Article Critique 4 pages / Article is attached

Information Polity 21 (2016) 153–170 153 DOI 10.3233/IP-160385 IOS Press

The diffusion of civic technology and open government in the United States

John G. McNutta,∗, Jonathan B. Justicea, James M. Melitskib, Michael J. Ahnc, Shariq R. Siddiquid, David T. Cartera and Angela D. Klinea aSchool of Public Policy & Administration, University of Delaware, UD, USA bSchool of Management, Marist College, NY, USA cMcCormack Graduate School, University of Massachusetts, Boston, MA, USA dAssociation for Research on Nonprofit Organizations and Voluntary Action, USA

Abstract. Civic technology is a nascent force in the relationship between governments and communities. Elements of the civic technology ecosystem include open data, related information and communications technology (ICT) innovations and the organizational boundary-spanning practices of civic technology. This paper reports the results of an exploratory study of civic technology adoption by local governments in the United States. The research compares the 113 U.S. city governments recognized for their exemplary fiscal year 2012 popular annual financial reports (PAFRs) with the 49 municipalities in the U.S. state of Delaware that operate websites. Results suggest that a long term commitment to citizen involvement in government data and the size of the community are important predictors of adoption.

Keywords: Civic technology, open data, e-government, e-governance, local government, transparency

1. Introduction

The term civic technology describes an emerging set of practices that proponents assert has the poten- tial to remake communities and local-government organizations throughout the world. Civic technology applications allow citizens to schedule public services, communicate with neighbors and other stake- holders, and participate in government and a wide variety of other tasks, via civic practices that build upon an extensive and growing set of online applications and data. More importantly, it offers the pos- sibility that citizen groups, local nonprofits and businesses and other participants can help to reinvent government from the outside. The promise that communities can be made more responsive and effective at meeting the needs of all their members via coordinated technological and organizational innovations is an appealing one. This article reports on an effort to learn where and to what degree that promise has been realized to date by measuring the adoption and diffusion of civic-technology innovations among municipal governments in the United States.

What is civic technology? Civic technology is, “The use of digital technologies and social media for service provision, civic engagement, and data analysis [in ways that have] the potential to transform cities and the lives of their low income residents” [1, p. 3]. Generally, it includes an array of technology and

∗Corresponding author: John G. McNutt, School of Public Policy & Administration, University of Delaware, UD, USA. E-mail: [email protected].

1570-1255/16/$35.00 c© 2016 – IOS Press and the authors. All rights reserved

154 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

nontechnology interventions including (1) open civic data and transparency; (2) civic and service apps and advanced technology solutions such as municipal general-information and non-emergency service- request call centers (referred to here as 311, which is the telephone number used in the U.S. and Canada for such services) and database-interface standards to facilitate service reporting using other types of ICT (open 311), and open and civic mapping; (3) and organizational innovations such as civic hacking and Hackathons, Code for America fellowships, and similar arrangements [1–4]. While many identify civic technology as referring primarily to a set of information and communication technologies (ICT), it also denotes the social arrangements that make it “work” within the community. These include hackathons, Code for America, civic hackers and so forth, as well as active engagement among the local governments, individuals, and for-profit and not-for-profit organizations involved in developing and implementing civic technology. It is the social-technology dimension of civic technology that potentially represents the most significant change to established institutions and techniques of local governance.

Civic technology, and the forces propelling it, appear as though they could together lead to a revo- lution in the way that governments, community groups and nonprofits relate to each other. While there has been considerable discussion of the relationship between sectors and how the distinctions between sectors have blurred [5–7], developments like civic technology could if fully realized radically alter that relationship. The possibility that sectoral and organizational boundaries can become more permeable or even become indistinct through a merger of sectors and/or radical openness of organizations must be considered. Thus widespread adoption of civic technology, if and when it occurs, might be the next phase of a truly evolutionary process to blur the distinction between institutions in society.

This is an emerging area and one that has yet to develop much of a literature of its own indepen- dently of the work of its advocates. On balance there is a relationship with technology that is already well established and well understood, such as Web 2.0 (also called social media), online communities, ICT-facilitated transparency and open data initiatives, and various aspects of e-government. While none of these innovations are new in and of themselves, their more recent evolution and integration into an emerging system of civic technology, in which open data, technology, and new forms of civic engage- ment mutually reinforce one another, is. The relative dearth of available research and the novelty and ongoing emergence of the civic-technology innovation itself present an opportunity for exploratory re- search aimed at more clearly defining and measuring the phenomenon, and understanding the prospects for its continuing development and diffusion.

The present study sought to address this gap in knowledge by exploring two broad research questions, after first solving a basic methodological problem. The first exploratory research question asked how widely and in what ways civic-technology innovations have been adopted by municipal governments. The second area of exploration used diffusion-of-innovation theory to begin understanding the adoption and spread of civic technology to date and its prospects for continuing diffusion. In order to explore those questions, it was first necessary to devise a reasonably valid and reliable operational definition of civic technology and the instrumentation to measure it. Accordingly, the research began with a recursive process of developing, testing, and refining an operational definition of civic technology and the instru- mentation to measure it through a survey of municipal officials and the content analysis of municipal websites.

