mHealth Recap: Takeaways and suggestions for moving forward

It will be years before mHealth becomes as pervasive as those of us working in the mobile sphere probably think it can and should be. After listening to several presenters, speakers, panelists, colleagues, among several others, my overall takeaway is that there are some manageable hurdles facing mHealth.  A few key hurdles boil down to establishing foundations and standards to build and refine the tools and practices. Below I’ve summarized the themes and suggestions on how an entire community can address them as they move forward:

  • Operate and build within a multidisciplinary framework. For instance, several apps I saw have a questionnaire component. The survey research industry has established standards and practices for mobile surveys, which could be borrowed to optimize those specific aspects of an app. Likewise, behavioral scientists are working to establish their own foundation for computationally-enabled theoretical behavior models (as I discussed in my recap of Day 2 of mHealth) in response to the new types of new technologies offer, and the volume and speed at which they can be captured.
  • Strive for seamless processes in mHealth. The word “interoperability” was mentioned several times throughout the conference. It’s important to keep in mind the role of APIs, and communication between apps, mobiles devices, wearable devices etc. Think of your app or device as a person – you want it to be an extrovert, and it should work well with others.
  • Strive for a seamless mHealth user-experience. mHealth shouldn’t be intrusive. Instead of adding to, integrate with whatever experiences you’re working within. What do designers Jony Ive, Dieter Rams, and Charles Eames have in common? Their designs were perfectly simple, and their creations seemed to just fit into our lives.
  • Remember: mobile devices are communication devices. There is a social aspect to health. As was noted during one of the sessions on wearable devices, there was a 10% increase in respondents reporting they workout with a friend. Without getting into the literature on social support mechanisms for health, I wanted to point out (as Mark Gorman did in an earlier session on aging) that social networks make mobile networks useful in many ways. Also, 78% of Facebook users are accessing Facebook via a mobile device, which should tell us something.
  • At its core, your mission is to decrease mortality and increase quality of life. That’s it.

mHealth Day 2 – Building a foundation for mHealth research

As I mentioned in my recap of Day 1 of mHealth, there is a reconciliation to be had between the speed of technological development, and the speed of science. After two days of talks, presentations, and panels at mHealth, the current state of mHealth seems somewhat held down by a lack of cohesiveness and standardization. By this I mean there are several great ideas, but the rate at which the community is building on and refining these great ideas is often surpassed by the evolution of technology itself. Much of what we’re seeing seems to be happening independently! One session today, however, seemed to take notice of this trend in mHealth, and presented a convincing approach to building a foundation for how we can standardize the utilization of mobile for health behavior research.

The current state of mHealth looks like the early internet. If you remember back to the early and mid-90s, the internet was packaged to us in a variety of ways. There were industry giants, small startups, wheels reinvented, novel approaches (some flourished, some faded), a growing consumer base, and a confused demographic that either just wasn’t interested or didn’t understand why this internet thing was important (that’s what the library is for!).  Overtime, the consumer demanded simplicity and more ubiquity as they realized the potential efficiencies and conveniences the internet offered, and a more structured form of the internet would emerge – Web 2.0. Now the internet is the mode for platform communication, which is more structured through APIs, and the internet is packaged in relatively fewer disparate pieces. It it now a user-driven experience that seems to have a fluidity that allows for the omnipresence we see today.

So what can mHealth learn from the evolution of the internet? For starters, mobile solutions for health should be approached with a uniform understanding of standards and practices, which starts with optimizing our understanding of the behaviors we’re trying to measure, influence, or otherwise evaluate in the mobile sphere. What these standards and practices can provide is a structure for connecting efforts and reducing duplication. Just as Facebook’s API allows for a simplified login process across the Web, understanding when, where, and how (i.e., best practices) to capture location data, for example, can create a foundation to optimize efficiencies and build and refine mhealth approaches.

One group of researchers attempting to develop such an understanding reported findings from a workshop on computationally-enabled theoretical behavioral models. I’ve summarized the summary of their workshop (it is by no means complete!), which I think serves as a great example of how the mHealth community could address such a complex technological shift in a way allows great ideas to support one another, rather than compete for space with one another:

Development and Planning the Workshop:

  • Behaviors are tied to several influencers (e.g., stress, social context, timing etc), which can now be understood in real-time through our digital footprints (e.g., social networking sites, sensors, cookies) that are often created on mobile devices, or other devices tethered by a mobile device. These digital footprints are essentially tracking patterns, and doing so across time (longitudinal) leaves contextualized data trails.
  • Current behavioral models are based on snapshots, or static interpretations of behavior. The challenge is to extend behavior theory to measure and evaluate these contextualized data trails.
  • To address this challenge, US and EU researchers were paired to address these challenges and come up with solutions to the challenges to extending behavioral theory to include a world where data is becoming virtually unlimited.

Outcomes of the Workshop:

  • To better create a foundation for measuring and evaluating contextualized data trails, computational models need to be developed and refined using rich, dense, longitudinal shared datasets, and build a vocabulary ontology.
  • Tools for data capture need to be made affordable and scalable.
  • Develop standardized quantifications of key factors (e.g., psychological state, motivations, values, emotions, social issues, physical context, behaviors, behavioral intentions).
  • Sensor fusion – develop and refine methods for inferring measures via proxy factors (i.e., combining a set of sensors to infer a difficult-to-quantify behavior from an algorithm that pulls from other conetextual and behavioral factors).
  • Vocabulary – concepts like habit and timing should take on a more universally understood meaning. Very quick measures that can be hard to capture via more traditional data collection, such as surveys, are now possible, which suggests our concept of things such as timing can take new forms.
  • Breaking down the differences in how these measures are interpreted in different fields.

The future:

  • It’s important to keep in mind that all data is health data. Data that is either tied to behavior either directly or indirectly is playing a role in our health.
  • Self-monitoring is an intervention.  You’ve heard the phrase “out of sight, our of mind.” Well, bringing health outcomes back in focus, and allowing people to see the connection their behaviors have to outcomes can also be an intervention. We see this with wearable tech.
  • Feedback can be crowdsourced! Yet another use of crowds, which is made possible by mobile technologies. For those of you familiar with Reddit, what if there was a tangible outcome tied to your meals either being upvoted (healthy) or downvoted (unhealthy) by all of your friends, or a set of experts? You just might change your eating diet! For more on crowdsourcing, check out Mike Keating, Bryan Rhodes, and Ashley Richard’s chapter in our new book Social Media, Sociality, and Survey Research.

You can read more about what this group is doing here.

mHealth – Day 1 Recap

With Day 1 of mHealth wrapped up, I’d like to share my thoughts on some underlying photo (1)themes and takeaways from the sessions and presentations I was able to catch.  Specifically, I was able to attend sessions on shaping care coordination with mobile tech, chronic disease management, and wearable tech and fitness devices, as well as a few short presentations at the NIH pavilion.

Following his talk on mobile clinical decision support tools, Robert Furberg mentioned the need for a reconciliation between the speed of technological developments, and seemingly glacial speed of science to adequately and appropriately adopt such develops. While there is a growing recognition for the need for tech integrations such as that mHealth seeks to promote, the scientific and healthcare communities are not always adhering to set guidelines, standards, and best practices in science and medicine when adopting new technology.

Today I was reminded of the impact mobile technology can, will, and is already having on health improvement, healthcare delivery, and healthcare costs (and the list goes on). From mobile tool kits utilizing tablets for administering patient questionnaires, reminders, and messages, to gaming approaches to health interventions, each presentation discussed how mobile can bring solutions to issues that were once too complex to efficiently and effectively address. The reason this is possible is because mobile is where a majority of us have consolidated several everyday tasks, including one of the most natural tasks: communicating.  And where we communicate is where we exchange information (i.e., data). In fact researchers from a variety of disciplines are facing the same reality (several of which we discuss in our new book as it relates to survey research).

So if we’re all in a agreement that a trend toward mobile solutions is a necessary one, what are some of the more immediate hurdles we face that are keeping us from getting to a world where, as one speaker put it, “mHealth is as taken for granted as the internet is today?” Well, in my opinion, a more ubiquitous mHealth means seamless integration as both an experience and a process.

Personally, given my work with creating and developing applications for data collection that utilize Facebook’s API, I was happy to hear the term “user-centered design” and “APIs” in some of today’s presentations. Specifically as it relates to mobile applications, I feel these are where the experience (user-centered design) and process (APIs) seams exists:

User-centered design: How can information be collected and utilized in ways that promote things such as application use, response rates, and decrease things like application fatigue? At the NIH pavilion in the Exhibit Hall (a cool way of adding on short presentations and demos BTW! It’s not candy or trinkets, but a giveaway that will last a lifetime – knowledge!), there was an underlying theme I noticed in the short 10 minute presentations – user experience (UX). Seamless integration into the mobile experience is critical to service delivery and the collection of data, which means objectives such as message delivery for behavioral interventions  (e.g., smoking cessation, exercise, or diet monitoring) can potentially hit on all cylinders except UX, and produce results that are ineffective. So if your application isn’t gaining any traction, consider how the end-user feels. And remember, superb functionality for a researcher does not always mean a superb app for the user!

In his presentation, Furberg noted that the decision support tool was well received by its users, and that to optimize effectiveness of such apps, researchers and developers should also concern themselves with APIs (application programming interfaces).

APIs:  Why are APIs important? Understanding API structures can mean the difference between optimizing systems communication and having a smooth transfer and utilization of data for all stakeholders, and a clunky system everyone complains about. When platforms can effectively speak with one another, researchers can efficiently and effectively obtain the information needed. This goes for any process that receives, disseminates, and utilizes data.  This could be a survey instrument collecting and sending data to a server, which then goes to a case management system where it is analyzed in ways that improve survey delivery, and in turn, data quality. It could also be a mobile app intended to collect patient data to successfully implement guidelines for physicians as they care for a patient, and having that data shared with administrators, and possibly even researchers for longitudinal tracking.

That’s all for Day 1 – I look forward to hearing and seeing what mHealth has in store for Days 2 and 3!

Market Research in the Mobile World – Horizontal Thinking

If mobile has arrived, where do we go from here? And if we’re tethered to our devices (as I suggested yesterday), how do far do we go? If wearable tech and other digital enhancements of our environment is the next phase, the answer is not too far.  But the extent to which insights can be had from a mobile device alone is still limited by the depth of their integration into everyday life. Sean Conry (Confirmit) pointed out context as a driving force behind device selection, particularly time (when or how long we intend to interact digitally), goals (what we want to accomplish), location (where we are), and attitude or state of mind (people tend to have more personal attachments to smartphones). So every utilization of technology has its limitations, which means there’s a broader view to be had.

Marie Ng (Millward Brown) demonstrated just that (a 5 week study with only 1 week of mobile diary data collection supplemented by data capture via online surveys) and suggested that mobile allows researchers to get more granular, but it also compliments other types of research, both of which a multi-platform approach can accomplish. Moving beyond a multi-platform approach, as Catherine Winfield from the MIT Mobile Experience Lab described, there is an entire ecosystem surrounding the mobile experience that we need to consider. In this sense, mobile is the touch-point where our experiences converge. As we digitally enhance our environment and the bridge the gap between the tangible and digital, more and more of our experiences will converge into a single space – likely a mobile device, considering its ubiquity in our everyday lives.

I think the lesson to be had here is that a narrow focus can certainly uncover nuances, but horizontal thinking, as Catherine Winfield put it, allows researchers to move discovery beyond understanding the state of things, to discovering the next big thing. Researching phones and apps only gives you phones and apps, but researching tablets, apps, and cooking gives you Chipchop – a digitally enhanced cutting board that measures ingredients as you cut them (very cutting-edge).

Market Research in the Mobile World – An Emerging Theme

A common theme I’ve seen throughout this afternoon’s presentations is that we’re tethered. If I were to make an observation about 2013, this is the year researchers are realizing people are tied to their devices (make note Reg Baker!). We’re starting to see a transformation from a device with useful functions, to humans that don’t function the same without the device. The implications of this span from the possibilities (and ethical concerns) of passive data capture, to the competition for the end user’s time, not to mention the relationship between the two. After all, one way to avoid competing for a study participant’s time is to acquire data unobtrusively.

Roddy Knowles of ResearchNow demonstrated just that – tapping into behaviors such as app use and location verification to better understand what shoppers are doing with their smartphones while they shop at certain stores. But as Robert Clancy and Lisa Wilding-Brown of uSamp pointed out, mobile devices in themselves are, in a way, naturally less intrusive – competing for time isn’t so obvious when that’s where people spend time (the alternative being non-mobile modes and keeping people from their tethered devices altogether).

Then again, sometimes it’s very obvious. As Deborah Powsner (SessionM) put it, mobile ads that appear outside of the user typical flow are vampires – sucking the time away. Likewise, asking study participants to take part in any task, be it active health monitoring or taking a survey, we’re indeed asking for precious time. And with the countless distractions smartphones have to offer, true downtime is a rare commodity.

What does this all mean? Well, when passive data capture is an acceptable alternative, it means striking a balance between informed consent and truly passive, or other less intrusive data capturing. During the panel discussion on setting standards for a new era of data collection, the notion of passive data collection was discussed, but our understanding of public perceptions of privacy are still confused. To what extent do we want to perceive ourselves as being tethered to our devices? And to what specific ends and under what circumstances are we willing to allow such access to our lives?

Market Research in the Mobile World – Recapping Day 1 Keynotes

It’s the morning of Day 1 of the Market Research in the Mobile World (MRMX) conference,and the keynote speakers have done an excellent job explaining the context of a mobile world. To warm up the crowd, conference MC Mary Evans Kasala from Capella University asked everyone to engage in some show and tell – show your neighbor your mobile phone and tell them why you like it. My neighbor had a flip phone he’s had for 10 years and an iPod he used for his mobile internet. For the record, his wife has an iPhone, so he wasn’t exactly shunning the smartphone revolution.

The mere fact that everyone readily had a mobile phone to show and tell (some of us with multiple mobile devices) was a clear indicator that we’re tethered to these devices. And as Guy Rolfe of Kantar Mobile pointed out, some of us are so tied to them that checking picking up our phone was probably the first thing we did in the morning. When we’re tethered to our devices so much so that most of us check our mobile devices before we accomplish anything else in the morning, or we email or text someone in another room rather than physically walk to another room to speak with them, you might say we’re living in a mobile world. But what does that mean for researchers?

Jeanine Bassett (VP of Global Consumer Insights at General Mills) suggests that, for a variety reasons, mobile is THE way to do research in many respects. If not for trends in mobile adoptions, or that Millenials have little to no grasp of more dated modes of communication, it’s the versatility of mobile devices (photos, internet, email, video etc) and the ability to capture people in the moment. In fact, that’s what companies like General Mills and Electronic Arts are doing. Lisa Spano, Head of Consumer Insights at EA, discussed how in-app surveys are providing more context appropriate and expedient evaluations of mobile games by embedding surveys within games (sometimes getting thousands of responses overnight!). But is that the extent of mobile research? Surely we’ll develop more creative ways to utilize mobile phones in research, but to what extent is further research and development into mobile methods going to pay off when other technologies are on the horizon? This is the very question General Mill’s is asking, and why they’ve decided to sunset their mobile research agenda after 2014.

So why, exactly, is General Mills sunsetting their mobile research agenda when there is clearly work to be done and insights to be had? Perhaps, as Guy Rolfe suggested, it’s because wearable tech is on the horizon – indeed, the next big thing is on its way! While mobile research methods are certainly not drawing to a close, research on a global scale will continue to require methods that integrate the technology we’re actually using and the ways in which we use such technology. Increasingly, this includes the latest and greatest in the tech world, as developing countries tend to bypass other bridge technologies that most of Western countries experienced (e.g., landlines, dumb-phones, wired internet access etc), and move straight to the fun stuff!

