Cairns Blog

From Russia with Love

A few days ago I gave a presentation on “Solving Public Problems with Crowdsourcing” for 80 assembled governors of the Russian Federation outside of Moscow at an event on Contemporary Management Technologies.

The presentation was a sweep of the global landscape covering ways institutions are engaging with the public on (and off) line to brainstorm, collect data, undertake tasks, generate economic wealth, raise money and elicit opinions.  In other words, I was focusing on how government institutions can collaborate with citizens to solve problems through crowdsourcing.

It was a diverse group to address. It included the governor of Arkhangelsk, an Arctic Circle state, with no roads that can only be accessed by helicopter as well as the governor of Krasnoyarsk, a state so big that it takes 4 hours to fly across it. The federal president appoints these governors. But a new law (recently reversed in part by Putin) is slated to establish direct elections this autumn. Hence many of these leaders are thinking about their electoral chances in a time of increasing dissatisfaction with the Kremlin and, above all of the need to deliver better services to improve Russian quality of life with dwindling resources.

My goal was to be concrete and specific (thanks GovLab Research) about how to use technology to solve real problems in practice not in theory. My perspective was informed by my experiences in public service where I learned how hard it is to get stuff done on a hundred topics a day without access to good information or the time to get it. I wanted to convey at this propitious moment with the promise (threat?) of elections around the corner that engagement is a way, not to get beaten up by the press and public, but to collaborate to the end of solving real problems.

Crowdsourcing itself is a slippery term. A recent article counted 40 different definitions of “crowdsourcing” and there are lots of terms from open innovation to collective intelligence to crowd crafting to human computation, which get substituted interchangeably. My approach to defining crowdsourcing centered around the following variables:

Institution + Platform + Incentive + Citizens/Communities = Crowdsourcing

Using this model, I wanted to get the audience away from thinking that citizens can only complain and, instead, to remind them that people are smart and will give if asked. Crowdsourcing, is in essence, about strategies for working with citizens, companies and organizations — whether those with specific expertise, passion and know how or the general public (didn’t know how to say Joe Q Six Pack in Russian) — to collaborate with a government institution to solve a problem.

Important to point out is what crowdsourcing projects have in common (and this I learned clearly fromDaren Brabham‘s excellent short book on Crowdsourcing) is that crowdsourcing projects must be directed by an institution and depend upon the institution to organize the practice and set the rules. They aren’t bottom up exercises like Wikipedia nor are they top down activities like a corporate contest to pick a logo. To quote from one of my favorite slides that I often use: “it ain’t a thousand flowers blooming or command and control.”

Rather, Crowdsourcing exists in the middle between institution and citizens. This liminal space between government and civil society is such a fertile and little understood idea, especially in the public sector, where there can be resistance to saying “I don’t know” and “I need help” from anyone outside my professional cadre. To do crowdsourcing well, the institution has to play an active role in planning. This demands work but it also creates an incentive to public leaders who fear that engagement might lead to plebiscite or free for all where either they get beaten up or are drowned in information they can’t use.

For the future, GovLab and others need to create materials that outline the anatomy of specific crowdsourcing projects.  For institutional leaders flirting with the idea of greater openness and engagement, it will help to understand concretely both the steps to take and the steps taken by others before. For example, it may be heartening for the wary and would-be innovator to know that Veterans Affairs ran several important challenges to decrease the backlog in Veterans benefits claims with a 3K piece of software and part-time assistance of one employee. It’s crucial to understand that the open health government data movement got started with one tireless individual working to convene a few dozen people to use the data to demonstrate what’s possible. Today that Healthdatapalooza movement has growth exponentially and others outside of government have stepped up to help. But it doesn’t have to be hard to get started.

To inspire real change, vision must be married to tactics.. In short, the more concrete, the more granular and the more specific we can get by telling stories of how innovations got done and why they worked, the more likely we are to encourage leaders to understand that engagement is a must have tool of successful governance.

So please share examples of the “how to” of crowdsourcing – who did what, how much did it cost, what were the tools and, above all, what were the outcomes — that can be used to inspire any leader to become more collaborative and, as result, more effective. Comments here or links cc @thegovlab. Thanks!

Wiki Style Government

The Public Administration Select Committee of the UK Parliament issued a Report on Public Engagement in Policymaking calling for “wiki style” approaches to policymaking and creating greater engagement in the workings of government.

The Report, which is the result of a extensive consultation, including three live testimonies before the Committee (Full disclosure: I had the pleasure of being a witness) and extensive written submissions advocates a smarter government strategy of using online tools to elicit more information without ceding responsibility for decisionmaking. Rather, civil servants should steward the process of engagement in the public interest.

To govern is to choose. Open policy-making should take debate outside Whitehall and into the community as a whole, but ultimate responsibility and accountability for leadership must remain with Ministers and senior civil servants. Once again, we emphasise the importance of leadership in Government; of effective strategic thinking, which involves choosing between different arguments, reconciling conflicting opinions and arbitrating between different groups and interests; and of effective governance of departments and their agencies. A process of engagement, which can reach beyond the “Westminster village” and the “usual suspects”, will itself be an act of leadership, but there can be no abdication of that leadership.

 

“Groupy Once More”: J. Richard Hackman on Group Behavior and Performance

This week family and friends of the late Richard Hackman, who passed away in January, gather in Cambridge, Massachusetts to celebrate his life and work. He devoted his career as a social and organizational psychologist to studying what makes groups from spies to orchestras to sports teams (he was a huge fan of Harvard Women’s Basketball) work well both in terms of their ability to accomplish a goal and to enhance individual satisfaction.  For a bibliography, see his website and buy his books here. Thankfully, he still had opportunity to complete the definitive catalogue raisonee of group behavior research, a chapter on Group Behavior and Performance in the Handbook on Social Psychology (2010).

In his honor, I went back to re-read this masterful survey (co-authored with Nancy Katz) on the study of groups. It covers the major theories, research questions, thinkers and their writings. For anyone working on democratic engagement and participatory democracy and thinking about crafting mental models for how to engage groups in governance, the chapter advances our evidence-based understanding of:

1.     When should groups be used and when not? What kinds of tasks do they perform well?

2.     How does group performance compare to that of individual engagement?

3.     What differentiates groups that realize their full potential and the potential of their members from those that do not?

In the chapter, Hackman outlines the history of organizational psychology.  The exigency to leverage scientific expertise to combat Nazism in World War II gave rise to the formal study of what makes teams – in that case infantry squads – work. In 1945, Kurt Lewin founded the Research Center for Group Dynamics at MIT. This “action research” center was devoted to conducting controlled experiments in the laboratory and the field on group communication. Challenged by the experience of the Holocaust, Lewin was explicitly normative: “research that produces nothing but books will not suffice.”

What started as a project in psychology, however, has pervaded other disciplines, including political and policy sciences.  Social psychology’s interest in groups waned by the 1970s in response to the pervasive individualism of the 1960s. Interest in groups did not disappear, however, and today there are scholars in business schools and computer science departments who look at permanent and temporary (what Hackman called sand dune teams) groups online and off in a variety of organizations. (I studied in a law school where Richard’s work on purposive groups had a huge impact on my own interest in groups as the fundamental unit of democratic culture.)  It is not surprising that those concerned with re-imagining governance and who want to invigorate democratic engagement are looking to this literature on what makes groups work.

Studying dynamic, living, changing organisms like groups is, Hackman explains, a hard and imperfect science and fraught with ideological conflict. Hackman contends that there will never be a “definitive” study of groups given the different metrics for success, the different tasks groups perform and the diverse ways, including synchronous and asynchronous, that they work. Some researchers focus on whether the group accomplished its productive purpose while others measure to what extent the experience of working in a group contributes to individual well being.  Those who focus on process also study whether the way the group was organized made it function more or less well. (Citizen engagement researchers could do well to learn from this typology. We tend to focus either on the outcome, such as what app we built, without thinking about the benefits of getting engaged. Alternatively, others focus on the salutary benefits to individuals of participating in a deliberative process without inquiring as to the outcomes).

Hackman created another very useful typology for understanding the multiplicity of tasks that people acting in groups undertake. Groups do different things. Track teams perform “disjunctive tasks” where performance is measured by the best member. There are “conjunctive tasks” like a roped together group of climbers, who perform at the level of the least competent member. “Additive” tasks like a tug of war or many crowdsourcing projects are the sum of the parts. “Compensatory” tasks are those specific kinds of crowdsourcing projects that average the results of member contributions like a prediction market. “Complementary” tasks involve the assembly of smaller parts into a whole such as we find on Wikipedia. When we begin to take apart the different ways that groups function, (Hackman studied all of these), we begin to gain insight into how to design ways for groups to participate in different tasks of governance.

Over the years, Hackman and Katz discuss in the chapter, how we’ve evolved different approaches to studying the different aspects of group behavior. While action research emphasizes the normative goal of improving the work of the group, the psychodynamic approach studies the emotional state of groups as epitomized by Le Bon’s study of The Crowd: A Study of the Popular Mind. The network approach analyzes the relationship between group members and tracks the changing patterns of network ties. The process approach is to look at how the group interacts, often using computational tools. Its major advance is to “capture nuance and meaning and to produce a rich and data-based understanding of how group members interact.  The decision-analytic approach focuses either on “qualitative, case-based methods” or “highly quantitative and largely laboratory-based” methods. It is this model that gave rise to Irving Janis’ work on groupthink that has had such a negative influence on policymakers willingness to embrace more citizen consultation.

Finally, the complex systems approach is a more recent trend emerging from the natural sciences to model mathematically the evolving patterns of group behavior over time. The chapter contains even more comprehensive yet concisely articulated analysis of the approaches to group analysis and recent studies of compositional diversity, the role of moderation, the effect of training, and the impact of the broader values of the system in which the group operates. As groups move more online, these different approaches are being revisited with an eye towards understanding how working at a distance changes (or doesn’t) these earlier insights.

One of the important insights in this chapter to thinking about the role of groups in democratic life is the way groups work together to create a shared mental model.  “When group members share the same mental model, coordination is enhanced and the likelihood of miscues is lessened.” Today we lack these shared mental models for how networked publics outside of government can work together with professionals inside government to solve problems.

We understand what it means to vote in an election or sign a petition. But we are only just beginning to comprehend how the kinds of tasks that groups can perform together (see above) can be applied to addressing public problems. Instead of guessing jellybeans in a jar, we are learning that groups that aggregate predictions can help us understand complex policy issues. Instead of additive social tasks, we are now beginning to realize that people can also crowdsource the collection of data to fight disease and improve people’s quality of life. Thanks to Hackman’s research and the rise of technologies for distributed communication, we now have an opportunity for more people to engage in the life of our democracy and make our democratic culture stronger and more robust.

Creating an organizational culture that emphasizes shared, public interest over individual self-interest is the essence of democracy. But designing the processes for more and more distributed purposive groups to do public work and take on the tasks of governance requires drawing on Hackman’s research.

Without insights from the social sciences we cannot know how to compose teams, how to organize their work together, how to define their tasks and how to create incentives for group based collaboration that achieves a goal and enhances satisfaction. We surely don’t know how to do this online without drawing on what he’s taught us about team effectiveness offline. “It is reasonably well established, for example, that performance is facilitated when teams are small and compositionally stable with clear, but permeable, boundaries and interdependence for some consequential, shared purpose.”

While in the chapter Hackman and Katz describe the field as a whole, it is stunning to note how much of the literature and how many important insights he contributed to the field. As we stand on the threshold of the opportunity to redesign our democratic institutions to incorporate the work of outside groups in governance, his analysis, passion and humor will be profoundly missed. While technically a psychologist, Hackman should go down in history as one of the great and important political thinkers of our time. 

IRS: Turn Over A New Leaf, Open Up Data

From the Forbes blog

Co-authored by Stefaan Verhulst

The core task for Danny Werfel, the new acting commissioner of the IRS, is to repair the agency’s tarnished reputation and achieve greater efficacy and fairness in IRS investigations. Mr. Werfel can show true leadership by restructuring how the IRS handles its tax-exempt enforcement processes.

One of Mr. Werfel’s first actions on the job should be the immediate implementation of the groundbreaking Presidential Executive Order and Open Data policy, released last week, that requires data captured and generated by the government be made available in open, machine-readable formats. Doing so will make the IRS a beacon to other agencies in how to use open data to screen any wrongdoing and strengthen law enforcement.

By sharing readily available IRS data on tax-exempt organizations, encouraging Congress to pass a budget proposal that mandates release of all tax-exempt returns in a machine-readable format, and increasing the transparency of its own processes, the agency can begin to turn the page on this scandal and help rebuild trust and partnership between government and its citizens.

Every year in the United States approximately 1.5 million registered tax-exempt organizations file a version of the “Form 990” with the IRS and state tax authorities. The 990 collects details on the financial, governance and organizational structure of America’s universities, hospitals, foundations, and charities to the end of ensuring that they are deserving of tax exempt status. We are missing an opportunity to analyze this data so that decisions about whom to investigate can be based on evidence rather than conjecture, on patterns rather than prejudice.

Currently, hundreds of thousands of the largest tax-exempt organizations are required to file their returns electronically. The IRS should release this data in bulk as a free database immediately.  If the IRS were to make these 990 data available in a form that could be easily downloadable and processed by computer programs for visualization and statistical analysis, researchers could quickly do more extensive, in-depth empirical research to better understand the sector and spot fraud, waste and abuse more systematically. Knowing who runs a nonprofit can help detect fraud. Attorneys General have occasionally found the same person collecting full time salaries from several different nonprofits.

While the IRS is using robo-audits, catching large evasions still happens mainly by happenstance. With open data, they could be detected, first, through computer analysis. By using technology to expand the regulator’s toolkit, it becomes possible to target limited enforcement resources to where problems really are.  The Securities and Exchange Commission has, for instance, developed an improved capacity to detect and prevent insider trading more effectively by making public information computable and easier to mine. In addition, open data creates the means for government and citizens to collaborate on spotting problems. As the adage goes, with many eyes, all bugs are shallow.

Similarly, Form 990 requires charities to disclose loans to or from current and former officers. Making these and other transactions that correlate with instances of fraud like these would save government resources at the state and federal levels.

With a 990 database, it would also be easier to run queries to understand which executives receive the highest compensation. By combining 990 and other data,  such as lobbying data, it might become possible to spot impermissible political activities.

President Obama’s 2014 budget calls for requiring all tax exempts to file electronically, but also requires that the IRS makes these already public returns available in a timely, machine-readable format. These data would create a corpus of open, computable information that could be used to understand where nonprofits are providing services and where there are gaps. Enabling more people and organizations to analyze, visualize, and mash up the data, creating a large public community that is interested in the nonprofit sector and can collaborate to find ways to improve it.

In sum, the data that the IRS collect about nonprofit organizations present a great opportunity to learn about the sector and make it more effective.

Making IRS data open won’t solve every problem; the recent scandal has proven that the IRS must be more transparent about both the information it collects, but also how it manages that information. A commitment on day one to share the data it collects in a machine readable manner would show true leadership by Mr. Werfel and help solidify the Obama administration’s legacy as an open government