Forbes: Big Data And Health Care Get Engaged

Ben Kerschberg,  Forbes, June 12, 2012

The Commonwealth of Massachusetts and MIT announced recently a joint initiative to tackle the enormous opportunities presented by Big Data. “The hour of big data has arrived in Massachusetts and it’s a very, very exciting time,” said Democratic Governor Deval Patrick. With a particular focus on creating jobs and promoting innovation, the initiative will be able to address some of the many questions posed by Big Data: How can we work smarter? How can we improve our processes? What are we leaving on the table?

The answers to these questions can be found precisely where the rubber hits the road: at the intersection of Big Data and business intelligence (“BI” or analytics). Moneyball, or the application of BI to uncover and leverage previously hidden or underappreciated data, is a concept already familiar to the Commonwealth. Almost a decade ago, former Boston Red Sox General Manager Theo Epstein borrowed its now-famous principles from Oakland A’s GM Billy Beane to bring Boston the first of two World Series crowns.

For Danvers, Massachusetts-based Eliza Corporation (“Eliza”), understanding Big Data and health care has always been a top priority. With over 700 million interactions with health care consumers under its belt, Eliza is mining its data to address through Health Engagement Management one of the most important questions in the health care space: How do we get Americans to make healthier choices? In order to understand this landscape more clearly, I caught up recently with Eliza’s Founder and Chief Visionary Officer, Ms. Alexandra Drane.

Alex, what makes it so difficult to engage health care consumers and modify behavior?

Often, it is what makes you strong that makes you weak. We in the health care space are very mission-driven people who want to change the world by helping people get healthier. That’s a good thing. But, because we have this obsession with health, we can project that onto the rest of the world, and fool ourselves into thinking every individual out there thinks about health as much as we do. Of course, that is not usually the case, yet the health care space still often approaches people from that framework.

Put this approach in perspective next to health care’s competition – the food, tobacco, and beverage industries, which spend 200 times more in U.S.-based marketing than the health care industry. They have a much easier sell – sit down on the couch, eat a bacon double cheeseburger, and have a third glass of wine. Meanwhile, we’re saying, “Get up, don’t press snooze, go exercise, eat carrots instead of potato chips, and get your colon cancer screening.” We have a different problem to solve and we have to overcome a lack of both information and inspiration to get there.

The health care industry also is still very academic. We’ve grown up with peer review and double-blind trials in controlled environments that don’t necessarily reflect how real people think or live. Traditionally, the materials the health care industry has sent out to people—an informational brochure with a picture of a diseased kidney, for example—have had a certain finger-wagging dynamic to them and that can put people off. Individuals may feel condescended to on a problem that they’re really not inspired to address anyway   – and since changing health-related behaviors is hard, we can be putting ourselves at a disadvantage out of the gate.

At Eliza, we know we have to push past that, and we do so successfully – even with some of health care’s biggest challenges. Let me give you a quick example. Patients who receive Eliza follow-up outreach after a hospital stay are 25% less likely to be readmitted to the hospital within 30 days of being discharged. Stats like these show that when Health Engagement Management is done correctly, we can have lasting effects on lives, and the health care system.

Where are you able to have the most impact? What segments of the population? Which diseases?

Overall, we’ve recognized a population-wide impact because we know that if you listen hard, people will tell you what they need to be inspired to make change.  Sitting on top of more than 700 million interactions gives Eliza an opportunity to do a whole lot of listening, and we tailor our health engagement programs to those learnings in different ways—but always with a singular focus on improving an individual’s happiness, health, and productivity—regardless of their demographic, regardless of their condition.  What we see is that if the message comes at the right time, using the right tone, and in the most appropriate medium, people will pay attention, and they’ll act accordingly. Some quick examples of our impact include:

·       increasing the number of patients who get their recommended diabetes screenings by 76%;

·       increasing colon cancer screening rates in a Medicare population by 137%;

·       quadrupling participation in an online smoking cessation program; and

·       we’ve even seen engagement among young, generally healthy individuals boost their perceptions of their health plans’ brands far beyond what more expensive, traditional advertising campaigns could deliver.

Understanding how engagement can vary by population segment is an important ingredient to effecting change – and segments don’t always reflect the big broad swaths that we sometimes cut, for example, grouping people simply because they have a condition. 

Specifically, let’s say that I have congestive heart failure (CHF). Naturally, that’s more likely to be top of mind because I can feel the negative consequences of this condition all day, every day. On an engagement level, we are successful if we can get people to start thinking about new approaches to their challenges. But in terms of creating outcomes for that particular population, it’s more difficult because they can be really, really sick. So, the success metric for the CHF population in particular might be to help them to stay home, to keep them out of the hospital, to respect their feelings and to keep them feeling as physically and emotionally positive as possible.

Now, let’s look at the other end of the spectrum – a young adult with high cholesterol. Most people with high cholesterol in its early stages don’t have any symptoms at all. Their lives are not defined in any way by their condition, and so, naturally, it’s more difficult to engage them in conversations about their health.  Now, the fix for high cholesterol you could say is actually somewhat simple – just get them to take medication. But the process of telling someone that they’re going to take medication—and maybe for the rest of their life—is a harder sell for the average American who wants to see him or herself as incredibly healthy. We’ve all seen the numbers showcasing how abysmal adherence rates are – hovering around 50% for most conditions. That’s because while the idea of telling someone to take a medication sounds easy, the reality of being the one who actually takes it every day—particularly for a condition that is asymptomatic—is quite the opposite. So while the engagement process for someone with high cholesterol is more challenging, and we might think that the treatment is simple, the actual process of getting someone to adhere to taking their medication in this situation is likely to be more difficult. But not impossible.

We tested many approaches and learned that a multimodal approach tailored to individuals’ barriers can be very impactful, in fact we saw a 16% increase in Statin medication adherence over a six-month period.  Through that testing process, we also learned things like 58% of people think they can stop taking their Statin once their cholesterol is reduced.  That is data we all should be paying attention to.

We also can share that we’ve found that the most defining and predictive variables driving health behavior are frequently not even associated with disease (as traditionally defined) in the first place. We call these The Unmentionables.