Top 5 Must-reads for Research in a Mobile World

In preparation for the Market Research in a Mobile World conference next week in Minneapolis, I decided to share my top five must-read articles and blog posts about the mobile world today (including a couple from SurveyPost itself). The more I went back to read older articles, the more questions I had, which turned into more searching. My goal was to create a top five list that provides a well-rounded overview of what it means to do research in the mobile world, whether academic, non-profit, or market research. I’m certain there is something missing, so your input is welcome!


1. Pew’s Report on Smartphone Ownership in 2013 – Aaron Smith

Given the continued growth and the current breadth of smartphone ownership in the US, it’s no wonder why we’re so concerned with the mobile world. As Pew’s reports suggest, smartphone ownership has seen a steady increase over the past few years, while those with “other” phones or no phones have seen a steady decrease. A majority of Americans now own a smartphone, and this report breaks down adoption rates and platform trends across several demographics, including age, income, and education. To begin to understand the mobile world in 2013, start with this report.

2. Cell phone data mined to create personal profilesHiawatha Bray

This recent Boston Globe article is a good review of the data-capture implications of widespread adoption of smartphones. To summarize, cell phones are data capturing machines, and the possibilities seem endless. Privacy concerns aside, the unobtrusiveness of passive data capture in our everyday lives is appealing to me as a researcher. This approach shifts the burden of research from the participants to the researcher that analyzes data, which eliminates some biases of previous modes of data capture. While new biases are certainly introduced (i.e., a personal profile will never be complete, thus measurement error will always exist), the idea of developing personal profiles via passive data capture, and the potential to do so on a massive scale have huge implications for researchers of all walks of life. This article provides overviews of what some researchers from Harvard, IBM, and the MIT Media Lab are doing now, and where they see mobile research in the coming years. Although not a study involving mobile technology, Deb Roy’s research tracking how language is learned by monitoring his son’s speech (presented here during a TED Talk) is just a glimpse into the discoveries made possible by streams of passive data capture.

3. Over-reporting of Mobile Use – Methods for ImprovementAshley Richards

Ashley’s discussion on SurveyPost of a Boase and Ling’s article about measuring mobile phone use provides additional insights into why cell phone use is over-reported, and suggestions on how to improve our measures.  As is mentioned in her blog post, over-reporting cell phone use definitely warrants more research. There are several social commentaries to be had here, considering our attachment to mobile devices and our apparent inability to accurately perceive their immersion into our everyday lives, but that’s a different post for another day!

4. The history of Mobile in 8 easy Google searchesReg Baker (The Survey Geek)

It’s just an in image, but a picture is worth 1,000 words, right? In this blog post, Reg Baker shows us that mobile continues to surprise us, so much so that each year we proclaim it to be “the year of mobile” in one respect or another. What seems apparent is that mobile is a mainstay, perhaps not in its current state (it’s technology – it’s ever-evolving), but we’re constantly coming up with new ways to harness mobile (e.g., mobile panels, mobile advertising, mobile media, and mobile marketing). So what will 2014 and beyond be? Will there be a year of the mobile profile when comprehensive data capture exhibits some sort of positive and welcomed benefit for mobile users (e.g., downloading apps to build health profiles for doctors or insurance companies)? Is the year of mobile connect on the horizon, where we’re connecting to Google Glass or an iWatch to further augment real life?

5. Making the Switch to Smartphones: Tips for Managing RiskAmy Hendershott

In this blog post, practical implications of making the switch to smartphones as a platform for survey data collection are considered. While we like to think the entire world is on board with the concept of mobile research when it’s the preferred method, there are several considerations outside of trends and the latest and greatest mobile feature that should be made when deciding how to approach mobile research. Here, Amy Hendershott discusses several considerations for transitioning a survey data collection effort to smartphones. Although it would be nice to have a data collection app running on the latest operation system on the latest smartphone, the reality is every survey is constrained by budgets, time constraints, and human ability and error.

