in one story, for the same video to be rerun one hour later with no credit. Sometimes it’s “ video”, sometimes it’s just flagged “amateur video”, and sometimes it’s ”unverified pictures”. There appears to be no consistent usage guidelines – regardless of channel or time of day.

As one person suggested to us, these inconsistencies are a product of the stresses of the news-gathering process, something we absolutely understand and appreciate.

For instance, the horrendous pictures of the Scottish police helicopter sticking out of the Clutha Vaults pub in Glasgow came on to our screens late on a Friday evening in Europe. We all know that newsrooms are staffed to the bare bones at that time – and getting help back in to the office at that time is going to be hard.

We can be sure (and we’ll be able to explore this this more in depth in the qualitative interview part of the project) that sourcing and adding on-screen credits for pictures found on Twitter was not the primary concern of producers, editors and reporters at that time. The same pictures were running two days later, still without credit however.

Our goal is to finish the coding and to publish our first report from this quantitative part of the research by the end of February. This is before embarking properly on our many interviews with journalists, editors and news managers at news organisations across the world. We’d be really interested to hear from anyone who would like to talk to us more about the issues we’ve raised here.

Claire Wardle and Sam Dubberley are Tow Fellows working on the Tow Center’s AMATEUR FOOTAGE: A Global Study of User Generated Content in TV News Output at the Tow Center for Digital Journalism.  The Single-Subject News Network is a project made possible by generous funding from both The Tow Foundation and the John S. and James L. Knight Foundation. The goal of AMATEUR FOOTAGE: A Global Study of User Generated Content in TV News Output is a global study into the integration of User Generated Content (UGC) in news output in television broadcasts and online. Follow Claire Wardle on Twitter @cward1e and follow Sam Dubberley on Twitter @samdubberleyTo learn more about the Tow Center Fellowship Program, please contact the Tow Center’s Research Director Taylor Owen: