About the SA Blog Network



Blogging At The Intersection Of Psych and Pop Culture
PsySociety Home

Knowledge, Knowledge Everywhere: Do Social Networks Spread or Drown Health & Science News?

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

We live in an age of constant data. Between television, the Internet, and  our “real-life” social circles, society has never before had as much access to health and science news as we now enjoy — and it has never been so easy for anyone to access an entire encyclopedia of information about any health or science topic. Feeling sniffly and want to know if you might be facing pneumonia? Just WebMD your symptoms! Want to know more about what “Obamacare” actually entails? Check the Wikipedia page! Curious what people mean when they keep talking about the “God particle?” Just Google it!

But as much as we all might like to think we are living in an enlightened age of constant access to accurate and helpful information (ha!), the sad reality is that our “access” to this knowledge is far from equal. In fact, certain people are far more likely than others to be bombarded with health and science information (whether correct or incorrect) — and, conversely, certain groups of people might be unsettlingly more likely to be kept in the dark. In his new book, Social Networks and Popular Understanding of Science and Health: Sharing Disparities, Dr. Brian Southwell explores the various reasons why there might be such huge differences in the extent to which health and science information gets “spread around” in different populations, starting off with individual personality & demographic differences that exist between different types of sharers themselves, touching on important network characteristics like density & cohesion, and finishing up by describing particular characteristics of “shareable” messages, like rhetorical structure and emotional content.

Southwell opens up the book with the example of the milestone June 2012 Supreme Court decision upholding the major tenets of the Affordable Care Act (also known by many as “Obamacare.”) Although many people throughout the country (and the world) learned about this decision through mass media (like television, the radio, or the Internet), many heard of the decision directly from friends or coworkers — and many more received erroneous (or incomplete) information. Perhaps even more surprisingly to those reading this post, who are likely situated within networks that tend to share health- and science-related content (which is presumably how you have stumbled upon a Scientific American blogpost), some people were actually firmly situated within social networks that ensured they would not be receiving any information about this decision at all. Even though most of us have access to the Internet, television, and  other sources of news, why do some people receive more “shared” information about health and science topics than others? And why, along the same lines, do some people tend to be more likely to be sharers?

These are the questions that Dr. Southwell explores in this book. The process of reading it in its entirety is one of those  unsettling-yet-elucidating intellectual experiences where you know by the end of it that you have objectively learned quite a bit of information, yet you still somehow feel more confused, overwhelmed, and unsettled about the reality of health and science communication than you were when you started. With so many factors in play, how can we ever ensure that our messages are getting out and hitting the populations that need to hear them the most? Is it even possible to keep everything important in mind as we craft stories, articles, and information campaigns? Probably not. But Dr. Southwell has given us a good primer on how to start seriously engaging with these issues.

I actually had over a dozen questions for Dr. Southwell about the material covered in this book, and we’ve spoken about these topics extensively over the past week or two. In fact, I had so many questions for him (and he had so many answers in return for me) that I ended up deciding that there is too much interesting information and thought-provoking material involved in this book to cram into just one post! For the time being, I will now post a short snippet of my conversation with Dr. Southwell to give my readers a taste for some of what Dr. Southwell discusses in this book, but over the next month I will be posting a series of longer posts delving into more detail with him on some of the issues, topics, problems, and controversies involved in using social networking to diffuse health and science information.

Melanie Tannenbaum (MT): Since this is a book on social networks and all…what would you say is the “Twitter Version” of your book? How would you sum it up in 140 characters or fewer?

Brian Southwell (BS): People now think social networks make communication campaigns easy. That isn’t true, and in fact we might reify disparities if not careful.

MT: OK, so you’re saying that the improper use of social media for health campaigns can actually make existing socioeconomic disparities in important health outcomes even worse? With that in mind, what are your three “best practice” recommendations for effective social media usage & networking based on what you’ve learned through your research & the process of writing this book?

BS: First, invest in network infrastructure before you need to rely on [the network for information diffusion]. Community building can be inherently valuable. Second, meet people where they are. Frame messages in ways that resonate with everyday conversations for those who don’t work in academic settings every day. And third, set expectations appropriately and recognize that many health and science campaign interventions will continue to benefit from mass broadcast. Viral marketing alone is not likely to be a compelling way to gain all the reach that is necessary, but it does help.

