Facebook disputes WSJ's Instagram story, claiming internal research shows Instagram made difficult times better for teen girls on 11 of 12 well-being issues

- Contrary to The Wall Street Journal's characterization, Instagram's research shows that on 11 of 12 well-being issues …

About Facebook 2021-09-27 Pratiti Raychoudhury

Context & Ripple Effects

Facebook's rebuttal follows the WSJ report on internal findings of harm for a sizeable group of users, especially teen girls. The disagreement is not only over Instagram's effects, but over which portions of the company's research should define the public account.

Facebook's response was challenged for addressing some findings while not releasing all of the data it cited; the contemporaneous critique of that response specifically flagged addiction as an unresolved issue.

First-order effects

Facebook and Instagram face an immediate credibility test: their claim of improvement on 11 of 12 issues is now evaluated against internal findings that the WSJ says documented harms for some teen girls.
The WSJ gains a clearer evidentiary contest with Facebook as the company selectively shares research while the paper publishes the documents underlying its reporting.

Second-order effects

Instagram's efforts to attract and retain teens, including recommendation changes intended to bring them back, face greater pressure to show how engagement decisions account for the well-being findings.
Selective publication of research shifts the dispute toward document access and methodology, giving critics a basis to challenge Facebook's framing rather than only its conclusions.

Third-order effects

If companies increasingly contest reporting with partial internal research releases, product-safety debates will hinge more on disclosure practices and the interpretation of internal evidence.
Teen-focused growth tactics may become harder to separate from well-being accountability when the same internal research is relevant to both product design and public claims.

The trend: Platform companies are being pushed to treat internal user-well-being research as externally contestable evidence, particularly when it conflicts with engagement-focused teen growth efforts.