This article reports on the development of that definition and instrumentation, and the findings from using the resulting content-analysis instrument to explore and compare the adoption of civic technology by two groups of municipal governments in the United States. The results of the research indicate that our operational definition and content-analysis instrument were a reasonable first step for an emergent area of research, that civic technology has not yet been widely adopted by U.S. municipal governments beyond a

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 155

handful of early adopters, and that patterns of adoption to date are consistent with established theories of the diffusion of innovations. The remainder of the article is presented in four sections, the first of which reviews the available relevant literature relevant to civic technology and the diffusion of innovation, and develops exploratory research propositions. The second section reports the research methods and data for the study, the third section presents findings, and the fourth section provides discussion and conclusions.

2. Review of the literature

Civic technology as a system encompasses a wide variety of interventions aimed at using technology to meet community needs and engender social change. All rely on the existence of open civic data. The arena is still evolving but there have been a number of attempts to codify the field [1,3,8]. This is a difficult task given the evolutionary nature of some of the efforts and the strong degree of symmetry with current movements toward open government, smart cities and evidence based policy and practice. To be clear, civic technology is not about ICT alone. It is also about social interventions aimed at using the technology to reinvent government and communities, and about the open civic data that powers the entire process.

2.1. Civic technology as a cumulative innovation

Civic technology is a recent and still emerging socio-technical innovation which represents a greater- than-the-parts whole – a system – that builds cumulatively upon previous innovations, which are them- selves continuing to develop, in three areas: (1) ICTs, including the Internet, and the social media and Web 2.0 applications and practices made possible by the Internet; (2) open civic data that supports the technology, and (3) the social practices of collaborative governance, open organizing and problem solv- ing that make use of and help to realize the potential of the enabling ICT and open-data innovations. Civic technology as a whole builds upon those foundations by developing the social technologies and practices that transform those underlying innovations into constructive civic action. The three compo- nents of civic technology (open civic data, the underlying information and communications technology, and civic-technology practices) create a system in which these various forces play out in creating the effect. These components may be available independently, but all three are essential for developing a civic technology effort.

2.2. Explaining civic technology

Given its recent emergence, there is as yet little scholarly literature that deals directly with civic technology. It is possible, however, to understand the civic technology phenomenon and its origins and adoption through the lenses of three well-developed and established sets of scholarly literature. First, there is the literature examining e-government and e-governance practices and applications, including recent movements toward new forms of online transparency and open data. Second, there is the literature examining the roughly contemporaneous development and diffusion of new forms of ICT hardware and software, including the emergence and development over the past two decades of now-ubiquitous Web 2.0 and social-media tools. Third, there is a literature exploring the use of those ICT developments to facilitate and enrich civic collaborations by extending established models of citizen engagement, partici- pation, and interactions among individuals and between citizen groups and government organizations. In addition to the literature that defines and shapes civic technology, this research is guided by the literature

156 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

on diffusion of innovations, which provides a means to understand whether, when, and how this type of activity is adopted by municipalities. In the following three subsections, the three literatures presaging civic technology are reviewed in turn, and the implications of those literatures for and their extension to the civic technology innovation itself are discussed. Following this, diffusion of innovation theory is offered as the theoretical framework for the research effort.

2.3. Open civic data as a foundation for civic technology

The development of electronic government or e-government is a worldwide movement that has sig- nificantly affected the conduct of government and the theoretical base of public administration. Starting from modest beginnings, the technology that e-government offers has revolutionized the way that gov- ernment information is provided and how services are delivered [9]. Whether civic technology is simply the next stage in the evolution of e-government or something broader is an open question. Surely some elements (such as transparency, open government, and developments like 311-type call centers and open 311 public database interfaces at the very least) fit within traditional e-government approaches, but the idea of citizens and community organizations developing technology for government may not be so easy to accommodate within a conventional e-government framework. Certainly earlier e-government maturity models, such as those of Layne and Lee [10] and Moon [11], would have a difficult time ac- commodating civic technology, and even the later models, which place far more emphasis on public participation, might not easily encompass the full scope and potential of civic technology. Clearly, if civic technology becomes the force that it might be, new thinking will be needed. Lips’s comment is salient here [12, p. 248].

The critical importance of institutional shaping for e-Government, whether transformational or not, implies that scholars with a research interest in e-Government phenomena should be focusing on how the use of ICTs in the public sector and its external relationships are being shaped by the particular institutional settings, processes, actors, and arrangements.

The development of systems like civic technology has the potential to upset the ordered world that traditional public administration was designed to deal with. If it does nothing else, civic technology will blur the boundaries between government, the nonprofit sector and the proprietary sector, and other subsets and configurations of the public. It will complicate the role of those who want to manage in the same old fashion.

An important development in the e-government arena is open civic data: the practice of making gov- ernment data available to the public in a format that is easy to use and accessible. Open civic data is not only good in itself, it is a critical ingredient of civic technology. The move toward open civic data is part of a larger effort to promote open government and transparency [13–18]. Open civic data can be part of open government but is not, per se open government.

Governments collect a substantial amount of information about all aspects of our lives. This includes data on taxation, voting, population, economic activity, air and water pollution, educational achievement, crime poverty, disease incidence and prevalence and so forth. This information is frequently important in identifying problems and documenting the need for community change or changes in the conduct of government affairs. Some of that data is available easily, while other data is not. There are laws that control the release of some data, while other data is not released for administrative or political reasons. In the past, government data was difficult to obtain and in many cases not available at all.

New efforts are pushing government toward a more open stance on releasing data. Efforts in the United States, the EU and other nations have led to a situation where long held data is being released in

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 157

Table 1 Resources requested in the U.S. City Open Data Census (2015)

Asset Disclosure Budget Business Listings Campaign Finance Contributions Code Enforcement Violations Construction Permits Crime Lobbyist Activity Procurement Contracts Property Assessment

Property Deeds Public Buildings Restaurant Inspections Service Requests (311) Spending Transit Zoning (GIS) Web Analytics

Source: Code for America http://www.codeforamerica.org/blog/2014/02/27/ mapping-the-open-data-landscape/.

substantial quantities. The growth of transparency has been accelerated by technology. Technology tools often make transparency less costly and less time consuming. An example is what the U.S. Public Interest Research Group has termed Transparency 2.0 [19], an effort to make checkbook-level information on government spending available to every citizen via the Internet. This moves past traditional provision of budgets and other expenditure data. It is very unlikely that this system would be feasible without technology.

Table 1 lists the open-data resources that are the benchmarks for the U.S. City Open Data Census [20], a project supported jointly by the Open Knowledge Foundation, the Sunlight Foundation, and Code for America. These are primarily administrative data that have been generated by local governments for many years, but not widely available outside of the government organizations, and now can be made available less expensively and more widely through the use of contemporary ICT tools. Such data, es- pecially if available in machine-readable formats, or supported by user-friendly online visualization and analysis tools, can be useful in a variety of ways. For instance, citizens might use the data to identify neighborhood conditions they need to know about or to identify patterns and problems in crime, neigh- borhood conditions, or service delivery, as a complement to official efforts and/or as a way to check on official performance or accounts of performance and problems.

The technology involved in these tools includes a variety of systems and applications but depends heavily on the technologies frequently described as Web 2.0 or social media, which is the subject of the next subsection.

2.4. Social media and Web 2.0

The extensive and growing literature on Web 2.0/social media is relevant to this exploration of civic technology. These tools to some extent facilitate and anticipate the participation goals of civic technology but are not in and of themselves sufficient to achieve the desired effects. This technology regime grew out of earlier developments in technology and evolved over a number of years [21]. Both social media and Web 2.0 are more popularized and less exact terms, often used inaccurately. The most accepted definition is offered by technology thought leader Tim O’Reilly, who suggests the following definition [22]:

Web 2.0 is the network as platform, spanning all connected devices; Web 2.0 applications are those that make the most of the intrinsic advantages of that platform: delivering software as a continually- updated service that gets better the more people use it, consuming and remixing data from multiple sources, including individual users, while providing their own data and services in a form that allows remixing by others, creating network effects through an “architecture of participation,” and going beyond the page metaphor of Web 1.0 to deliver rich user experiences.

158 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

Table 2 Illustrative Web 2.0/Social media applications

Blogs Wikis Microblogging (Twitter) Social Networking Sites (Myspace, Facebook; Pinterest, Google+) Meetup Podcasting

Image Sharing (Flickr) Videosharing Sites (YouTube, Vimeo) Artificial Worlds (Second Life) Storify Instagram Vine

The applications that are generally referred to as social media or Web 2.0 technologies are a set of stable and well understood systems with some similar characteristics. While they are more interactive than most previous systems, interactivity alone does not qualify a technology for inclusion. All Web 2.0 technologies are based on the Internet (the web as platform to use O’Reilly’s term [23]), allow the harnessing of collective knowledge or collective intelligence by facilitating the creation of user- generated content, and promote software as a service and so forth [22,23]. Table 2 lists a number of examples of Web 2.0/social media applications. These can be combined with each other or additional types of applications to build new systems through a process known as a mashup. The mashup process could yield an almost unlimited supply of Web 2.0 applications.

Social media/Web 2.0 represents much of the current set of technology applications that power civic technology. Earlier technologies, particularly websites and databases, are also critical players. As the move toward larger files of open civic data and more sophisticated applications develops, Web 3.0 (the semantic web) will become even more central.

Civic technology is a system that incorporates and extends the Web 2.0 model into a larger system of collaborative innovation. Leading edge interventions represent the forefront of Web 3.0 develop- ments [24], applications of the Internet of Things [25] and pervasive technology [26]. Both sides of the technology ecosystem work together because the availability of data is fundamental to making the user-oriented applications work [27].