How Simple Tech Integrations Can Improve Field Survey Accuracy and Reduce Costs

As any good methodologist would attest, ensuring accurate estimates is the primary goal of survey research. Typically, the challenge is to obtain the highest quality for the resources available.  For several years I have been investigating solutions to increase data quality while simultaneously reducing costs on the North Carolina Seat Belt Survey. This survey is conducted annually to calculate statewide estimates of seat belt usage and other traffic safety statistics. Each year, we’ve introduced tech-related elements to the study intended to improve some aspect of the data collection process, and over the past five years we have seen a reduction in data collection-related costs each year.

The sample for the Seat Belt Survey is comprised of road segments, from which we select a location to collect data on vehicles, drivers, and passengers. Where paper maps and in-person site visits were once our only tools and methods to navigate to sites and perform quality control checks, the integration of technology into data collection procedures and management has afforded us several opportunities to reduce costs and improve data quality. When we initially set out to integrate technology into data collection a few years ago, our goal was to reduce navigation errors and travel times that were often a result the types of errors that have always accompanied written directions and paper maps (e.g., miscalculated distances and driving times, incorrect street names, and outdated routes). To remedy this, we introduced GPS units programmed with the longitude and latitude of each sampled site.

In more recent years we have transitioned data collection itself from paper and pencil to tablets. This transition allowed us to combine both navigation and data collection on one device, as well as provide a communication mechanism (email) and a location to store and access digital versions of all project-related documents (e.g., manuals). This year, we’ve made upgrades to our data collection application that are allowing us to improve data quality and perform quality assurance tasks by analyzing data as they come in. Using the longitude and latitude of each sampled site we previously used to pre-program navigation routes for data collectors, we programmed spot checks to ensure data is collected within a reasonable distance from the site. As data collectors begin entering data for a site, they are informed of their distance to the site. Additionally, each observation made is accompanied with a location stamp. Together, these measures ensure data are collected at the right place, and with a timestamp, the right time.

We anticipate that these programmed spot checks will reduce error, ensure the location of data collection is accurate, and reduce data cleaning costs on the backend. In April and May of this year, we conducted the Nighttime Survey using location spot checks, and no location discrepancies were found between sampled locations and the geo-tagged observations. Interestingly, at the outset of data collection, several field staff called me while in the field to ask if their distance from the site was reasonable, which was somewhat reassuring!

A Birds-eye View of the 2013 AAPOR Twittersphere

As Joe Murphy mentioned in the preceding post, Twitter can provide a unique and efficient glimpse into conference dialogue that may be difficult to capture in other ways. We can measure when conversations occur between tweeters, the subject of those conversations, and other interesting patterns such as the most frequent words appearing in Tweets or the most popular Tweet (as indicated via retweets). There’s one analytic possibility, though, that’s frequently overlooked: social network analysis (SNA).

As I discussed in my poster presented at this year’s AAPOR conference, something inherent and potentially very useful to researchers is the network aspect of social networking sites. Inspired by a growing Twitter presence among the AAPOR community, and Joe’s analysis of the AAPOR Twittersphere, I decided to put some of the SNA concepts I discussed in my poster to practice. Specifically, in my attempt to examine public opinion in a different light, I asked the question “who is talking to whom?”

For those of you familiar with SNA, or are just curious as to how the measures I discuss are actually quantified, I’ve pulled the equations and explanations from my poster and inserted them for reference. In the first image below, a birds-eye view captures what the AAPOR Twitter conversation looked like. By conversation, I’m specifically referring only to Tweets where one user Tweeted to another using the hashtag #AAPOR. For instance, if (hypothetically speaking) Joe (@joejohnmurphy) Tweets to me “hey @AdamSage, your poster was A-M-A-Z-I-N-G #AAPOR,” the link connecting Joe to me would be directed from Joe to me.

At first glance, this birds-eye view looks kind of cool (or like a big mess depending on your perspective) – but what does it mean? Each dot, or node, represents a Twitter user that at one point used the hashtag #AAPOR and mentioned another Twitter user in a Tweet. Each line represents the association that’s created when such a Tweet occurred. The degree of each node is number of different connections made as a result of these Tweets. In the AAPOR Twittersphere, the average degree, or the average number of people to whom a given Twitter user is linked via these dialogues, is slightly over 2 (2.017). This is important because an average degree of 1 would mean we’re not a community, but rather a group of 1 on 1 dyadic cliques, which makes ideas difficult to spread. While these did occur, I excluded them in these graphs and focused what is called a giant component in SNA, or the largest group of connected individuals in a network.

