2023 in (Science) Journalism Research
A whistle-stop tour of this year for the newsletter
Welcome back to TITLE-ABS-KEY(“science journalism“), a newsletter about science journalism research. In the previous issue, I read my first science journalism paper in French and, for the most part, vigorously agreed with everything in it.
This time, I look back at 2023 in journalism research as well as lots of gratitude to scientists and to you, my dear readers.
Today’s paper: Not a paper but a whole roundup of journalism research from 2023, courtesy of the Reuters Institute for the Study of Journalism.
Why: Feels useful and appropriate for the last issue of the year!
Abstract: Journalism has faced many challenges in 2023, including the changing nature of social media, a declining press freedom in many countries and a growing distrust of news media around the world. Despite these trends, so many brave reporters have kept doing their jobs even under the worst possible circumstances, as we have seen in Gaza in the last few weeks. The research team of the Reuters Institute has documented the changes in the news ecosystem in seven reports, two factsheets and several peer-reviewed articles in academic publications. Here are 20 findings from our research in 2023 which will still be relevant in 2024.
Here’s how this is going to work: I’ll take the findings one by one (might combine a few for the sake of sanity and this email not getting way too long!) and try to contextualize them specifically for science journalism, based on my own experience and other research I’ve read and reviewed for this newsletter. I’ll link to the underlying source — a RISJ paper, report, or analysis — so that you can check out the particular rabbit hole that interests you.
1. More and more people say that search and social media is their main way of getting news (Digital News Report 2023)
Essentially, at some point between 2020 and 2021 social media overtook direct access to sites or apps as the main way of getting news online, and now it’s 30% vs 22% (search is at 25%, and aggregators at 8%). The summary of this finding refers to social media, search, or mobile aggregators as “side-door routes” to news, but I wonder whether that’s helpful in 2023 — I mean, you (probably?) can’t reorient your house after it’s been built to have its front on the side, but you should definitely pay more attention to the doors that are used more and more.
Science journalism is very sensitive and vulnerable to context collapse, and social media is generally where context dies a particularly gruesome death, so this is not good news for accuracy or nuance. And since good science journalism tends to be expensive, I fear the continued economic fallout from losing direct access to readers will keep hitting the (remaining) science desks and positions.
2. We see big differences in the level of news ‘platformization’ in different markets (European Journal of Communication)
This seems to be an extension of the previous point, rightfully pointing out that, since most media do not operate globally, aggregate numbers aren’t that useful to them. One of the factors playing into how much your readers rely on platforms for news, the authors write, is the historic strength of your newspaper market. So effectively this is a rather depressing version of the Matthew effect where strong journalism markets can hang on for a little longer while weak ones hardly ever stood a chance. Again, not great for science journalism.
3. Visual networks such as YouTube, TikTok and Instagram are increasingly important for news (Digital News Report 2023)
This very useful finding is also disaggregated into geographies and age groups; the highest numbers are 20% of people aged 18-24 who have used TikTok for news and, for all ages in highest-scoring countries, between 20 and 30 percent of respondents have used TikTok for news. So to me, these are not scary numbers, particularly because, without diving deep into the methodology, I suspect at least a portion of these users may admit to “using TikTok for news“ (1) unintentionally, i.e. they don’t really open the app to Find Some News but rather just see newsy stuff and pay attention to it because they like the app and the videos; (2) based on a fairly loose definition/understanding of news.
What is scary to me is the fact that my own understanding of this landscape is absolutely at the “how do you do fellow kids“ level. I’ve never had TikTok on my phone and only watched videos people sent me on other platforms. I can barely handle making Instagram Stories, I only watch Lofi Hip Hop Radio and music videos on YouTube, and I occasionally watch reels (of strong women lifting big weights at the gym) on Facebook. I don’t think of myself as a representative science journalist, but if I am, what exactly is happening to our field there and how can I know without installing TikTok? I will obsess over this for the rest of my holiday break, surely.
4. These visual networks were a key focus of newsroom leaders in 2023 (Journalism, media, and technology trends and predictions 2023)
Oh well, good to know that at least someone in the newsrooms is paying attention to TikTok (I do hope, however, they are not getting their news from there). There are some doubts, though, that journalism management is in fact well equipped to lead us in exploring these new and shiny frontiers.