MT: So, that last point actually ties into a question I had in relation to your discussion of reifying disparities. You mention that some “viral marketing,” like peer referral programs for services like free mammograms, end up widening existing health disparities. However, it seems (from my reading) that this is mostly due to pushing the “upper class” higher, rather than pushing the “lower class” lower. For example, you describe a study conducted through the Minnesota Department of Health, in which they attempted to advertise their free mammogram services through a “Refer-A-Friend” program. Although thousands of women took advantage of the opportunity to refer friends, socioeconomic status was a significant predictor of sharing — the women who most often referred friends to get free mammograms were well-off and possibly even already insured, and they were often referring other women who were also relatively wealthy and already insured (though, admittedly, often underinsured).

Are there specific examples of times that social media practices can actually harm disenfranchised/disadvantaged groups? Or is the widening disparity just a result of “the rich getting richer”? If so, are there any ethical concerns involved in not wanting this phenomenon to occur? After all, wealthy women receiving mammograms who otherwise might not are still women whose lives are being extended and whose health is important. Is it more important to focus on fixing health disparities, or on getting more women (regardless of class level) into the mammography room?

BS: I agree with your point: Serving even some with life-saving information can be better than not doing anything at all. I wouldn’t argue that we are necessarily causing damage with such efforts. But, I will say that we are not necessarily helping to the extent that some assume [based strictly on the numbers]. That said, there are also instances in which misinformation can spread through peer networks. In those cases, I worry that differences in networks might negatively impact certain groups, as it could lead to differences in the extent to which people are showered with misleading information.

MT: You talk about a ton of interesting studies and findings in your book. Do you have a favorite?

BS: It’s not necessarily a favorite, but that work you just referenced [with the Minnesota Department of Health on the mammogram "Refer-A-Friend" program] was definitely part of what got me interested in writing this book in the first place. That work simultaneously suggests some of the promise and limitations for using social networks in health campaigns. On one hand, it’s amazing that people even do it at all! You give them the chance to refer mammography to their friends, and many people end up doing just that. In fact, we’ve got evidence that people will refer their friends to get mammograms for free, without getting paid to make those referrals at all. They’ll just do it, they’ll just refer their friends, for no compensation or reward.

But on the other hand, that [program] is the first place where I also saw the potential for a disparity that could be really problematic. I always come back to that. Because, look: Here’s a state-level program that has embraced viral marketing. And it seemed successful. But then at the end we thought, gosh, what if we’re actually only serving those who are more well-off? So that’s something that I often return to in my head, when thinking about these issues.

MT: Did you learn anything that surprised you in the process of researching this book? What was it and why was it particularly surprising?

BS: As I was putting these ideas together, I really felt (and continue to feel) concern that folks are rushing into recommendations to use viral marketing and embrace social media. I think there’s a lot here for us to understand about networks and the possibilities, but I wrote this mostly because I was surprised by how widespread these recommendations are, and I wanted to speak a bit to that trend. One of the things that I found eye opening is that there’s a lot that you can pull together from various literatures to get at various aspects of network diffusion, but not a lot has been done on this empirically. You can organize research at the individual level (by studying individual behavior) or the macro level (by doing content analyses of the news), but the specific study of networks takes a specific set of skills, and it’s work that just hasn’t been done  as much. So it’s a little surprising, but also just a comment on the nature of the literature — and what is really left to be done.

Stay tuned for future posts with more information about Dr. Southwell and his work, including discussions of whether or not some people are really born to be “influentials” as hypothesized by writers like Katz and Gladwell, the differences between the types of messages that are “read” vs. the types of messages that are “shared,” and the role of fear (or other emotions) in health & science communication!

Dr. Brian Southwell is the program director for Science in the Public Sphere at RTI International in Research Triangle Park, NC. He has faculty appointments at Duke (with the Energy Initiative) and UNC (with Public Health and with Journalism and Mass Communication). His book, Social Networks and Popular Understanding of Science and Health: Sharing Disparities is available for purchase on Amazon. In the interest of full disclosure, I did not receive a free copy of the book in exchange for this review/interview, but I did receive a copy of the pre-publication PDF. All images in this post are courtesy of Dr. Brian Southwell, who granted permission for them to be used.