Excluding those conversations occurring outside of the giant component, the diameter of the entire conference Twitter conversation was 8. In other words, the furthest any two people were away from each other during this Twitter conversation was 8 steps. Imagine yourself at the largest Tweet-up in AAPOR history consisting of those AAPOR attendees participating in a dialogue on Twitter (using #AAPOR of course). As people traverse from conversation to conversation, everything you say would be no more than 8 conversations away from anyone in the room. So that rumor about my awesome poster could theoretically spread quite quickly. Hopefully it’s starting to become clear how great papers and keynote addresses can become popular Twitter topics in a given year. Great papers resonate and travel quickly; great keynotes are viewed by a lot of people and travel even faster.

But wait, it gets better. Ideas spreading among the AAPOR community wouldn’t traverse the network as you might think. Looking at the overall graph below you will notice several color-coded groups of tweeters. Each color represents a community, which is essentially defined as those tweeters who are mentioning each other in their Tweets more than others. For instance, community around @AAPOR (the green colored node with several links in the top middle) does not include many of the users with high betweenness and closeness. Betweenness measures the strength of one’s connection to the entire network (i.e., how interconnected or embedded one is to the entire network), and closeness measures one’s distance to the overall network (an average distance). They are measured as:



As you might expect, these people tend to drive conversations because they have many connections, which when analyzed using certain measures of network centrality, can reveal likely members of such conversations. For instance, just by measuring mere mentions within Tweets, clustering coefficients (defined as a given groups number of connections divided by the number of possible connections) allow us to distinguish “communities” from the rest of the network. Because this graph is directed, meaning I can tell if someone is doing the Tweeting or being Tweeted at, I know a majority of @AAPOR links were created by people mentioning @AAPOR in a Tweet, rather than @AAPOR mentioning others. The community around @AAPOR (the green colored node with several links in the top middle circled in red) does not include many of the users with high betweenness and closeness because @AAPOR’s propensity to draw Tweets to it naturally steals the thunder of everyone in close proximity.

While I haven’t analyzed the content of the conversations occurring among these communities, my guess is there are general topics (e.g., polling, elections, survey methods, statistics etc). After all, we tend to Tweet about what we’re interested in. In fact, just looking at the categories identified in Joe Murphy’s post, Tweets coming from @AAPOR and the highest frequency Tweeter (who I’ll leave unnamed, but I have circled in yellow) were generally categorized differently with the former much more often covering general conference information and the latter results from paper sessions.

While this is a quick analysis of the Twittersphere, these analyses can reveal characteristics of conversations and conversation participants that can potentially lead to other discoveries, such as the propensity for ideas to spread (or fade) among certain groups, identifying which individuals (and their behaviors) are key to group cohesiveness, and how we can eventually focus efforts to create a more robust environment for innovation, e.g., creating a hashtag to link people and ideas in the survey and public opinion research Twittersphere, much like market researchers use #MRX.

Tracking 2013 AAPOR Buzz on Twitter – Just for Fun!

For those of you who have attended an AAPOR conference since 2010, you may have seen the Twitter Board that RTI sponsors annually (see below). Over the past few years, many of us at SurveyPost have focused on discovering and exploring the potential of social networking sites (e.g., Facebook and Twitter) in research. Although we will be presenting on these topics once again this year, we will also be putting some of what we’ve learned about analyzing Twitter data to good use.

In cooperation with (and great thanks to!) AAPOR planning and executive committee members, we have developed a series of fun questions that we will field via Twitter. Aware of the limitations of using Twitter data, and knowing this is the first effort of its kind at an AAPOR conference, we hope many of you will find the results and analyses we do interesting and, well, fun!

If you’re excited to participate (which we know you all are), simply follow the #AAPOR hashtag. For these particular questions, we are also using the hashtag #AAPORbuzz. You’ll see questions coming from our Twitter handle @SurveyPost, so be sure to follow us. We look forward to your Tweets!