So net, net, it’s tough to offer a brief answer to your question about the success of different campaigns because they’re all so unique. We never presume that a broad-sweep solution is going to work on an individual level. That’s precisely why we come in and try to encourage an individual to share what’s specifically going on with them. We know that individuals need to be supported in ways that are tailored and relevant to them.

Eliza has moved beyond just speech recognition technology. But how does Health Engagement Management make up the gap that remains between it and speaking with a nurse, for example?

That’s a great question. Technology is never going to be able to replace a nurse, or doctor, or social worker, or anyone else who has devoted their life’s work to helping people better manage their health and wellbeing. Period, end of story. And so Eliza’s job, because it cannot be a human, is to do the very best we can with the tools we do have – tools that enable us to recognize how to help each individual to the extent they need—and to link those individuals back to the supportive infrastructure that exists—including humans. For example, we can figure out on a mass scale the specific individuals who would benefit the most from having a conversation with a nurse or a doctor or a behavior specialist or a pharmacist, and in real-time we can connect them with those resources – resources they might not have known how to access otherwise. Surrounding a care team with trusted and time-tested technology that extends their reach outside the doctor’s office and creates more constant feedback and interaction is a key part of driving sustainable health behavior change.

After more than ten years of reaching out to people, we listen hard to the data and the emotion –you can hear the opportunity to help. And it’s not always a call – it might be an email, a text, a mobile app, or a pointer to a web site where individuals can connect with people ‘like them’. Eliza captures and respects the ability to tie everything back to the individual. Every communication, every utterance, every claim, visit, or mailing is tracked in a longitudinal way so we can see what each person responds to, specifically and individually.

For example, we can screen for indications of highly co-occurring conditions. Take diabetes, for example. It is highly co-occurring with depression, which can be critical. Someone who is depressed just may not have the personal capacity required to address their diabetes. And, so we often encourage our customers to let us screen for depression as part of our condition-specific outreach – providing crucial resources to a population that could really use the support. And we’ve learned how to do this effectively – because talking about depression is not necessarily a natural turn in a conversation. If we’re speaking with you about your diabetes, you might be uncomfortably surprised if we suddenly start asking you if you’ve been feeling down, depressed or blue. So, we open the topic by sharing with people that even though the questions we are about to ask may not seem directly related to diabetes, they are incredibly important to our ability to support them. We’ve found that not only does engagement increase, but in some cases over 50% of our targeted population screens positive for depression. This lets us offer optimal support, which can include transferring that individual to a behavioral health specialist, right at that moment.

How do you quantify Return on Investment (ROI) for your customers?

When you engage with people in conversations about their health and inspire positive behavior, two things happen. First, it drives people to do the right thing – whether that’s getting a flu shot, keeping up to date with preventive screenings, filling their prescriptions, or even just visiting a social networking site where they can share experiences with people ‘like them’. Second, the things we are focused on driving people to do have proven to dramatically reduce long-term care costs – not to mention positively impact satisfaction, quality of life, and loyalty metrics like net-promoter scores.

ROI isn’t just a matter of reaching out to individuals to change aspects of their health. It’s about using Eliza’s insights to help our customers establish legitimate, valuable, loyal relationships with individuals who then stay with them. Member retention is a key indicator. Since our customers are responsible for the dollars associated with health-related behaviors, our ability to create positive behavior change, in a way that sustains or builds brand loyalty, is enormously powerful for them.

Much of that insight comes from our proprietary technology and data assets. When we deliver back to our customers granular data about how people are behaving, they can use our business intelligence to redesign their strategies in a way that reflects how real people want to interact with the resources that exist – for example, some people don’t want help from their doctor, while others do. Some go online; others do not. Some are traditional; others favor holistic approaches.

We interact with a lot of individuals over time using technology that analyzes how people are behaving in that moment. It gives us an extraordinary amount of perspective that can be translated into better overall management of those populations – management that reflects how real people live and make decisions.

On the subject of data, how has data transformed Eliza’s mission and responsibilities? How has the playing field changed?

Data forces us and our partners to get really creative, really fast. We are sitting on a best-practices database and we listen hard to what the data tells us about how people want to interact when it comes to their health – learning what works and what doesn’t.

One of the most important consequences of access to data is that health organizations can no longer ignore the reality of people’s lives. For a long time, the health care space felt that if it educated people, they would act accordingly. But when you begin to amass large amounts of data about real behavior by individuals in the ‘wild,’ you discover things that are not explained by traditional economic theories. Guess what? Humans aren’t rational. We are extraordinarily complicated. That makes it difficult for health organizations to influence behavior if all they’re using are traditional means built around old-fashioned assumptions.

One of the challenges our industry faces is we presume access to large amounts of data also comes with a button that says “Insights”. In some cases, the volume of data is almost creating learned helplessness – now I have all the data I’ve always wanted to have and the answers still aren’t jumping out at me so I Give Up! The beautiful thing about millions of data points on millions of people is it highlights the challenges (but therefore also the opportunities) of getting real people to change their behavior in a sustainable, cost-effective way. I think we’re in the middle of a very messy period when folks are getting that just having a lot of data won’t get people to do what they should be doing when it comes to their health on a day-to-day basis. Data doesn’t magically make it possible for us to influence behavior. That’s so hard. We all have the best intentions and so too do the people we’re reaching out to. It’s about using data to better understand who people are in a way that reflects how they see themselves, then developing data-driving messaging approaches that actually inspire behavior, and finally, focusing all this power and intellect on solving problems that real people want solved, which aren’t always the problems the healthcare space focuses on.

Please tell me about The Unmentionables. What are they and how will they push the envelope in the health care industry?

The Unmentionables are at the heart of Eliza’s mission. The health care space needs to expand the definition of health to consider the influence on our health of factors such as workplace stress, financial stress, relationship stress, or the fact that you’re caring for an aging parent.

We started a research and analysis initiative designed to investigate the extent to which these things are influencing people’s lives – and therefore their health. Forty percent (40%) of people said that they were experiencing between four and six Unmentionables at any point in time; 95% said that they had at least one.

Then we asked them to define on a scale of one to seven (1-7) how much these factors were defining their lives. Most responses were pegged up toward seven and then when we asked how well they felt supported by the health care space on these issues, the responses were in the 1-2 range. This led Eliza to create The Ostrich Index, which measures the delta between how much something negatively affects your life and the support you’re getting for it. The Unmentionables are off the charts on The Ostrich Index.