Discussion

@andymstone Andy Stone on x
@CaseyNewton We're sharing these decks with Congress in advance of the hearing this week and we're evaluating how we can release it to the public at some point.
@fbnewsroom @fbnewsroom on x
Contrary to The Wall Street Journal's characterization, Instagram's research shows that on 11 of 12 well-being issues, teenage girls who said they struggled with those difficult issues also said that Instagram made them better rather than worse. https://about.fb.com/...
@stevekovach Steve Kovach on x
Last night FB attempted a point-by-point analysis of why it thinks WSJ was wrong about Instagram harming teen girls. https://about.fb.com/... This morning it “paused” Instagram kids over concerns raised in WSJ series. https://www.cnbc.com/...
@danielsand Dan Sandler on x
@CaseyNewton Is it just me, or is their PR purposely written to be difficult to actually parse their message. They at one point talk about how the WSJ is basing on a super small N (<50) so their results are not significant. Then claim the same study shows they are doing good.
@jeffhorwitz Jeff Horwitz on x
Thinking a bit today about former FB employees, including @YaelEisenstat and @szhang_ds, who publicly criticized the company and got accused of grandstanding. Regardless of their seniority or tenure, everything the @WSJ team has written about FB is compatible with what they said.
@bplewis @bplewis on x
Well color me shocked to read that Facebook says a Facebook study about a Facebook property finds that said Facebook property is not a toxic cesspool for teenagers according to Facebook https://twitter.com/...
@rmac18 @rmac18 on x
FB spox suggests that the slides will be released at some point, but why wasn't it with this blog post? This just gives people more to be suspicious about. https://twitter.com/...
@mikeisaac Rat King on x
facebook goes on the offensive against the WSJ report on Instagram and teens research https://about.fb.com/...
@rmac18 @rmac18 on x
The problem with these blog posts is that Facebook is at such a trust deficit it can't just reference “research” it has done and ask people to believe it. It'd go a long way to release the full report (which the co already hid internally from employees). https://about.fb.com/...
@samidh @samidh on x
I'd find blog posts like this much more persuasive if they were co-authored by the individual researchers who actually did the analysis. That would be a signal that they are willing to stand behind the comms team's characterization of their work. https://about.fb.com/...
@oliviasolon Olivia Solon on x
Why doesn't Facebook just publish the frickin research? https://twitter.com/...
@rmac18 @rmac18 on x
The blog post is also fundamentally in conflict over the research. Facebook says that actually the research shows that 11/12 metrics showed that Instagram was good for teens. But later down it suggests the results are unreliable. Well... which is it? https://twitter.com/...
@seanrobinson1e4 Sean Robinson on x
When you have a documented history of lying and you don't show the data behind your self-serving claims, don't expect to be trusted. https://twitter.com/...
@jesselehrich Jesse Lehrich on x
Facebook brags that only 1 out of every 100 teens has had suicidal thoughts because of Instagram. https://about.fb.com/... https://twitter.com/...
@nkulw Noah Kulwin on x
I would say that sentences like “Body image was the only area where teen girls who reported struggling with the issue said Instagram made it worse as compared to the other 11 areas” are inadvisable but truthfully it seems like none of this shit matters https://about.fb.com/...
@andymstone Andy Stone on x
@KurtWagner8 Yes, this is the very slide the Journal quoted. If it's unrecognizable, maybe it's because the Journal didn't note that in 11 of 12 areas more teenage girls who said they struggled with that issue also said that Instagram made those difficult times better rather than…
@jamisonfoser Jamison Foser on x
@HowardKushlan @MikeIsaac I love that Facebook will tell you “we have the smartest people in the world doing all kinds of sophisticated research into our effects on things” and then when someone reports that research demonstrates bad effects Facebook will say “our N was small thi…
@jeffjarvis Jeff Jarvis on x
Right. Facebook has to learn that there is not trust without transparency. Make the arguments here with full release of the research. https://twitter.com/...
@jeffhorwitz Jeff Horwitz on x
Definitely curious why Facebook has restricted internal access to this research given that it's so positive for the company. https://about.fb.com/...
@jonfortt Jon Fortt on x
“Actually Cheetos are good for you” is about the worst take Facebook could have countered with, yet here we are: https://about.fb.com/...
@evelyndouek Evelyn Douek on x
We really need independent and transparent research into social media so we're not deciding if social media is good actually based on selective disclosure by two of the most powerful corporations in the world of what 40 teenage girls may or may not have said https://twitter.com/.…
@dlauer Dave Lauer on x
There is no “Facebook research” which can be believed. It's a fundamentally unethical firm with a history of fraud, as I detailed here: https://link.springer.com/... The only reasonable approach would be to release the data. & as a monopolist, that should be a regulatory requirem…
@kevinmdraper Kevin Draper on x
@MikeIsaac Also they knew the WSJ stories were coming for weeks (months?) and the best they have is...an unconvincing blog post a full week after publication?
@jamestitcomb James Titcomb on x
Question isn't whether Instagram is a net positive - lots of problematic things are net positive - it's why 10s of millions of teenagers find it makes them feel worse https://about.fb.com/... https://twitter.com/...
@matt_cochrane7 Matthew Cochrane on x
@BDubes82 The WSJ criticisms were based on these same studies. You can't write hit articles that cherry picks data. Good reporting should look at the entire picture.
@chrismessina Chris Messina on x
Reminder that last week's @TechmemePodcast Experience had a robust discussion about these and related topics with @justinhendrix @AuthorPMBarrett and @EmilyTav and research on Facebook you may have missed: https://pod.link/... https://twitter.com/...
@carnage4life Dare Obasanjo on x
I'd love to see corresponding research from Twitter, TikTok or Reddit from users as to if they felt better, worse or neutral about their lives after using the service. Without that baseline it's hard to reason about if these results typical or unusual. https://about.fb.com/...
