Monitoring social media misinformation during the UK elections
Spotting misinformation on social media can be difficult.
During a General Election, it’s even harder.
Full Fact have now fact-checked two General Elections and two referendums. They are the UK’s independent, non-partisan fact-checking organisation. Their fact-checks come in easy to digest blogs, news pieces, infographics and videos.
Montfort’s Ben Matthews was part of the team that helped manage the digital aspect of Full Facts election campaign, particularly around the crowdfunding the £100,000 that the independent charity needed to expand their presence during the snap election.
First Draft, who were funded by Facebook and Google to tackle fake news during the recent French elections, are a global network of newsrooms, technology companies, human rights organizations and academic institutions who work to verify the accuracy of news and digital content. Montfort’s Jaz Cummins provided digital consultancy to the First Draft team when the initiative was first launched.
During the UK General Election, the two organisations joined forces and helped people answer these sorts of questions:
- Are the parties’ manifesto claims about the NHS correct?
- Is the image of people missing after the Manchester attack real?
- I’m seeing this Facebook image about government spending a lot. Can I trust it? i
As First Draft’s Claire Wardle said:
“We don’t want to give oxygen to rumours or fabricated content that is not getting attention or engagement online, but when we see that misinformation might trend, we want to alert newsrooms, who have the ability to debunk or reference it in their output.”
And according to Full Fact:
“Social media is how many people get their news, so factcheckers need to be on top of what is being discussed there. And social media verification can be held to a higher standard by the use of factchecking: rather than just linking to the article where a meme got its numbers from, factcheckers can find the primary source and determine the accuracy of what is being shared.”
But how do they decide what to fact-check?
Full Fact are already developing an internal set of automatic tools to help them factcheck things like the TV debates and Question Time. But getting to the bottom of some stories, can take a specialist fact-checker hours, or even days, so selecting what to put time into is an important choice.
On top of that, the speed at which people’s timelines move, means to really reach people you have to get your fact-check out there as soon as possible. Getting a head-start could be the difference between a couple of hundred eyes, or a couple of million, seeing the factcheck as well as the original claim.
So any tool that can speed things up, or help choose what to put effort into, is of the utmost importance. Here are the tools that helped the team identify and combat misinformation during the UK General Election.
Tools to monitor and identify fake news
Newship Spike gives predictive interactions and what is up and coming in next few hours. This is perhaps the most useful for identifying the original source of a piece of content. A bit of caution needs to be applied here though. For example, a story might look like it’s bubbling up, but die away if it doesn’t break out of the initial echo chamber.
Trendolizer is more flexible in terms of different sources, as you can scan Facebook, Twitter, Reddit or put your own sites in to monitor. You can also filter by what content is most shared by the hour, which helps identify trending content.
CrowdTangle lets you see what’s trending across all social channels. It’s a good snapshot of what’s doing well and a lot of the major UK national news media have access to this as well.
Tweetdeck is a great free tool to use once you have spotted a piece of content bubbling up. Once a piece of ‘viral’ content is identified, you can see who else is sharing it on Twitter and identify accounts to follow for future updates, as they’ll likely be sources of spreading misinformation. You can also see how much engagement the content gets and who else shares that content from the original source.
Google Trends is great as a tool for retrospective analysis, but it isn’t so great for finding out what is happening now – or what will be big soon.
Signal – Facebook’s own tool for monitoring Facebook. (How meta!) Signal makes it easy to source and curate content from major public events, from which journalists can then embed online and integrate it into their outputs.
Social media platforms – As well as Signal, the team used the social media platforms themselves, as well as particular news platforms to monitor trends. Facebook, Instagram, YouTube, Reddit, satirical sites, hyper-partisan sites, union websites, think tanks and other spaces that host public debate were all closely watched. This was really important as not everything could be picked up by the tools listed above.
With Instagram, a lot of misinformation could be spread using images, but without relevant text or hashtags accompanying the image for the trending tools to track. As such Instagram was much harder to monitor as a channel, and there was no replacement for looking at it like any other user.
To set each of these tools up to be effective assistants in the fight against fake news, the team made choices about what hashtags, keywords, accounts and websites to follow. They tried to be as non-partisan as possible when choosing each of these, making sure to cover all sides of the political spectrum.
From a core set of keywords and hashtags, they then added more sources and keywords that were appearing frequently, so this set of search terms increased over time. Carefully trying to find a balance between expanding the search terms and making sure the team actually had enough resources to assess the data that comes through.
The questions the team asked themselves when monitoring these sites include:
- what’s trending right now?
- what might trend soon?
- what makes that content spread?
- how is trending content being spread?
- how does that content spread from fake or partisan channels into the mainstream?
- what content is getting the most emotional reaction and engagement?
- what are the connections between websites?
- what echo chambers exist?
- what content is being shared across echo chambers?
- who or what is sharing viral content?
All of this gave the team a good overview of misinformation that needed to be debunked.
Once identified, they could work with the different social media platforms to head off inaccurate claims quickly, as well as be better informed about how and when claims needed to be factchecked – and communicated about if the information it contained was fake.
The future of factchecking
The collaboration was seen as incredibly useful during the snap General Election, and both parties think that this project is just the start. The precision of factchecking combined with the ability to quickly recognise claims that are spreading, can help us provide verified and factual information to the public.
Full Fact summed up the project in this way:
“This election was different for a few reasons. We saw terrorist attacks that brought a halt to public campaigning, but the Internet continued. There was also a noticeable lack of facts in the political campaigning. A lot of policy promises focused on the future without saying much about what these promises would need now, so much so that we wrote a piece on what the parties weren’t saying.”
First Draft’s use of tools such as Trendolizer and NewsWhip enabled us to see what people were really asking. Some of our most popular work was on a video about the NHS and an infographic on the UK economy that had gone viral. When we produced this content, it was shared by far more people than our traditional factchecks because we were talking to people about things they were already interested in.”