New paper! Multi-year controlled #SciComm experiment finds that tweeting about published papers does NOT lead to an increase in citations of those papers. Led by @trevorabranch.bsky.social , featuring me @melissacmarquez.bsky.social @solomonrdavid.bsky.social @danirabaiotti.bsky.social & others 🧪 - ThreadSky

whysharksmatter.bsky.social • 343 days ago

New paper! Multi-year controlled #SciComm experiment finds that tweeting about published papers does NOT lead to an increase in citations of those papers. Led by
@trevorabranch.bsky.social , featuring me @melissacmarquez.bsky.social @solomonrdavid.bsky.social @danirabaiotti.bsky.social & others

🧪

Comments

seshnadathur.bsky.social•343 days ago

True enough, in my experience. But tweeting about one of my papers probably played a role in one of my mutuals learning about my expertise in a topic they needed help with, and providing this expertise was sufficient for me to become co-author on that paper ...

phillevin.bsky.social•95 days ago

Interesting! I’m glad y’all did this. I think it will help sharpen our thinking about diverse forms of “impact”.

tofu-irl.bsky.social•343 days ago

Oh thank goodness

danielbolnick.bsky.social•343 days ago

But what about skeeting/ reskeeting

drfishsg.bsky.social•86 days ago

There's a citation in there somewhere.

tetonkali.bsky.social•343 days ago

But that doesn't count a lay person like me reading and learning from the parts I can understand. I love #SciComm but I like to get more than the press release or blurb. I learn for abstracts, discussion, and conclusions. Some times a little sometimes a lot.

toniklemm.bsky.social•343 days ago

I also think scientists who do Twitter/scicomm don’t just aim for citations as measures of success of their research, but want to support policy or public understanding.

toniklemm.bsky.social•343 days ago

As a researcher I’m not surprised. I go to social media to follow public trends and conversations in my field of research, not to find the latest science.

danielread.bsky.social•343 days ago

Can you rule out whether authors tweeting about their papers has an impact?

whysharksmatter.bsky.social•343 days ago

If me posting about a paper to 80,000 followers doesn’t increase citations, someone posting to a few dozen followers won’t either.

danielread.bsky.social•343 days ago

I think that remains to be established. The researcher posting their own work to people who might already be interested in that work might have an impact that your posts do not have.

whysharksmatter.bsky.social•343 days ago

An impact to whom? If no one is following you and no one sees your posts, it doesn’t much matter what’s said in them.

danielread.bsky.social•343 days ago

I think you have told me that you cannot answer my question. And that is completely fine by me since no study can answer every question.

whysharksmatter.bsky.social•343 days ago

I prefer to base my social media strategy on data and evidence. Others are free to base theirs on wishes and vibes if they’d like. I hope it works out for them.

whysharksmatter.bsky.social•343 days ago

You may have heard of "Twitter," lots of scientists talk about their work there and I'm sure there's a bright future for it. Anyway, years ago, there was a discussion on academia twitter: Does sharing your papers on social media lead to more people eventually citing your paper?

whysharksmatter.bsky.social•343 days ago

Several studies found a correlation, the most-shared papers on social media also had lots of citations. But that doesn't mean that tweeting about a paper *caused* more citations.

whysharksmatter.bsky.social•343 days ago

The alternative hypothesis, termed "good papers are good" by a colleague, claimed that some papers are more likely to be cited and tweeted about.

thinkingsideways.bsky.social•343 days ago

The idea that some papers are cited merely because they are well-researched and well-written is exactly the sort of elitist academic attitude that we've come to expect.

Theories should live or die in the marketplace of ideas based upon empirical standards such as the popularity of their authors.

pennywise.bsky.social•343 days ago

I think I've identified some sarcasm in the wild. I'm really hoping that's it, anyway

drmichellelarue.bsky.social•343 days ago

... same

whysharksmatter.bsky.social•343 days ago

It’s elitist to say that tweeting about your papers doesn’t lead to more citations?

K

thinkingsideways.bsky.social•343 days ago

Sorry. I should have added an /s, but I ran out of characters.

whysharksmatter.bsky.social•343 days ago

So we (a team of high-profile twitter science communicators led by Trevor) did a controlled multi-year experiment, randomly tweeting about or not tweeting about papers from a variety of journals, and measuring their ctation count years later.

whysharksmatter.bsky.social•343 days ago

We found that there is no statistically significant correlation between how much a paper is tweeted about and how much it is eventually cited. Scientists: there are lots of good reasons to talk about your work on social media, but increasing citation counts is not one of them.

supersciencegrl.co.uk•343 days ago

This is great!! I suspected this (as did most people I guess) but now we know 😍 thank you for doing this work!

mcusolito.bsky.social•343 days ago

Interesting.
I'm not a scientist (& I'm no longer on Twitter) but I DO read certain papers when I see scientists sharing them. (Sometimes I skim). But you reach me when I might not have seen it otherwise. I'm an author of science books for kids through young adults & science communicator. #SciComm

jameslonghurst.bsky.social•343 days ago

“Anyway, . . .”

whysharksmatter.bsky.social•343 days ago

Press release: https://fish.uw.edu/2024/03/three-year-experiment-finds-no-citation-bump-from-twitter-promotion/

envpolicycenter.bsky.social•86 days ago

interesting but needs replication in other fields. But even if replicated, I think the #scicomm benefits are not about citations. It is about linking knowledge to other types of interested parties, journalists etc.

whysharksmatter.bsky.social•86 days ago

Lots of people, for years, said "you should be on Twitter sharing your research because it will increase your citations."

We tested that and it is not true.

envpolicycenter.bsky.social•86 days ago

With all due respect, I think your conclusion of "not true" is overstated. Even your lit review points out a set of contradictory findings, which your paper now joins. Your study is limited to a relatively narrow field, and also mostly highly cited authors (possible ceiling effect).

envpolicycenter.bsky.social•86 days ago

And although not "significant" your effects are still positive and seem pretty big, so it depends on how much you weight you give to p.<.05. So my conclusion still remains "citation bump"...maybe, maybe not. Other benefits...such as collaboration, extension etc...probably...

whysharksmatter.bsky.social•86 days ago

Absolutely many other benefits

Comments

Posting Rules

Reply