The reason Eliza has become so obsessed with this notion of The Unmentionables is that when we look at our data for the past 13 years, it supports with laser focus our theory that such factors create bad health outcomes, negatively impact productivity, and lead to the misuse of health care resources – a direct counter to our mission to help people be happier, healthier, and more productive. We’ve also found that when you give people an opportunity to speak about these issues in non-judgmental way, individuals are eager to share the extent to which these things are affecting their lives. And they want help.  Eighty percent (80%) of people want help from their doctor, health plan or employer on financial stress – 95% do for the stress of being a caregiver.

Quantitatively, the data shows that if you look at the combination of these unmentionable life challenges coupled with your positive and negative coping factors, you get what we call The Vulnerability Index. In fact, The Vulnerability Index is more than three times as predictive of a person’s self-reported health than traditional health factors, which is our industry’s default approach. We think this kind of impact can’t be ignored by those of us responsible for helping people live their happiest, healthiest, most productive lives. Imagine the ROI of an approach that not only allowed us to focus our limited resources on those who needed it most – but enabled us to capture attention and influence behavior because we were addressing the fundamental issues that don’t only drive actual health and productivity – they reflect the issues on which we are begging for help!

Is it fair to say that you need to make Health Care more seductive?

People aren’t sitting around waiting for the health care industry to lecture them into better behavior. We have to start learning from our brothers over in tobacco, food, and beverage industries who do a much better job seducing. We need to learn how to be worthy of the mind share that we are asking people to give us in our efforts to seduce them into making healthier choices on a daily basis. Our goal is to harness the incredible creativity and passion in all of us who are working so hard to effect change and to use that creativity and passion to bring joy and inspiration back into the conversation we are having with the public about their health and wellbeing. Couple that philosophy with a data-driven approach and you will be amazed what you can do.

How has Health Care Reform changed Health Engagement Management?

The Affordable Care Act has had a tremendous impact on our industry – we are being forced to respond to real problems with real solutions. No more fluff. If the industry can’t create measurable, demonstrable, sustained, better health outcomes, it won’t get paid. We had a meeting about seven months ago during which a health care executive stood up and told us, “Quite honestly, I don’t care what you do – you just need to get started doing it right now because if we don’t get these measurements up, we are not going to make a profit next year.”

Whether the Supreme Court upholds health reform or not, many of the changes being made in preparation have set off a sequence of events that will continue regardless – and that’s a good thing. Take the idea of health care exchanges. If reform is upheld, most people believe some version of the exchanges will come into play in 2014. Many would also say that even if reform is repealed, the idea of creating a transparent marketplace driven by consumer choice is an idea whose time has come.  The idea of an exchange changes everything because historically, the way that health plans have gotten volume—the way that they have amassed lives to care for—is by selling to benefits managers based on cost, etc. The decision of which health plan to choose for coverage has not been up to individuals. Health insurers are realizing that if they want to stay relevant in a world driven by consumer choice, they have to be relevant not just to the benefits manager who previously purchased plans on behalf of thousands of individuals, but rather to the thousands of individuals themselves.

Big Data and Politics: Obama's data advantage

Lois Romano, Politico, June 9, 2012

CHICAGO — On the sixth floor of a sleek office building here, more than 150 techies are quietly peeling back the layers of your life. They know what you read and where you shop, what kind of work you do and who you count as friends. They also know who your mother voted for in the last election.

The depth and breadth of the Obama campaign’s 2012 digital operation — from data mining to online organizing — reaches so far beyond anything politics has ever seen, experts maintain, that it could impact the outcome of a close presidential election. It makes the president’s much-heralded 2008 social media juggernaut — which raised half billion dollars and revolutionized politics — look like cavemen with stone tablets.

Mitt Romney indeed is ramping up his digital effort after a debilitating primary and, for sure, the notion that Democrats have a monopoly on cutting edge technology no longer holds water.

But it’s also not at all clear that Romney can come close to achieving the same level of technological sophistication and reach as his opponent. (The campaign was mercilessly ridiculed last month when it rolled out a new App misspelling America.)

“It’s all about the data this year and Obama has that. When a race is as close as this one promises to be, any small advantage could absolutely make the difference,” says Andrew Rasiej, a technology strategist and publisher of TechPresident. “More and more accurate data means more insight, more money, more message distribution, and more votes.”

Adds Nicco Mele, a Harvard professor and social media guru: “The fabric of our public and political space is shifting. If the Obama campaign can combine its data efforts with the way people now live their lives online, a new kind of political engagement — and political persuasion — is possible.”

Launched two weeks ago, Obama’s newest innovation is the much anticipated “Dashboard," a sophisticated and highly interactive platform that gives supporters a blueprint for organizing, and communicating with each other and the campaign.

In addition, by harnessing the growing power of Facebook and other online sources, the campaign is building what some see as an unprecedented data base to develop highly specific profiles of potential voters. This allows the campaign to tailor messages directly to them — depending on factors such as socio-economic level, age and interests.

The data also allows the campaign to micro-target a range of dollar solicitations online depending on the recipient. In 2008, the campaign was the first to maximize online giving — raising hundreds of millions of dollars from small donors. This time, they are constantly experimenting and testing to expand the donor base.

For example, they have found $3 to be a magic number: Asking supporters for that paltry donation to win a chance to attend a fundraiser with the president and George Clooney or Sarah Jessica Parker, has generated tens of thousands of responses — people from whom the campaign can collect highly valuable data and then go back to.

“They are way ahead of Romney micro-targeting and it’s a level of precision we haven’t seen before,” says Darrell M. West, a leading scholar on technology innovation at the Brookings Institution. “[The Obama campaign has] been able to work on it under the radar during the Republican primary season.”

Obama already this year has outspent Romney significantly (and outspent his own 2008 levels) for online advertising, according to several market analyses the numbers. The hope for the Obama team, says West, “is that his online reach will outweigh any Superpac funding advantage Romney might have for television advertising, by reaching deep into communities with targeted online advertising, grass-roots organizing and fundraising.”

The other hope, of course, is that Obama can replicate some of the online excitement that propelled him into office four years ago.

Romney campaign officials acknowledge that they have had neither the time, nor the resources to build a complex digital operation as they were fighting their way through the prolonged primary season. “It wasn’t something we were going to put resources into if he wasn’t the nominee,” said one adviser.

It is also apparent that the Romney campaign will stick closely to the traditional campaign model of heavy and expensive television spending — with the assist of wealthy conservative super PACs that have signaled a willingness to spend hundreds of millions of dollars to defeat Obama.

While the campaign is beginning to increase staff in preparation for the general election, Romney digital director Zac Moffatt says they will do things differently than the Obama operation. For one, Moffatt says the Romney campaign will outsource much of their data management — instead of handling it in house, creating “customized solutions” to fit their needs.

“As a campaign we would not presume to know more than the collective intelligence and resources of the marketplace,” said Moffatt, who adds that they will target audience through “our partnership with an industry-leading data management platform.”

“In the end, what is most important is not how many people on any list or how many followers we have — but their engagement level. And our followers are engaged,” Moffitt said.

Harvard’s Mele, who at age 25 helped pioneer the use of social media technology on Howard Dean’s 2004 presidential campaign, said that even with some obvious attrition to Obama’s 2008 database, Romney is trailing.

“I’m not going to say he can’t catch up because with enough money and intensity, it may be doable — but it seems very unlikely to me,” said Mele. “(Obama) used the powerful narrative of his 2008 campaign to build a digital infrastructure that remains formidable, both in terms of data and sheer know-how and expertise.”

The challenge facing both campaigns in 2012 is the changing consumer — and the endless ways they receive information. Four years ago, supporters might have been satisfied with just “friending” a candidate on Facebook, but today most users expect a more sophisticated way to actually engage –and on their terms.

While Obama campaign officials guard details about its digital operation as fiercely as Romney guards his tax returns, they were willing to share some advances since 2008:

• Created a holistic, totally in-house digital operation that is the largest department at campaign headquarters. In 2008, much of the social media and video was generated organically from supporters. As one campaign official put it, “digital is no longer a part of the campaign. It is the campaign.”

• Hired a number of nonpolitical tech innovators, software engineers and statisticians. “It has been incredibly freeing, because all election campaigns are a slave to history, and the history here is just nonexistent,” says Obama campaign manager Jim Messina. “So, we’ve been able to kind of reinvent it.”

• Invested mightily in cutting-edge technology that scales the website to fit the screen of any device. With nearly half of the U.S. population using smart phones, “responsive design” allows a user to give money and volunteer without bifocals. “More than 40 percent of all our donors are new, and a lot of them are coming in because of things like this,” says Messina. “Call up our website and try to donate on your phone and then do Romney’s. … Those things are important, because people are busy and people want to help us and they think about — ‘Oh, yeah, I saw the president on TV. I want to give them money. How hard is it?’ ”

• Developed a more complex symbiosis between the campaign and Facebook, which is 10 times bigger than it was four years go, and has far more personal information available to mine. “Facebook was just a site to see friends four year ago now it is part of people’s DNA,” notes a senior campaign adviser. Obama invites supporters to log on to the campaign through their Facebook accounts, which gives the campaign one more avenue for data.

• Opened the first all-volunteer. all-digital office in San Francisco where knowledgeable techies drop in for a few hours and strive to develop new software for the campaign under the supervision of paid staff.

• Staffed a full-time digital director in each of about a dozen battleground states to effectively run mini-general election campaigns in those states.

“Last time, we had two campaigns,” Messina said. “We had the on-the-ground, door-knocking, person-to-person campaign, and then we had the digital campaign. But most of the digital campaign was really organized by [supporters] — by themselves.”

This time, says Messina, it’s the campaign that’s driving and controlling most digital content. “The goal is to burst through the wall of those two things.”

Twitter was just gaining steam in 2008 when campaign used the platform largely to notify followers about events. The campaign had 118, 000 followers at election time and about 2.4 million Facebook followers.

Today, Obama has 16 million Twitter followers to Romney’s 500,000, and Michelle Obama has nearly 1 million to Ann Romney’s 45,000. On Facebook, Obama has nearly 27 million followers to Romney’s 1.8 million. (It’s hard to know how many of either man’s followers are non-American.)

“We are building content to a variety of different channels, because what has really changed is the channels are different, and so some people are going to get all of their stuff through Facebook, some people now do Twitter, some people are going to go directly to our website, some people like it on email, some people like it on text,” Messina said.

Less certain for Obama is whether his vast digital empire can recreate the movement and excitement of 2008. Messina makes it clear that some tried-and-true methods still apply. “It’s even more important this time than last time to run a real grass-roots campaign that’s built on turning out our voters, persuading undecided voters and making sure they’ll vote on Election Day,” he says, “and that is the kind of the organizing we’re doing.”

Why Google Isn't Making Us Stupid.or Smart

Chad Wellmon,  The Hedgehog Review: Vol. 14, No. 1 (Spring 2012)

Last year The Economist published a special report not on the global financial crisis or the polarization of the American electorate, but on the era of big data. Article after article cited one big number after another to bolster the claim that we live in an age of information superabundance. The data are impressive: 300 billion emails, 200 million tweets, and 2.5 billion text messages course through our digital networks every day, and, if these numbers were not staggering enough, scientists are reportedly awash in even more information. This past January astronomers surveying the sky with the Sloan telescope in New Mexico released over 49.5 terabytes of information—a mass of images and measurements—in one data drop. 

The Large Hadron Collider at CERN (the European Organization for Nuclear Research), however, produces almost that much information per second. Last year alone, the world’s information base is estimated to have doubled every eleven hours. Just a decade ago, computer professionals spoke of kilobytes and megabytes. Today they talk of the terabyte, the petabyte, the exabyte, the zettabyte, and now the yottabyte, each a thousand times bigger than the last.

Some see this as information abundance, others as information overload. The advent of digital information and with it the era of big data allows geneticists to decode the human genome, humanists to search entire bodies of literature, and businesses to spot economic trends. But it is also creating for many the sense that we are being overwhelmed by information. How are we to manage it all? What are we to make, as Ann Blair asks, of a zettabyte of information—a one with 21 zeros after it?1 From a more embodied, human perspective, these tremendous scales of information are rather meaningless. We do not experience information as pure data, be it a byte or a yottabyte, but as filtered and framed through the keyboards, screens, and touchpads of our digital technologies. However impressive these astronomical scales of information may be, our contemporary awe and increasing worry about all this data obscures the ways in which we actually engage it and the world of which it and we are a part. All of the chatter about information superabundance and overload tends not only to marginalize human persons, but also to render technology just as abstract as a yottabyte. An email is reduced to yet another data point, the Web to an infinite complex of protocols and machinery, Google to a neutral machine for producing information. Our compulsive talk about information overload can isolate and abstract digital technology from society, human persons, and our broader culture. We have become distracted by all the data and inarticulate about our digital technologies.

The more pressing, if more complex, task of our digital age, then, lies not in figuring out what comes after the yottabyte, but in cultivating contact with an increasingly technologically formed world.2 In order to understand how our lives are already deeply formed by technology, we need to consider information not only in the abstract terms of terrabytes and zettabytes, but also in more cultural terms. How do the technologies that humans form to engage the world come in turn to form us? What do these technologies that are of our own making and irreducible elements of our own being do to us? The analytical task lies in identifying and embracing forms of human agency particular to our digital age, without reducing technology to a mere mechanical extension of the human, to a mere tool. In short, asking whether Google makes us stupid, as some cultural critics recently have, is the wrong question. It assumes sharp distinctions between humans and technology that are no longer, if they ever were, tenable.

Two Narratives

The history of this mutual constitution of humans and technology has been obscured as of late by the crystallization of two competing narratives about how we experience all of this information. On the one hand, there are those who claim that the digitization efforts of Google, the social-networking power of Facebook, and the era of big data in general are finally realizing that ancient dream of unifying all knowledge. The digital world will become a “single liquid fabric of interconnected words and ideas,” a form of knowledge without distinctions or differences.3 Unlike other technological innovations, like print, which was limited to the educated elite, the internet is a network of “densely interlinked Web pages, blogs, news articles and Tweets [that] are all visible to anyone and everyone.”4 Our information age is unique not only in its scale, but in its inherently open and democratic arrangement of information. Information has finally been set free. Digital technologies, claim the most optimistic among us, will deliver a universal knowledge that will make us smarter and ultimately liberate us.5 These utopic claims are related to similar visions about a trans-humanist future in which technology will overcome what were once the historical limits of humanity: physical, intellectual, and psychological. The dream is of a post-human era.6

On the other hand, less sanguine observers interpret the advent of digitization and big data as portending an age of information overload. We are suffering under a deluge of data. Many worry that the Web’s hyperlinks that propel us from page to page, the blogs that reduce long articles to a more consumable line or two, and the tweets that condense thoughts to 140 characters have all created a culture of distraction. The very technologies that help us manage all of this information are undermining our ability to read with any depth or care. The Web, according to some, is a deeply flawed medium that facilitates a less intensive, more superficial form of reading. When we read online, we browse, we scan, we skim. The superabundance of information, such critics charge, however, is changing not only our reading habits, but also the way we think. As Nicholas Carr puts it, “what the Net seems to be doing is chipping away my capacity for concentration and contemplation. My mind now expects to take in information the way the Net distributes it: in a swiftly moving stream of particles.”7 The constant distractions of the internet—think of all those hyperlinks and new message warnings that flash up on the screen—are degrading our ability “to pay sustained attention,” to read in depth, to reflect, to remember. For Carr and many others like him, true knowledge is deep, and its depth is proportional to the intensity of our attentiveness. In our digital world that encourages quantity over quality, Google is making us stupid.

Each of these narratives points to real changes in how technology impacts humans. Both the scale and the acceleration of information production and dissemination in our digital age are unique. Google, like every technology before it, may well be part of broader changes in the ways we think and experience the world. Both narratives, however, make two basic mistakes.

First, they imagine our information age to be unprecedented, but information explosions and the utopian and apocalyptic pronouncements that accompany them are an old concern. The emergence of every new information technology brings with it new methods and modes for storing and transmitting ever more information, and these technologies deeply impact the ways in which humans interact with the world. Both the optimism of technophiles who predict the emergence of a digital “liquid” intelligence and the pessimism of those who fear that Google is “making us stupid” echo historical hopes and complaints about large amounts of information.

Second, both narratives make a key conceptual error by isolating the causal effects of technology. Technologies, be it the printed book or Google, do not make us unboundedly free or unflaggingly stupid. Such a sharp dichotomy between humans and technology simplifies the complex, unpredictable, and thoroughly historical ways in which humans and technologies interact and form each other. Simple claims about the effects of technology obscure basic assumptions, for good or bad, about technology as an independent cause that eclipses causes of other kinds. They assume the effects of technology can be easily isolated and abstracted from their social and historical contexts.

Instead of thinking in such dichotomies or worrying about all of those impending yottabytes, we might consider a perhaps simple but oftentimes overlooked fact: we access, use, and engage information through technologies that help us select, filter, and delimit. Web browsers, hyperlinks, blogs, online newspapers, computational algorithms, rss feeds, Facebook, and Google help us turn all of those terrabytes of data into something more useful and particular, that is, something that can be remade and repurposed by an embodied human person. These now ubiquitous technologies help us filter the essential from the excess and search for the needle in the haystack, and in so doing they have become central mediums for our experience of the world.

In this sense, technology is neither an abstract flood of data nor a simple machine-like appendage subordinate to human intentions, but instead the very manner in which humans engage the world. To celebrate the Web, or any other technology, as inherently edifying or stultifying is to ignore its more human scale: our individual access to this imagined expanse of pure information is made possible by technologies that are constructed, designed, and constantly tweaked by human decisions and experiences. These technologies do not exist independently of the human persons who design and use them. Likewise, to suggest that Google is making us stupid is to ignore the historical fact that over time technologies have had an effect on how we think, but in ways that are much more complex and not at all reducible to simple statements like “Google is making us stupid.”

Think of it this way: the Web in its entirety—just like those terrabytes of information that we imagine weighing down upon us—is inaccessible to the ill-equipped person. Digital technologies make the Web accessible by making it seem much smaller and more manageable than we imagine it to be. The Web does not exist. In this sense, the history of information overload is instructive less for what it teaches us about the quantity of information than what it teaches us about how the technologies that we design to engage the world come in turn to shape us. The specific technologies developed to manage information can give us insight into how we organize, produce, and distribute knowledge—that is, the history of information overload is a history of how we know what we know. It is not only the history of data, books, and the tools used to cope with them. It is also a history of ourselves and of the environment within which we make and in turn are made by technologies.

In the following sections, I put our information age in historical context in an effort to demonstrate that technology’s impact on the human is both precedented and constitutive of new forms of life, new norms, and new cultures. The concluding sections focus on Google in particular and consider how it is impacting our very notion of what it is to be human in the digital age. Carr and other critics of the ways we have come to interact with our digital technologies have good reason to be concerned, but, as I hope to show, for rather different reasons than they might think. The core issue concerns not particular modes of accommodating new technologies—nifty advice on dealing with email or limiting screen time—but our very conception of the relationship between the human and technology.

FULL ARTICLE AT 

Social media and big data

Twitter Knows When You Sleep, and More

Somini Sengupta, The New York Times, June 7, 2012

Twitter knows when you nap. At least its engineers think they have a pretty good idea when you nap, because those are the few times when you’re not pouring your heart out in 140-character bursts.

Peering into its fire hose of tweets from four big cities around the world, a pair of Twitter engineers, Manuel Rios and Jimmy Lin, have discovered that the Twitter users in Istanbul don’t sleep much in August — and why would they, considering how many people are strolling along Istiklal Avenue on summer evenings. The denizens of Sao Paulo, Brazil, seem to take an afternoon siesta – tweets sharply drop off in the period after lunch.

New Yorkers tend to be as voluble on Twitter during the workday as they are during evenings. But that is not the case for the other three cities. The Japanese in Tokyo appear to be most disciplined about actually working during work hours – they tweet very little, relatively speaking, between 10 a.m. and 6 p.m. Their sleeping hours also appear relatively unchanged over winter and summer.

All this intelligence is potentially lucrative for a global communications business like Twitter. It can inform when to serve up advertisements on the site and potentially for what kinds of products. It also produces very cool graphs and charts.

The data at the disposal of social media companies are, in principle, valuable to other organizations, from soda pop companies to politicians to police agencies that seek to understand human behavior. Location is one of the most useful pieces of information, especially for marketers that want to target their advertising dollars most efficiently.

Facebook this week released a map that political advertisers would find useful in an election year. It matched its users political affiliations with where their physical locations. The map went far beyond the red state versus blue state divide, with far more fine grained information about where Facebook users checked in in various parts of the country.

California, predictably, had large swathes of Democratic blue, but with plenty of Republican red polka dots in in the Orange County and San Diego suburbs.

New statistical model lets patient's past forecast future ailments

Kurzweil Accelerating Intelligence Blog, June 6, 2012

Analyzing medical records from thousands of patients, statisticians have devised a statistical model for predicting what other medical problems a patient might encounter called a “hierarchical association rule model.”

Like how Netflix recommends movies and TV shows or how Amazon.com suggests products to buy, the algorithm makes predictions based on what a patient has already experienced as well as the experiences of other patients showing a similar medical history.

“This provides physicians with insights on what might be coming next for a patient, based on experiences of other patients. It also gives a predication that is interpretable by patients,” said Tyler McCormick, an assistant professor of statistics and sociology at the University of Washington.

McCormick’s co-authors are Cynthia Rudin, Massachusetts Institute of Technology, and David Madigan, Columbia University.

McCormick said that this is one of the first times that this type of predictive algorithm has been used in a medical setting. What differentiates his model from others, he said, is that it shares information across patients who have similar health problems. This allows for better predictions when details of a patient’s medical history are sparse.

For example, new patients might lack a lengthy file listing ailments and drug prescriptions compiled from previous doctor visits. The algorithm can compare the patient’s current health complaints with other patients who have a more extensive medical record that includes similar symptoms and the timing of when they arise. Then the algorithm can point to what medical conditions might come next for the new patient.

“We’re looking at each sequence of symptoms to try to predict the rest of the sequence for a different patient,” McCormick said.      If a patient has already had dyspepsia and epigastric pain, for instance, heartburn might be next.

The algorithm can also accommodate situations where it’s statistically difficult to predict a less common condition. For instance, most patients do not experience strokes, and accordingly most models could not predict one because they only factor in an individual patient’s medical history with a stroke. But McCormick’s model mines medical histories of patients who went on to have a stroke and uses that analysis to make a stroke prediction.

The statisticians used medical records obtained from a multiyear clinical drug trial involving tens of thousands of patients aged 40 and older. The records included other demographic details, such as gender and ethnicity, as well as patients’ histories of medical complaints and prescription medications.

They found that of the 1,800 medical conditions in the dataset, most of them occurred fewer than 10 times. McCormick and his co-authors had to come up with a statistical way to not overlook those 1,400 conditions, while alerting patients who might actually experience those rarer conditions.

They came up with a statistical modeling technique that is grounded in Bayesian methods, the backbone of many predictive algorithms. McCormick and his co-authors call their approach the Hierarchical Association Rule Model and are working toward making it available to patients and doctors.

“We hope that this model will provide a more patient-centered approach to medical care and to improve patient experiences,” McCormick said.

The work was funded by a Google Ph.D. fellowship awarded to McCormick and by the National Science Foundation.

Ref.: Tyler H. McCormick, Cynthia Rudin and David Madigan, Bayesian Hierarchical Rule Modeling for Predicting Medical Conditions, to be published in Annals of Applied Statistics, 2012

Ref.: Tyler H. McCormick, Cynthia Rudin and David Madigan, Bayesian Hierarchical Rule Modeling for Predicting Medical Conditions, 2012, open access version [PDF]

Panel: The promises and perils of big data

Rachel King, ZDNET, June 5, 2012

Summary: Big data is already enabling many businesses to make better decisions, but it has only made incremental changes in other verticals.

SAN FRANCISCO – Big data offers a lot of promise and opportunities for improving our society — especially for business and education purposes — but there are a lot of hurdles to get around first.