5. Twitter’s user base is not representative of the wider population (RISJ)
Ah, Twitter! It’s still there, but it has become so bizarre that every time I start a tweet about climate change or Russian science or whatnot, I find myself wondering if I’m just typing into some kind of black void. I still get quite a lot of useful stuff from looking into the void, but now that requires very specific search queries, whereas I used to rely on my feed and serendipity.
Given the particular ways in which Twitter’s user base is not representative (more male, more well-off and well-educated, as well as more interested in news and politics), there’s likely still some business sense for science journalism and journalists to stick around. But it’s definitely less useful professionally as scientists turn elsewhere and the discourse becomes more and more “black void“-y.
6. Fewer people are commenting on news stories in public (Digital News Report 2023)
Personally I couldn’t be happier about this trend as a whole as reader comments under my science news stories have literally never done me any good. For podcast-related reasons, I’ve had to engage with audience comments a little closer than I’m used to recently, and — no thank you, I’m happy without anonymous feedback with zero accountability.
Professionally, though, I am worried this signals less engagement (just as newsrooms and journalists have adjusted to tracking that), as the data under this point features sharing stories as well.
7. Growth in the percentage of people paying for online news seems to be slowing (Digital News Report 2023) + 8. Most of the people who pay are men with high incomes and high interest in politics (RISJ)
I think I may have cycled through stages of grief about the business models of journalism (and science journalism in particular) at least five times now, since my first PowerPoint slide on this issue in 2015. I held on to my faith in more business-oriented people magically figuring out the sacred ‘Spotify for news‘ puzzle for a little while, but now I am just stuck in “perpetual sadness while navigating the endless ethical maze of propping up my own journalism work with income from comms.“ Feels like that’s the case for quite a few science journalists these days. *sigh*
Also, the paying and thus sought-after audience skewing towards higher incomes has some important implications for the next two insights I think, especially when it comes to science.
9. Many people suspect media managers and media owners force journalists to cover the news in a way that favours their own agendas (RISJ) + 10. People from marginalised communities have very negative views about news coverage (RISJ)
So I’ve always found this idea that journalists are beholden to managers and media owners (and thus their coverage is shaped by those interests or agendas) simplistic in a rather concerning way. First and foremost, it would only be “not great, not terrible” if that was in fact the case, i.e. if editorial lines were dictated by commercial interests of the owner. For media outlets not owned by the state and trying to earn revenue, those interests and agendas would at the very least be somewhat aligned with those of the audience if the owners want to make money. (Not saying this would not mean appealing to and weaponizing people’s worst instincts.)
The real problem is that a lot of coverage is shaped by diverging PR/comms investments from third parties — businesses and political actors — into a mad patchwork of feeding those worst instincts and some consumerism and radicalization as the disgusting cherries on top. I guess if you question people about the agendas of media managers and media owners, you might end up here too, but it would be helpful if even more depressing for audiences to get to this point and keep it in mind.
Unfortunately the best way to fight off this tide is to fund journalistic work properly and directly, and we’re still having massive trouble with that (see all trends mentioned above, basically).
As for marginalized communities, the RISJ work did not focus on science, environment or health journalism, but I imagine if they did, it would be just as bad, with the added insult of condescension and disregard towards those “not smart enough” to understand and be interested in science or medicine. Science isn’t called an ivory tower for nothing, after all, and it can be easy to internalize the elitism.
11. People use ‘shortcuts’ when deciding what news sources to trust on social media (Journalism)
This is actually from a peer reviewed paper I might want to read in full next year for one of the issues of the newsletter; it looks into the kinds of cues in social media content that people use for “quick judgements about trustworthiness, without deep engagement with the actual content itself.” This might be one of the ways of learning a bit more about these platforms without real exposure to them (I am fairly sure TikTok’s infinite scroll will burn my mind.)
I also think it pairs well with the papers I already read on effectively trying to design those cues from scratch for online news about scientific research and stories that merit more transparency about the journalistic process (arguably all stories out there, especially in light of #9 and #10.)
