Less discussion, less content: how social media experiences of the election have changed

Over the last three weeks, we’ve conducted a small self-funded research project to explore how social media and smartphones are informing voting intention in the imminent general election.

We’ve visited and interviewed six respondents at length in their homes and analysed screen recordings showing some of what they have been seeing on their phones.

In the run-up to the 2019 general election, we conducted a similar exercise. On both occasions the work was featured in The Guardian. This year’s research is covered here and here.

Repeating the research has given us a fascinating insight into how media consumption and online behaviours – specifically around political engagement – have transformed over the last five years. And, although the sample was small, it’s been interesting to compare these changes with the trends we’ve seen more widely in that period in all our work on media use and the impact of technology on people, culture and society.

Before we discuss our observations from the research, it is important to acknowledge that there are potentially significant differences between the two elections that might have an impact on how engaged people are with politics in general. We feel it’s fair to say that the two leading parties at the last election were led by individuals whose politics were poles apart. In their own way they were both charismatic and somewhat eccentric. And while the polls tended to suggest a Labour win in the run-up to the election, it was also possible that it would be a close-run thing. We’ll leave it to you to decide how you think the current polling, parties and ‘personalities’ compare with 2019.

Our observations from the research this month:

1. We saw far less mainstream news content on the participants’ social media feeds.
The difference in five years was stark. None of our respondents were big users of X (formerly Twitter), and certainly not for news. In past work for Ofcom and media organisations and in self-funded projects we would have seen posts from mainstream news providers on Facebook feeds, often reposted by friends or other contacts on social media more generally.

When we did see news and election-related content, it was generally unbranded and lacked detail, and most of it was memes and repurposed content. This material was often designed to denigrate the politician featured, and in some cases to misrepresent them or their policies, through short edits of real speeches or images taken out of context. These were more ‘cheapfakes’ than deepfakes, although we did see a fair amount of AI-generated imagery in a couple of the participants’ feeds, particularly about the conflict in Gaza.  

There was a clear age divide in the sample. Older participants listened to the radio or watched TV for news – often the BBC – and were regularly using news sites or apps. In contrast, the younger participants had almost no engagement with mainstream news sources. Some weren’t aware of the names of party leaders and almost all ‘news’ was picked up from the reactions and comments by others on social media. 

“It’s from the reaction. I’ll have an opinion from the reaction…I suppose I’d be more interested in people’s reactions than, you know, the thing itself immediately.” Finley, 19 

2. We observed less discussion, less sharing, less posting.
In our previous general election research we saw and heard about people frequently sharing news stories or their views on news. We got a sense of discussions between friends on WhatsApp, for example. But this time, beyond the odd ‘like’ of a meme, there was markedly reduced engagement and less desire to discuss political views. Some participants told us they had tried to keep away from discussing politics or the news because they felt the tone of online spaces had become more toxic. 

“During the last few elections, I did unfriend a load of acquaintances… because people were so offensive when they had a belief that they would shout this out on Facebook pages.” Ava, 67

In other research we’ve done on image-focused social media (TikTok, Facebook, Instagram) we’ve seen a year-on-year shift away from social interaction, posting and commenting on posts. People have been trying to keep a lower profile because they’re concerned how others might react or judge them. At the same time, image-focused social media content has become increasingly professionally produced. We would describe social media as having become less social and more commercial.

This sits in contrast with discussion-focused spaces such as on Twitter, Discord and Reddit, which tend to be more tolerant of diverse views and language, and less moderated. People are more willing to put their views across, and to a certain degree they visit expecting to be confronted. The participants in this project weren’t using these spaces to discuss politics.

3. National issues concerned the oldest participants, younger voters were more interested in personal, emotional or local issues.
National issues such as the economy or the NHS appeared to be the primary concerns for the two oldest participants, and they were actively engaging with news content to feel better informed and decide how to vote. It was notable that for the other four participants the focus was almost exclusively on issues that were more personal, emotionally driven or local. This felt a long way from Westminster and the way the media typically covers politics.

It’s difficult to know whether the disconnect between the younger voters and mainstream political messaging is down to a lack of familiarity with the issues because it’s largely absent from the media they are consuming, or because politicians and media storytelling feels so irrelevant and uninspiring that lifestyle content on TikTok or Instagram is simply more appealing than mainstream sources and channels.

If you’d like to talk over these findings or their implications, or find out more about our wider work on media and digital trends and media literacy, please do get in touch.