The fourth part of a selection of pieces of analysis (with data and charts) quantifying different aspects of the Brexit coverage in the UK news media in 2018. Round 4: Is the news outlets’ focus on abstract concepts a reason voter opinions are slow to shift?

In this fourth part of the analysis, we look at the Brexit’s impact on the media industry. Do articles deal more with the abstract (Customs, Migration, Free Trade) or the practicalities (i.e. the sectors that readers work in)?

In terms of headlines, across 2,000 Brexit items, ‘No deal’ dominates (insofar as 7% of headlines can dominate anything). In not too distant second place, ‘Customs’ at 5%, and in third, ‘Immigration’, at 3% of headlines.

Percentage of Brexit headlines referencing topics. (January-August 2018)

Note that these are (on the whole) slightly smaller numbers than we saw in the previous pieces of analysis around MPs (Theresa May was cited in ~15% of headlines, Boris Johnson in 2%, David Davis in 2%).

While this isn’t an exhaustive list of topics, comparing to the references to prominent MPs is interesting — BBC is more likely to lead on a topic-based story than an MP-led piece, particularly compared to other outlets (notably the Daily Mail, Mirror, Sun, Buzzfeed).

Percentage of Brexit articles referencing key topic or prominent MP. (January-August 2018)

Let’s have a look at the contents of these articles (1,300 subset of them). Does the story change?

Well, a little.

‘Customs’ now leads, being mentioned in some form in 46% of articles. ‘No Deal’ and ‘Single Market’ come in second and third, at just over 30% of Brexit items, with ‘Migration’ fourth at 23% of stories — ‘Northern Ireland’ is just behind.

Percentage of Brexit article text referencing topics. (January-August 2018)

But also, ‘Free Trade’ (19%), ‘Tariffs’ (18%). ‘Red Tape’, despite being a mainstay of many EU-critical articles pre-referendum, was only mentioned in 3% of pieces…

So, what else isn’t being covered in the media? Well, sorry people in the creative sector, you are not being mentioned very much. Well, who cares about those people anywa… oh hang on, that is my industry!

But also, the ‘Nuclear Industry’ — 2%.

Science’ — 4%.

Fishing’ — 4%.

Aviation’ — 4%.

Automotive’ — 6%

This, I suspect, may be key — and likely influencing the very slow changes we have been seeing in public perception.

If articles deal with the abstract (Customs, Migration, Free Trade), and not with the practicalities (i.e. the sectors that readers work in)...

... perhaps it is little surprise that attitudes to Brexit are slow to shift.

At outlet-level, the BBC, as a whole, reflects specific issues more frequently than its peers. Perhaps partly due to the fact that it doesn’t focus on opinion-led editorials, but the BBC leads on references to ‘Customs’, ‘Single Market’, ‘Northern Ireland’.

Percentage of items referencing topic/issue. (January-August 2018)

The Sun, however, is way below average on ‘Customs’ references, and above average on ‘Migration’. The Daily Mail has noticeable skews towards ‘No Deal’, ‘Free Trade’ and ‘Northern Ireland’, and leads on ‘Jobs’ as a theme.

The ‘NHS’ — you may note — is oddly low on the list, with the Sun particularly unlikely to reference it in its Brexit pieces.

At the low coverage end of the issue spectrum, the Daily Mail is more likely to reference ‘Sovereignty’, ‘WTO’ and (interestingly) ‘£350m per week’, while the BBC has better than average coverage of ‘Aviation’ and ‘Fishing’ sectors — alongside ‘Border Checks’.

Percentage of items referencing topic/issue. (January-August 2018)

The Guardian is particularly likely to reference ‘Science and Technology’ relative to other news sites, while Buzzfeed picks out the ‘Nuclear’ industry more than others.

If I am honest, what surprised me was how frequently specific Brexit issues were mentioned in BBC News. My personal perception from reading their articles has been that BBC News Online has been less likely to challenge the government narrative than certain other news outlets…

… however, perhaps this is confirmation bias on my side?

While the data doesn’t refute the idea (it is not examining how concepts are presented), it certainly doesn’t support it. The BBC has wider topical coverage — on average — across its Brexit items.

So, this needs further investigation. An analysis of the tone and sentiment of articles is the next (more challenging) analytical step.🔷


  • I generated a list of nearly 2,000 headlines using Google’s site search feature (looking for ‘Brexit’) restricted to items published since Jan 1 2018. For 1,300 of these items (those not behind a paywall), I have assessed the article contents for keywords.
  • I have done my best to remove errant links/references to other items on each site from the text analysed, as this risks over-representing certain terms on certain sites. This does mean there is a margin of error in this analysis.
  • This isn’t a 100% exhaustive list of all items published by each news outlet on Brexit, but it is a good sample, representing at least one per day for most of the major news sites.
  • Each article is only counted once per politician — regardless of how many times they are referenced in that article. I took into account the fact that full names might not be used in articles.



Embed from Getty Images

(This piece was first published as a Twitter thread and turned into the above article with the purpose of reaching a larger audience. It has been minorly edited and corrected.)

(Cover: Dreamstime/Chris Dorney.)