Given the range of government and community problems, it is not surprising that there are a range of efforts under the civic technology umbrella that adapt the tools and approaches of social media/Web 2.0 for solving urban problems. Living Cities [1, p. 3] found that civic technology tools fall into three broad categories: (1) “Improving quality of and accountability in public service delivery,” (2) “facilitat- ing resident-driven improvements to neighborhood quality-of-life,” and (3) “deepening participation in public decision-making.” These technology-based tools run the gamut from applications that can be used on a smart phone to back-end technologies that manipulate billions of data points. In the United States, some of the leading-edge technology tools on the citizen side are service apps, public engagement or participation apps, civic mapping, and 311 and open 311.

Service Applications allow citizens to contact and be engaged in municipal services. Some are offered by government but others are not. One of the earlier and better known of the latter type of service applications is “Fix My Street” (https://www.fixmystreet.com/), which allows participants to report road issues and makes the reports publically available. This is a system that began in the United Kingdom and spread worldwide. FixMyStreet is operated by My Society, a project of the nonprofit UK Citizens Online Democracy. This application identifies issues for local government but also pressures government to take action [28].

As Desouza and Bhagwatwar note, there are a variety of service apps for a wide variety of issues, including political participation, open and civic mapping, civic crowdfunding, neighborhood forums, and data visualization [27]. In the United States, these apps often interface with 311 and open 311.

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 159

The 311 (or 3-1-1) system provides direct, real-time, interactive citizen access to local-government in- formation and services for nonemergency issues. This can be conducted by telephone, mobile apps or other type of contact. This frees emergency services for more pressing matters and builds citizen support for government. The 311 effort is part of the traditional-government response of local gov- ernment and can be a one-way discussion, but open 311 makes for a more participatory conversation (http://www.open311.org/learn/) [29–32]. Open 311 has the potential to involve people in the remedi- ation of social problems much in the way that C. Wright Mills discussed the way that private troubles become public issues [33].

Political participation apps to promote public involvement are a growing aspect of e-government and are used by a wide range of public interest groups (see Desouza and Bhagwatwar [34] for an excellent overview). The technologies in this group are designed to connect citizens to their government.

Open and civic mapping technologies allow communities to crowdsource geographic information and produce spatial collective intelligence [35]. The most well-known example is OpenStreetMap, an open source application that has enjoyed popularity in a large number of community efforts. Another system is provided by Ushahidi (www.ushahidi.com) which provides a set of crowdsourcing and crowdmapping applications that have been used to combat violence and deal with disease outbreaks and disasters [36, 37].

Civic crowdfunding is an emerging idea in both the nonprofit and governmental sectors that extends an already well developed strategy of crowdfunding used in the commercial sector [38]. This allows community members to fund projects that they find interesting and useful. Funds are raised via a tech- nology application that demonstrates what possibilities are available and permits residents to invest in a given service or program.

Neighborhood forums allow residents to share resources and experiences, develop relationships and create social capital. They are also useful in dealing with community problems.

Data visualization tools (including those developed for dealing with big data) are another core civic- technology application. The management and visualization of both open civic data and data created by the use of the tools or provided voluntarily by participants is crucial to the civic technology process. This means the ability to manage, understand and apply data to issues and concerns. These tools, in combi- nation with the existing platforms and open access to public data described in the previous subsection, are critical to the success of many civic technology efforts. In addition to these technologies, however, civic collaborations and social interventions are important as well. Those innovations are discussed in the next subsection.

2.5. Civic technology practices

Collaborations and social interventions separate civic technology from other types of technology- enhanced interventions and from traditional approaches to e-government. Some of these elements are unique to civic technology while others are found in other aspects of intersectoral relationships. Civic technology depends on relationships with both commercial firms (typically technology companies) and nonprofit organizations. The latter include not only formally constituted nonprofit organizations but also informal voluntary associations. Some of these unincorporated associations are virtual organizations and online communities. It is this rich set of networked relationships that makes civic technology happen.

Civic hacking and hackathons are an important part of the civic technology movement [2,4,39–41] and represent a unique form of community engagement and an important way that civil society can redefine government. Civic hackers develop new technology for government and for their fellow citizens.

160 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

Hackathons are events where civic hackers work on issues identified by communities who then take advantage of their skills, enthusiasm and abilities. Hackathons are often associated with contests that provide an opportunity to compete against other civic hackers. The National Day of Civic Hacking (http://hackforchange.org/) might well be the most influential forum for civic hackers but there are many instances of events to promote civic hacking.

Code for America fellowships are another innovation that is strongly associated with civic technol- ogy. Code for America (www.codeforamerica.org) is a nonprofit founded in 2009 by Jennifer Pahlka. The fellowship program is one of their signature efforts. It places technologists with select local gov- ernments for a period of time and a set of intended developments. The fellow creates the technology for the local government using the Code for America design philosophy incorporated in Code for Amer- ica’s seven “design principles for 21st Century government” (https://www.codeforamerica.org/govern ments/principles/):

1. Design for people’s needs 2. Make it easy for everyone to participate 3. Focus on what government can do 4. Make data easy to find and use 5. Use data to make and improve decisions 6. Choose the right technology for the job 7. Organize for results Within this framework, Code for America fellows develop applications that meet citizen and govern-

ment needs. Code for America also sponsors local brigades that put local technologists in touch with governments that need assistance. While Code for America is a U.S.-based nonprofit organization, it has an international effort called Code for All. While there are certainly others, Code for America is clearly a thought leader in the civic technology realm. Its ideas are influential in the overall development of the field.