12. News audiences want impartiality – but impartiality means different things to different people (Journalism Studies)
Another peer reviewed paper that could have come with a Captain Obvious sticker — if I weren’t pretty sure that, given #14 and #15 that we’ll get to shortly, it’s not actually that obvious to newsroom leaders. And this is achingly relevant for science journalism where the allure of evidence based reporting is often strong enough to blind us to any and all other considerations, including the interests of our audience. (To be crystal clear, I am not advocating for evidence-free reporting — I’m saying it’s hardly ever enough to be accurate and true to facts for the story to have any impact.)
Also something to file under “next year for this newsletter“: I’ve read a fascinating paper on how readers operationalize fact checking, i.e. how they decide if a claim is true or false, and this would be a complementary insight into readers’ mental models of impartiality.
13. News use increases awareness of misinformation, but does not increase belief in it (The International Journal of Press/Politics)
Somehow I’ve written surprisingly little so far about misinformation and disinformation in this newsletter, but I have The Oxygen of Amplification on my reading list for 2024, and perhaps this is also a chance to come back to the topic with my moderately informed views and opinions. But TLDR of course this is very encouraging: we don’t boost uptake of misinfo by fact checking or debunking it (social media algorithms are happily doing that for us, although if the Facebook Elections Project is any indication, it’s quite a bit more nuanced than we tend to think.)
14. Almost four out of five of the top editors across 100 brands in five markets are white. (RISJ) + 15. Only 22% of 180 top editors across 240 major outlets in 12 markets are women (RISJ)
Again, I’ve been through a few cycles of grief regarding this problem that (1) contributes greatly to audience grievances described in ##9, 10 and 12; (2) signals persistent disparities and discrimination within the industry itself; (3) is plainly annoying as hell. I don’t think science journalism is doing significantly better on this front; there are dedicated spaces for this work and these conversations (truly mainstreaming them would be great, and The Open Notebook is actually doing a lot on that too) as well as research on the flip side of our stories — source diversity.
As you can tell from the next two points, newsroom leaders have, at best, a mixed understanding of some of the underlying systemic reasons for these inequalities:
16. Most newsroom leaders think their organisations lack a proper plan to retain diverse talent, especially at the top + 18. Most newsroom leaders think flexible work has made recruiting easier (Changing Newsrooms)
(also sliiightly suspicious of the use of the word easier with recruitment and flexible work in the context of the pandemic — who are you mostly recruiting in journalism and for what?)
17. Most newsroom leaders are not prepared yet for AI disruption (Changing Newsrooms)
Oooof, AI disruption in newsrooms. Here I’d note the interesting interpretation offered by RISJ who basically say AI disruption — AI fundamentally changing every role in the newsrooms — is imminent because they equate “not prepared“ with “remain[ing] cautious about the potential impact of artificial intelligence on their workplaces“ and not having guidelines and training programs. Of course, I’m prepared to take their word for it but you already know what I think of AI-assisted fact check and live reporting (I’d qualify as cautious and thus unprepared, most likely).
Also, transformation? Absolutely, but feels like disruption is still a bit worn out from all those earlier disruptions.
19. Audiences who follow climate news regularly are more aware of the effects of the climate crisis + 20. Politicians are less trusted than other sources but often quoted on news stories about climate change (RISJ)
On these points, I’d like to note that (1) I’m on the lookout next year for studies that would at least describe if not attempt a proper experimental design allowing to test for causation in the first association; (2) we should arguably keep quoting politicians in climate stories as well as all policy-relevant science stories, but we should always strive to be better at it.
It is critically important that politicians keep getting these questions, that audiences see/hear them get these questions, and that we do our best to get proper answers, including by refusing to repeat their nonsense or bullshit when offered or pushed to do that, or at least always calling it out for what it really is.
That last bit sounds like an actual New Year’s resolution I guess. 🎄
That’s it for this newsletter and this year — this was a lot of fun! I look forward to more cool papers and more (hopefully at least somewhat cool) jokes in 2024. Thanks again for reading this newsletter!