A few of those pitfalls and perils were discussed in detail during a panel discussion at The Economist’s two-day summit about information, which kicked off on Tuesday afternoon.

Economist data editor Kenneth Cukier got the ball rolling during a series of rolling interviews exploring the promise and perils of the data deluge.

Jeff Hammerbacher, chief scientist of Cloudera, posited that the most interesting thing going on in big data today is making data preparation more granular.

“The more you zoom in, the more pathologies you find,” Hammerbacher said.

When Cukier pressed if this meant that big data can’t always be trusted for one reason or another, Hammerbacher countered by asking, “If you can’t trust the data, what can you trust?”

“It’s hard for me to think of a case where measuring more makes you less certain,” Hammerbacher asserted.

One of the barriers towards big data at the moment, Hammerbacher hypothesized, is that we’re in this “brief pocket of time” where people involved in statistics and computer science have an advantage over most of the population in this field, allowing them to draw information and conclusions that most people can’t.

“It’s fun to exploit this inefficiency,” Hammerbacher joked, but he predicting that our tools will automate big data analytics for everyone within 10 to 15 years.

Hammerbacher continued the rolling interview style by following up with Geoffrey Nunberg, an adjunct professor at the School of Information at the University of California, Berkeley.

Nunberg commented that there has been progress in using big data sources in linguistics and how language is used as a tool in different communities, but that change has been more incremental rather than a “quantum leap.”

Looking at Google Translate, Nunberg admitted that the online translation tool is “so much better than systems we had a long time ago.” But the problem is that it is still hard to tell the difference because Google Translate still doesn’t usually produce results that even a “first year” language student would offer.

As an observer of language in the cultural scene, Nunberg said that he uses data analytics this all the time to see the way words are changing in meaning. For example, Nunberg cited the word “elite” used to be usually modified by financial terms, but now it is more associated with the media.

Turning towards social networks, Juliette Powell, author of the book, 33 Million People in the Room, cited social media as a medium “to co-create the future together” using connected technology.

When it comes to understanding data sets, Powell outlined four core elements that any Internet user would recognize that are vital for understanding big data analytics going forward: digital literacy, digital trust, platform transparency, and access.

But what is most interesting to Powell are the people themselves, adding that we shouldn’t get caught up in data sets and forget the relationships.

“It’s all about trust,” Powell asserted, explaining that we choose the information we share and that fact needs to be kept in mind when analyzing big data.

One topic that always seems to come up when discussing either social media or big data is privacy. Circling back, Cukier argued that this is not the central problem to big data, but rather “propensity” and the possibility that we’re going to have algorithims predicting our behaviors.

See also: http://www.economist.com/events-conferences/americas/information-2012

How Big Data Gets Real

Quentin Hardy, The New York Times, June 4, 2012

The business of Big Data, which involves collecting large amounts of data and then searching it for patterns and new revelations, is the result of cheap storage, abundant sensors and new software. It has become a multibillion-dollar industry in less than a decade. Growing at speed like that, it is easy to miss how much remains to do before the industry has proven standards. Until then, lots of customers are probably wasting much of their money.

There is essential work to be done training a core of people in very hard problems, like advanced statistics and software that ensures data quality and operational efficiency. Broad-based literacy in the uses of data should probably happen too, along with new kinds of management, better tools for reading the information, and privacy safeguards for corporate and personal information.

That such a huge number of tasks are taking place is a good indicator that, even with the hype, Big Data is a big deal. Last Friday, a number of technologists gathered at a forum hosted by the University of California, Berkeley, iSchool and talked about ways many of these jobs are being done. (Disclosure: I lecture at the iSchool, which is the school of information science, and moderated several panels there.) They talked about the progress so far, and identified a number of good ideas and businesses left to pursue.

In some ways, Big Data is about managing all kinds of weird new data, like social media updates from a mobile phone. It is hard to categorize in the first place, and may be used in lots of different ways, from advertising to traffic management. The so-called unstructured database of choice is by now pretty clearly Hadoop.

Cloudera, a leading software producer, is now training 1,500 people a month, mostly online, in how to use both the database and associated applications. According to Amr A. Awadallah, Cloudera’s chief technical officer, over 10,000 people have been trained on its system.

Data quality from new diverse sources is still a big problem, as is persuading companies and organizations to let others see data that might be more valuable in a commonly shared algorithm. “I’ve tried paying money for it, but it’s easier for companies to decide not to share,” said Gil Elbaz, the founder of Factual, a company that seeks to hold lots of online data. “The only way that works is to get them to take risks in exchange for data that is valuable to them.”

Much of the fear about exposing data, he said, has to do with competitors learning secrets. Mr. Elbaz thinks there is a good business in developing “de-identifiers” that can make data anonymous, and privacy insurers specializing in covering the costs of exposure.

On a personal level, others think the government or a trusted private institution will hold the personal identifiers of things like medical data, releasing it to trusted parties. “It’s a little scary that right now a cab driver using Uber knows more about you than a doctor, who has to take all of your information for the first time,” said Peter Skomoroch, principal data scientist at LinkedIn.

Another data-improving business consists of moving the world’s older data online. A company called Captricity aims to couple image-capture from things like cellphone cameras with cheap workers in Amazon.com’s Mechanical Turk service, in order to put older handwritten documents into digital databases. The company’s early business is from government and charity sites in Africa and India, but there is no reason why it shouldn’t be valuable for most medical records. If someone took the trouble to write it down, the company figures, that is a good way to assume it is valuable data.

There are other businesses trying to take the arcane side of Big Data into the mainstream, with easy-to-use statistical tools and new ways of visualizing data that make it easier to understand. Companies like ClearStory and Platfora “want to make it possible for businesses, for history majors, to use,” said Ben Werther, chief executive of Platfora. “We’re in the pre-industrial age of Big Data.” Martin Wattenberg, creator of a well-known wind map and who is now at Google, talked about a necessary revolution in design of data outcomes that have yet to become widespread.

Away from big online companies like Google, some of the earliest data-driven businesses include hedge funds and insurance companies, which have the money and tradition of using lots of math. Another, Mr. Skomoroch said, is the Mormon Church. “It is the focus on genealogy data,” he says. “If you move, they’ll always find you.”