Melanie Tannenbaum About the Author: Melanie Tannenbaum is a doctoral candidate in social psychology at the University of Illinois at Urbana-Champaign, where she received an M.A. in social psychology in 2011. Her research focuses on the science of persuasion & motivation regarding political, health-related, and environmental behavior. You can add her on Twitter or visit her personal webpage. Follow on Twitter @melanietbaum.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 8 Comments

Add Comment
  1. 1. kirkMmaxey 2:20 pm 11/21/2013

    My experience with the online community is that it’s an incredibly narrowly selected, very liberal, literate elite. This comes from the results of an online survey about mating choices I have been running on my blog. Although the women (all self-recruited from twitter) were amazingly accurate in estimating the volume of a male testicle (to within about 5%)they were also virtually infertile – with scarcely one living child per 10 participants. This puts them in an outlying centile on both counts. This is not to say that the online community is not made up of good people – they’re exceptional people. But they are certainly not in the middle of anyone’s bell curve.

    Link to this
  2. 2. RCWhitmyer 5:24 pm 11/21/2013

    Perhaps the type or placement of the survey is affecting the type people responding?

    Link to this
  3. 3. Melanie Tannenbaum in reply to Melanie Tannenbaum 5:30 pm 11/21/2013


    I’m not sure if you’re responding to this post or to Kirk’s comment above, but regarding the broader topic of the book, it’s a summary of many different studies conducted in many different samples all around the country. I hope to go into more detail on some of the specifics in later posts. I hope you keep reading!


    Link to this
  4. 4. tuned 9:29 pm 11/21/2013

    Social networks spread more disinformation than is good for real people.
    Case in point: social networks like Reddit identified the wrong guy as a Boston bomber. That guy was murdered within a couple of days.
    I wouldn’t use a medicine dose without consulting a Doctor, even if I saw in Web MD or the NIH websites.

    Link to this
  5. 5. Squish 10:25 pm 11/21/2013

    The recent IPO of Twitter has reflected the public perception that social media has enormous potential. There is a huge amount of data that can be mined and used. Hacking a Twitter account and sending a false signal can cause the stock market to lose billions of capital. A geographical map of Tweets can show where earthquake damage has been severe; we can visual global thought trends that could have financial correlations. So this perception is warranted.

    I believe that the assumption that it will all be for good is still unfounded. Reading this magazine, Wired, FastCompany, and other science or tech-oriented publications, I see an enormous amount of misinformation in the comments. This magazine seems to have less creationist noise, but there is a dominant non-empirical global warming denialist voice. On mainstream news sites comments often veer towards racism. On Facebook I see a lot of garbage. Young people do not tend to prioritize the vetting of sources. The left wing falls into its crowd; the right wing falls into its own echo chamber and the whole system becomes more polarized while only the “outrage industry” profits.

    I support a free marketplace of ideas where they can compete. However, it seems that this marketplace is not working efficiently – freedom is not delivering truth. This market is awash in prejudices, emotional decision making and insecurity-driven out-group hatred – these are things that may be second nature to us but yet not be healthy.

    (For those who think this is a paradox, think of the analogy of how we evolved to crave sugar, fat and salt in one context, but now suffer in a new context where we have an overabundance of calories; we may face similar problems due to our wiring regarding sex, greed, etc.).

    To distill the above: we humans are still caught up in our biases – and often social media can amplify them (such as spreading pluralistic ignorance faster than it would have spread otherwise). I think the key to fulfilling some of the promises and expectations of social media – and avoiding a future where we are deafened by more viral noise – is to take a meta-level approach: by exposing these trends and amplification processes and making them subject to study, scrutiny and a basis for informed action.

    To that end, thanks Melanie.

    Link to this
  6. 6. rkipling 12:30 am 11/22/2013


    Could you elaborate on the nature of the referenced informed action?

    Link to this
  7. 7. Southwell 8:41 pm 11/27/2013

    Certainly, social media offer both potential and peril, as you all adeptly note. It also is worth noting that the book explores social networks beyond those available online through social media as well; it isn’t just an argument about Facebook. See for even more information. Thank you for continuing this important discussion.

    Link to this
  8. 8. KatieBarden 9:56 am 07/21/2014

    I recently posted an article looking at what happens when a doctor spreads product misinformation and HCP social media monitoring could have picked up on the conversation earlier and stopped it spreading:

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article