@ashleevance Ashlee Vance on x
Facebook begins research effort on technology that could make slide decks available to the public https://twitter.com/...
@girlsreallyrule Amee Vanderpool on x
We should charge Facebook under RICO.
@fbnewsroom @fbnewsroom on x
This research, like external research on these issues, found teens report having both positive and negative experiences with social media. We do research to find out how we can improve the experience for teens, and our research has informed product changes and new resources.
@jason_kint Jason Kint on x
Seems to be the Bill Barr strategy. Release an out of context and deceptive summary ahead of time to plant your headlines then release the details once public no longer knows the difference. https://twitter.com/...
@bostonjoan Joan Donovan on x
I remember in the 1990s, a lot of people were worried that airbrushing photos in teen magazines would give girls unrealistic expectations about their bodies. Girls were told in high school that we had to work on our self-esteem. The magazines kept printing such horror shows. http…
@lmatsakis Louise Matsakis on x
A lot of reasonable criticism to make here about transparency, etc. But I think what this also shows is that it's very hard to measure and make sense of how social media impacts society. It's obviously a mixed bag, because these sites have swallowed up so many cultural functions …
@sheeraf Sheera Frenkel on x
I can say this because we are on a group chat and he knows it to be true- my first thought was exactly the same as Ryan. Facebook releasing the research on teens and Instagram would be infinitely more convincing/impactful than this blog post. https://twitter.com/...
@benedictevans Benedict Evans on x
Coming out punching. It would be interesting to see the full reports - FB says selective quotation and shows the actual slides, but not the full PDF... https://twitter.com/...
@mikeisaac Rat King on x
i think this is one of the major problems i have with this rebuttal either you find the research and it's methodologies useful and instructive, or you don't, right? isnt it flawed to slap down sample size in some areas of the report as non representative but play up others? https…
@gilbertjasono Jason O. Gilbert on x
@MikeIsaac Does Facebook cause high schoolers to develop clinical depression? Sure. On the other hand, we also found that teens using Facebook were “75% more likely to be interested in Bitcoin”
@kurtwagner8 Kurt Wagner on x
Facebook just published a blog post that argues its research shows that IG is not as bad for teen girls as WSJ reported last week. They offered up this research slide, which is the same slide it claims WSJ cited in its piece https://about.fb.com/... https://twitter.com/...
@prestonmaddock Preston Maddock on x
If that was the overwhelming conclusion of the research, why'd it take Facebook 2 weeks of PR hell to write this blog? Why don't they just release the original research report? https://twitter.com/...
@somospostpc @somospostpc on x
Translation “what the WSJ published was so wrong it took a team of highly experienced crisis managers and facebook executives to come up with these empty words and, still, zero proof against what the WSJ reported” ok https://twitter.com/...
@scottmonty Scott Monty on x
Hey, look everybody: Facebook does corporate gaslighting! Imagine how much we'd respect or trust them if they acknowledged what we all know. https://twitter.com/...
@benedictevans Benedict Evans on x
It might be a good idea for Facebook to post the full reports that it says the WSJ misinterpreted, with annotations to explain all the shorthand and caveats. Then we could make our own minds up.
@eric_seufert Eric Seufert on x
Tristan Harris on Real Time with Bill Maher, discussing the Facebook Files reporting from WSJ: “This is a Cambridge Analytica-sized moment with these WSJ releases” -> So a wholly inflated narrative about something totally inconsequential?
@moonalice Roger McNamee on x
@sheeraf Releasing the research will most likely result in one of two outcomes: either confirming the worst suspicions about impact of Instagram on teens or confirming the worst suspicions about the quality of FB's research.
@mikeisaac Rat King on x
wild to hear facebook argue “Instagram mostly make teen girls feel better about themselves except in the area of body image issues”
@bdubes82 Billy Duberstein on x
@Matt_Cochrane7 This may be true but should you really put that much credence in a study done by Instagram on itself
@qwongsj Queenie Wong on x
I asked Facebook if there's a way to view the research. Spox: “We're sharing these decks with Congress in advance of the hearing this week and we're evaluating how we can release it to the public at some point.” https://twitter.com/...
@zamaan_qureshi Zamaan Qureshi on x
Facebook seems to want to go on the offensive on Thursday ahead of their Senate Judiciary hearing on this subject. But there are parents and teens who are asking for answers and they deserve to know what the full research shows. /4
@shiringhaffary Shirin Ghaffary on x
For a second I got my hopes up and thought Facebook might actually be sharing its raw research on Insta and teens — in full — but alas... https://twitter.com/...
@fbnewsroom @fbnewsroom on x
Here is a slide the WSJ did not publish, which shows that Instagram helps many teens who are struggling with some of the hardest issues they experience. The one exception was body image. https://twitter.com/...
@benlikestocode Ben Oberkfell on x
when you read this, replace the word “Instagram” with “cigarettes” https://twitter.com/... https://twitter.com/...
@matt_cochrane7 Matthew Cochrane on x
What? You mean the truth might be nuanced on a complex issue and not black and white? $FB https://twitter.com/...
@davidlazer David Lazer on x
Putting aside the particulars of this dispute between @Facebook & @WSJ, what is distressing here is that the @Facebook response isn't “of course, we have a lot more & better research on this important topic than a 40 person focus group & two modest sized surveys” https://twitter.…
@caseynewton Casey Newton on x
Stop telling us what's on the slides and just publish them so we can all read for ourselves https://twitter.com/...
@benedictevans Benedict Evans on x
The irony of calling the WSJ's set of internal FB docs ‘a new Cambridge Analytica’ moment is that CA was, in the end, nonsense - a hoax. None of the things people worried about had happened, but the idea it might have been true changed people's attitudes...

Chronicles