A final category of collaborative social interventions includes local peer-to-peer groups that bring together technologists and people with other interests to address civic issues with technology. Code for America’s brigades are one such effort. This creates a local community of practice for innovators.

These practices are the third leg of the civic technology stool. They are interrelated parts that together form the nucleus of a force for changing community.

2.6. Technology and the changing nature of civic engagement

A defining characteristic of civic technology is the collaborations among governments, civic groups, nonprofit and business organizations, and individuals that knit together the technology-driven compo- nents in order to create the larger innovation of civic technology as a system. Civic engagement has traditionally meant a limited direct role for citizens to appeal to their government (such as petitions, let- ter writing and voting) and a more powerful indirect method of membership organizations that advocated for their members [42]. The development of public interest groups [43] limited the latter possibility, but community organizations represented and continue to symbolize a robust force in many communities. In the past few decades there has been concern that public engagement is waning in America [44,45]. Peo- ple are becoming less involved in their communities, in local government, and even with their neighbors. This could make having a viable democracy difficult.

On the other hand, other observers have pointed to the growing amount of participation in online engagement and suggested that participation is merely going to other venues [46]. Recent research

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 161

demonstrates that online participation is robust. Whether online participation is as meritorious as tra- ditional efforts is open to argument, but advocacy groups, political parties and local organizations have embraced technology with gusto throughout the world.

Earlier technologies became part of politics and civic engagement in the late 1980s. Even before there was widespread Internet availability there was technologically based community engagement through the Freenet/Community Networking movement. The worldwide advocacy movement MoveOn was founded in the late 1990s, years before social media was available. Web 2.0/social media probably began to be used as a mass political technology in the United States with Howard Dean’s innovative campaign for the Democratic nomination for President in 2004, which made use of many of these new tools. After that, it was an expected part of political organizing. A similar path occurred in much of the world. Probably the most dramatic application was the Arab Spring demonstrations in Northern Africa and the Middle East in the past few years.

While most modern approaches to e-government account for public participation, it is difficult to take into account the evolution of participation in the online world. Research suggests some substantial technology-enabled changes in political participation in the last decade or so [42,47–51].

2.7. Theoretical framework: Diffusion of innovations

Rogers’s diffusion of innovation theory [52] provides the theoretical framework to guide the research design in this exploratory study. This is a well-respected and well understood theory that explains well the phenomena of interest. Rogers considers something to be an innovation if it is new to the system of interest. Rogers and others in the diffusion of innovation school contend that an innovation is commu- nicated to successive population groups via series of communication networks [52–55]. Each of these groups decides to adopt or not. Different innovations are adopted at a quicker or slower rate depending on innovation characteristics. People and organizations go through a series of steps in deciding to adopt an innovation. There is a point where the innovation’s adoption experience reaches critical mass and further adoption becomes easier. On balance, many innovations are lost in the shuffle and never become adopted.

Innovations are introduced by a change agent who comes in from outside of the system. In this case, organizations like Code for America and the Knight Foundation might function in the role of the change agent. The Code for America brigades and fellows are clear examples of this type of activity.

The change agent brings the innovation to a population group called the innovators. Innovators are those who are usually willing to experiment with new ideas. The innovators are crucial to subsequent adoption. These would tend to be municipalities who experimented frequently and probably have a de- cent resource base. This might also be governments in communities with a substantial technology sector. Also governments who have invested in good management practices will be likely to be innovators or early adopters.

The next group is the early adopters. Early adopters are organizations or individuals that, while part of the mainstream population tend to be trendsetters. Rogers [52] observed that this part of the population usually has many opinion leaders. This capacity is critical in bring in the early majority.

The early majority is the first major population group to adopt. They accelerate the adoption of the innovation and bring in the late majority. The last group to consider adopting is the laggards. The process proceeds gradually until a critical mass of the population adopts and the process quickens. Each of these conversions depends on communication networks and the functioning of opinion leaders within those networks.

162 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

What determines how quickly (if at all) an innovation is adopted? Rogers [52] observes that five factors are important: trialability (the ability to try before committing), compatibility (with existing practices), relative advantage, complexity, and observability. This would mean that easier to use technology would adopt at a faster rate. It would also probably mean that organizations with more proficiency and expe- rience with technology would be more likely to adopt. Larger organizations with more slack resources tend to be more innovative but there are a host of additional factors such as structure and environment to consider.

We would expect from diffusion of innovation theory that in a population of municipal governments, different municipalities would be at different points in the adoption process. This means that adoption rates would vary. It would also be likely that different parts of the civic technology bundle would be adopted at different rates. Those technologies that were more complex, less triable, less compatible, with lower perceived relative advantage and lower observability would not be adopted as readily and, in some cases, would be discontinued.