What Big Data is seeing now looks like the classic industrial curve. There is the first discovery of something big, leading to establishing principles like scientific rules. Science moves toward engineering as a means to manufacturing, resulting in mass deployment. Then things really change.

Rethinking Privacy in an Era of Big Data (Danah Boyd)

Quentin Hardy, The New York Times, June 4, 2012

Some years ago an engineer at Google told me why Google wasn’t collecting information linked to people’s names. “We don’t want the name. The name is noise.” There was enough information in Google’s large database of search queries, location, and online behavior, he said, that you could tell a lot about somebody through indirect means.

The point was that actually finding out people’s names isn’t necessary for sending them targeted ads. It can probably lead to trouble, as Google’s own adventures in Wi-Fi snooping show. Even without knowing your name, increasingly, everything about you is out there. 

Whether and how you guard your privacy in an online world we are building up every day has become increasingly urgent.

“Privacy is a source of tremendous tension and anxiety in Big Data,” says Danah Boyd, a senior researcher at Microsoft Research. Speaking last week at a conference on Big Data at the University of California, Berkeley, she said, “It’s a general anxiety that you can’t pinpoint, this odd moment of creepiness.” She asked, Iis this moving towards a society that we want to build?”

If conventional understanding chafes at the idea that our names are mere noise, consider the challenge in Ms. Boyd’s point about the self in a highly networked society. Take personal genetic data. “If I give away data to 23andMe, I’m giving away some of my brother’s data, my mother’s data, my future kid’s data.” For that matter, “Who owns the e-mail chain between you and me?”

Privacy is not a universal or timeless quality. It is redefined by who one is talking to, or by the expectations of the larger society. In some countries, a woman’s ankle is a private matter; in some times and places, sexual orientations away from the norm are deeply private, or publicly celebrated. Privacy, Ms. Boyd notes, is not the same as security or anonymity. It is an ability to have control over one’s definition within an environment that is fully understood. Something, arguably, no one has anymore.

“Defaults around how we interact have changed,” she said. “A conversation in the hallway is private by default, public by effort. Online, our interactions become public by default, private by effort.”

There other ways in which we can lose control of our privacy now. By triangulating different sets of data (you are suddenly asking lots of people on LinkedIn for endorsements on you as a worker, and on Foursquare you seem to be checking in at midday near a competitor’s location), people can now conclude things about you (you’re probably interviewing for a job there) that are radically different from either set of public information.

What is to be done? Ms. Boyd has made a specialty of studying young people’s behavior on the Internet. She says they are now often seeking power over their environment through misdirection, such as continually making and destroying Facebook accounts, or steganography, a cryptographic term for hiding things in plain sight by obscuring their true meaning. “Someone writes, ‘I’m sick and tired of all this,’ and it gets ‘liked’ by 32 people,” she said. “When I started doing my fieldwork I could tell you what people were talking about. Now I can’t.”

That is a placeholder solution, and Ms. Boyd sees only one certainty for which we should prepare. “Regulation is coming,” she says. “You may not like it, you may close your eyes and hold your nose, but it is coming.”

The issue is what the regulation looks like, and how well it is considered. “Technologists need to re-engage with regulators,” she says. “We need to get to a model where we really understand usage.” Right now, even among the highest geek circles, “we have very low levels of computational literacy, data literacy, media literacy, and all of these are contributing to the fears.”

Knight News Challenge data is now open

The Knight News Challenge is being offered three times this year in short, focused rounds to better mirror the pace of innovation. Winners of Round 1, which focused on networks, will be announced June 18. Here, Journalism and Media Innovation Program Director John Bracken writes about the Knight News Challenge: Data.

Today, we are opening the Knight News Challenge on data. We are asking just eight questions - and 500 words - for your share of $5m. And you have three weeks to enter before the challenge closes at noon EST June 21.

We’re looking for ideas that help make data more useful, by collecting, processing, visualizing or otherwise making it available, understandable and actionable. 

We don’t have a vision for what we hope to fund through the contest— if we did, we wouldn’t need a contest.

Here are some examples of data being useful: 

·       The quantified self and “datasexuals.” Dominic Basulto (via Chris Messina) notes that the “the same cultural zeitgeist that gave us the metrosexual…is also creating its digital equivalent: the datasexual… They are relentlessly digital, they obsessively record everything about their personal lives, and they think that data is sexy…Their lives - from a data perspective, at least - are perfectly groomed.” Devices like  Fitbit and Nike FuelBand and apps like MapMyRun , RunKeeper and The Eatery feed and accelerate the ability to collect, understand and share data about our bodies’ input and output.

·       “Statistical noisemaking as protest.”Alexis Madrigal recently explored how anti-corporate activists, by coordinating their clicks on online ads, could  “muddy data” and thereby confuse advertisers. “How long before [Occupy activists] realize that many businesses are valuable more for their data than their storefronts? Banks are troves of data being mined for profitable strategies. How long before activists see that making their data harder to analyze could be a political tool?”

·       Seatguru collects data from different sources to help travelers understand what we’re getting ourselves into and to make decisions that can have real impact on the quality of our lives: Will I be able to plug in my laptop? How late does my flight usually arrive? Which seats have the most legroom?

·       Forbes’ Jon Bruner built a story about the background of journalists out of data he teased out of LinkedIn, concluding that “Financially, the media industry looks much like the rest of America. In terms of educational attainment and cultural preferences, it’s very different.” (Jon did his Ignite talk at NewsFoo on the same topic.)

“Before going to News Foo, I built a Python script to scrape the LinkedIn profiles of everyone at the conference. Ten percent of the conference’s attendees went to Harvard….Journalists are also much more likely than Americans in general to live in New York or Washington…[A] reporter is nearly twice as likely as the average American to live in New York and three times as likely to live in Washington.”

·       Governments are moving to make their our data more available and useful to citizens. One example is the White House’s digital government strategy. Announced last week, it calls for “agencies to make as much data as possible public and machine-readable through application programming interfaces — APIs — and other tools.”

So send us your ideas on how to make data useful. We’ll be holding a Google Hangout to answer questions at 11 a.m. EST Tuesday June 5, on the Knight Foundation page. You can also reach us @knightfdn or newschallenge@knightfoundation.org - and check out our FAQ.