This framework, along with the literature on civic technology, leads to the general proposition that municipal governments will most readily adopt the open-data components of civic technology, given that these are the most familiar and the easiest to implement technologically and organizationally, es- pecially when the data released as electronic versions of traditionally generated static documents (such as PDF files of budgets). Following the same logic, the practices of civic technology are likely to be adopted last and least widely, given that they involve more significant degrees of technological and or- ganizational innovation. Technology applications, which involve relatively smaller leaps from familiar organizational practices of public relations and citizen engagement, are accordingly expected to lie in between. Take-Up rates and earliness of adoption for every component of civic technology are expected to be positively associated with an organizational disposition toward openness and citizen engagement. Finally, organizational size, both as a correlate of slack resources and technical sophistication and as a factor that can lead to deriving greater relative advantage from efficiency improvements through ICT innovations, is expected to be positively associated with the adoption of civic technology.

3. Methodology

This is an exploratory descriptive study of the use of civic technology in a set of municipalities in the United States. The study uses a cross-sectional design. The study employs a multiple methods approach. The study was conducted in 2015.

3.1. Subjects

The subjects were the 113 U.S. cities in 28 states plus the District of Columbia that earned the Government Finance Officers Association (GFOA) Popular Annual Financial Reporting (PAFR) Award (http://gfoa.org/pafr) for fiscal periods that ended in calendar year 2012 (the latest available at the time the sample for this research was developed), and a comparison group of all 49 municipalities in the State of Delaware with a web presence. The PAFR exemplars are a national group of municipalities that have been recognized for their efforts to produce reports that have no purpose other than to inform and engage ordinary citizens. This presumptively selects a comparison group of exemplars that will have varying levels of resources and technical expertise, but share a commitment to transparency and citizen engagement. No Delaware government participates in the PAFR award program. The total number of subjects was 162. Individual municipalities were the unit of analysis.

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 163

The populations of the 113 PAFR award-winning municipal governments ranged from 992 to over 600,000 (Denver and the District of Columbia), with median of 41,011 and mean of 83,871. All 113 PAFR-awardee governments had official websites. Delaware’s municipalities tend to be small: they range in population from 74 to 70,851, with a median population of 1,193 and mean of 4,476. Only three of Delaware’s municipalities had populations over 20,000 as recorded by the 2010 Census of Population and Housing: Wilmington at 70,851; Dover, the state’s capital, at 36,047; and Newark at 31,454 (with about half of those being students at the University of Delaware). Of Delaware’s 57 incorporated munic- ipal governments, 49 had websites in early 2015. The eight Delaware municipalities without websites were among the smallest municipalities in Delaware, ranging from 74 to 774 in population. The smallest Delaware municipality with a website had 122 residents. Eighteen of the Delaware municipalities with websites had fewer residents than the smallest of the PAFR award-winning municipalities.

3.2. Research methods

This is a mixed method study. For the purpose of this analysis, civic technology was operationalized as comprising three dimensions: open civic data, technology, and civic-technology practices. Data was collected on all of the variables within each category for each of the 162 cases in the sample through a content analysis of municipal websites and analysis of data from secondary sources. (The original research design also included an electronic survey of the municipalities. The response rate was very low (15.2%), so the decision was made to drop this element from the research.)

3.2.1. Secondary data analysis Information was collected and analyzed from several external repositories. Data points for the civic-

technology practices category included information for Code for America fellowships and brigades taken from Code for America’s website and information for civic hackathons taken from the National Day of Civic Hacking website. Population data was obtained from the US Census Bureau.

3.2.2. Content analysis of municipal websites An instrument for the analysis of the websites was constructed using available literature. Definitions

of data points for the open civic data category were taken from the U.S. City Open Data Census [20] and for the technology category from the Living Cities report [1]. Careful pretests were conducted and the instrument was iteratively developed, pilot-tested, and refined several times during the first half of 2015.

The data reported below was collected by four trained evaluators, who examined each website for the civic technology attributes of interest during a two-week window in mid-August, 2015. Average pairwise inter-coder reliability across all content-analysis items was 0.79. All website data was coded and cleaned and combined with data from the other data sources. The results were combined for final analysis.

4. Results

The results are presented below in two parts. First we report our findings about the adoption of ele- ments within each of the three dimensions of civic technology: open civic data, technology applications, and civic-technology practices. Finally, we compare the Delaware and PAFR municipalities in terms of their adoption of civic technology, and examine the relationship between municipal size (population) and the adoption of civic technology.

164 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

Table 3 Coded open civic data categories

Component of open civic data Number of sample % Number of websites % cities’ websites with with data that can any information is be downloaded and/or

available in a analyzed online category (coded 1) available (coded 2)

Asset Disclosure 7 4.3 0 Budget 133 82.1 6 3.7 Business Listing 64 39.5 5 3.1 Campaign Finance Contributions 15 9.3 1 0.6 Code Enforcement Violations 8 4.9 2 1.2 Construction Permits 25 15.4 7 4.3 Crime 74 45.7 11 6.8 Lobbyist Activity 12 7.4 0 Procurement Contracts 52 32.1 4 2.5 Property Assessments 31 19.1 7 4.3 Property Deeds 19 11.7 2 1.2 Public Buildings Addresses 28 17.3 4 2.5 Restaurant Inspections 1 0.6 0 Service Requests 33 20.4 5 3.1 Spending 69 42.6 4 2.5 Transit 35 21.6 3 1.9 Zoning 83 51.2 29 17.9 Web Analytics 0 0

4.1. Open civic data

Open civic data remains the bedrock of civic technology, providing the information that makes the technology possible. It also facilitates the collaborations that develop the technology. Coding of open data items was based on the U.S. City Open Data Census [20]. The wording of the definitions was taken verbatim from that project’s top-level description of each item but our coding scheme differed. In our coding scheme, 0 indicates that there is no information relevant to this category available directly from the municipality’s website without charge or restrictions on use; 1 indicates that some kind of relevant information is available at least for viewing online and/or that document files are available to read, print, and/or download without charge; and a score of 2 indicates that within a category of information at least one dataset is available to be downloaded and analyzed/manipulated without charge or restrictions on use. Table 3 reports the results for open civic data.

Data was collected on both (1) data made available in any format and (2) data provided in a format that was machine readable and therefore facilitated further analysis by users. For example, materials provided in PDF often have to be retyped in order to do further analysis. The most frequently available resource is the municipal budget (86.4%), which is most frequently available in non-machine-readable PDF format (82.7%). This is followed by zoning maps (51.2%) and crime reports (45.7%). This included a number of cities that made mapped crime-report data viewable on online maps maintained by third- party vendors, but not in downloadable or analyzable formats. As Table 3 shows, there is comparatively little data available in machine-readable formats. Zoning (17.9%) is the most frequently available in a usable format – typically in the form of downloadable shapefiles usable by geographic information system (GIS) software.

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 165

Table 4 Coded civic technology applications

Technology N % Social Media 114 70.4 Application/Service Provision 53 32.7 Application/Citizen Participation 21 13 Crowdsourcing 13 8 Neighborhood Forum 20 12.3 Crowdfunding 1 0.6 Open or Civic Mapping 8 4.9 Big Data and Data Visualization 5 3.1

Table 5 Coded civic technology practices

Practice N % Civic Competition 1 0.6 Hackathon 8 4.9 Code for America Fellowship 8 4.9 Peer to Peer Group/Brigade 10 6.2

Table 6 Zero-order correlation matrix

Population PAFR award Open civic Technology Civic-technology Civic data practices technology

Population 1 0.343∗∗ 0.427∗∗ 0.414∗∗ 0.738∗∗ 0.560∗∗

PAFR Award 1 0.709∗∗ 0.352∗∗ 0.154 0.675∗∗

Open Civic Data 1 0.398∗∗ 0.275∗∗ 0.922∗∗

Technology 1 0.301∗∗ 0.707∗∗

Civic-Technology Practices 1 0.436∗∗

Civic Technology 1 ∗∗Correlation is significant at the 0.01 level (2-tailed).

4.2. Technology

Table 4 presents the results for technology applications that were made available on and directly findable from cities’ websites. We scored a municipality as having a technology if it was findable on the website by direct observation or by using the website’s internal search functions, Clearly social media/Web 2.0 (70.4%) is the most frequently found application, followed by service applications and neighborhood forums. The other applications were found infrequently.

4.3. Civic-technology practices

The final category encompasses civic-technology practices. Civic competitions were coded from the municipal websites. Hackathons were obtained from the 2015 National Day of Civic Hacking website. Code for America fellowships and brigades were taken from the Code for America Website. Table 5 contains the results. A small number of municipalities had any of the national efforts and most of these were larger cities. Only one municipality had a competition.

4.4. Correlation and regression analysis

Three indexes were created using the results from data collection above: open civic data, technol- ogy, and civic-technology practices. We also created an overall civic technology index that combined

166 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

Table 7 Regression coefficients

Model Unstandardized coefficients Standardized coefficients t Sig. B Std. Error Beta

1 (Constant) 2.565 0.347 7.390 0.000 Population 1.307E-5 0.000 0.372 6.786 0.000

PAFR 4.418 0.442 0.548 9.991 0.000

Note. n = 162. Adjusted R2 = 0.573.

the three subindexes. Zero-Order correlations were computed among the three subindexes, municipal population, and whether a municipality was a PAFR awardee. The correlations among the subindexes, population size, and PAFR-awardee variable are found in Table 6. The three subindexes were all posi- tively correlated with the overall civic technology index and with each other, and all correlations were statistically significant at the 0.01 level. The subindex for open civic data is most strongly correlated with the overall index (0.922), followed by the technology applications (0.707) and civic practices (0.436). Population was strongly and significantly (p < 0.01) correlated with the overall index (0.560) and each of the subindexes (0.427, 0.414, and 0.738 respectively). PAFR-awardee status was strongly and signifi- cantly (p < 0.01) correlated with the overall index (0.675) and with the civic-data (0.709) and technology (0.352) subindexes, but was not significantly correlated with the practices subindex.

An OLS regression model with simultaneous inclusion of all variables was fitted to the overall civic technology variable as the dependent variable, and population and PAFR awards as the Independent variables. The results of this procedure yielded an R of 0.760 with an R2 of 0.578 and an adjusted R2 of 0.573. The F-Test (F = 108.971) was significant (p = 0.000). The regression coefficients are in Table 7. Both predictors are statistically significant, and PAFR Membership has the largest beta weight. Regression collinearity diagnostics were conducted and all values were within normal ranges.

5. Discussion and conclusions

This exploratory study examined the diffusion of civic technology among 113 PAFR-awarded munic- ipalities throughout the United States and a comparison group comprising all 49 municipalities in the U.S. state of Delaware that had websites at the time the research was conducted, none of which had received the PAFR award. Absent a reason to believe the two groups examined comprise uniquely late adopters, the findings reported above suggest that the overall incidence of civic technology adoption is still probably low in the U.S., particularly in smaller communities. It is also likely that larger jurisdic- tions and those predisposed to pursue citizen engagement are leading the way. Clearly within this sample the civic technology practices were more often found in larger communities.

Some open civic data is found in a surprising number of communities, but the coverage is uneven. Budget data is the most frequently provided, but it is often provided in a form (such as a PDF document) where it cannot be manipulated for analysis. We appear to be a long way from the ready provision of open data.

In terms of technology, it is clear that social media/Web 2.0 was the most frequently identified element in the technology dimension for municipalities of all types, followed by service applications. Given the wide use of social networking and other Web 2.0 technologies, this should not be surprising. The adoption of other promising technologies is far lighter.

The statistical analysis suggests that overall civic technology adoption is more strongly associated with the production of an award-winning PAFR than with population size. This suggests at least tenta- tively that the decision to adopt civic technology is driven more by an inclination to engage the public

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 167

meaningfully than by the size of municipalities. Municipalities in the PAFR-awardee group were signif- icantly more likely than the non-PAFR (Delaware) municipalities to make open civic data available on their websites and to employ relevant technologies.

Most of the applications of civic-technology practices are more likely to be found in larger communi- ties. The communities in our study that had hackathons appeared to be larger, although small communi- ties do participate in the National Day of Civic Hacking.

These findings are completely consistent with Rogers’s diffusion of innovation theory [52]. Smaller organizations are less likely to innovate than larger organizations. The latter have more slack resources, a higher level of internal expertise and a larger cushion to absorb any consequences of failure. They also provide better opportunities for change agents – in this case organizations promoting technology – to demonstrate proof of concept. In terms of Rogers’s population groups, they might be considered the innovators or, perhaps, the early adopters. Since the level of adoption is so low, civic technology as an innovation is probably far below critical mass.

5.1. Limitations of the study

There are very few exploratory studies without flaws, and ours is no exception. These must be taken into account when considering the findings. We investigated two purposively selected populations of municipal governments in the U.S. – 113 that had been honored for the quality of their fiscal-year 2012 popular financial reports, and 49 that were located in Delaware and had websites. Accordingly, we cannot generalize our findings about the extent to which civic technology has been adopted to all U.S. munici- palities or to municipal governments outside the U.S. In particular, the association in our data between the PAFR and greater civic-technology adoption could potentially be biased by the composition of the sample. At the same time, however, we do not have reason to believe that our findings concerning overall adoption rates, the dimensions and sub-dimensions of civic technology most frequently implemented, or the positive association of population size and civic-technology adoption would be wildly different if we had used a probability sample of all U.S. municipalities. The potential downward bias in measured overall adoption rates resulting from oversampling smaller governments here may counteract to some extent the potential upward bias of oversampling PAFR winners.

In addition, there were fairly common coding issues with the website analysis. Since these are evolving concepts, it is difficult to construct appropriate categories, define operational measures, and develop valid and reliable instrumentation. This complicated the analysis, although the iterative revision and testing of operational measures and instrumentation did in fact prove useful as a way of clarifying the concept.

As in other instances, the secondary data can have unknown errors and other issues. We used a proxy measure for hackathon participation that is both reasonable and defensible. A comprehensive list of hackathons was not available, so we used participation in the National Day of Civic Hacking as measure of involvement.

5.2. Going forward

The clearest conclusion from this exploratory research is that civic technology is still emerging both as concept and as practice. Enthusiastic proponents have suggested that civic technology holds the potential to improve multiple dimensions of local governance and municipal performance by increasing the flow of usable data from government organizations to citizens and civic groups, the flow of actionable infor- mation and ideas from the outside-in, and the scope and significance of opportunities for meaningful mutual engagement and collaborative governance [1–4,40,50]. Proponents’ definitions of the concept,

168 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

however, have tended to be of the “we know it when we see it” variety and to be broadly aspirational, focused more on the attractive novelty and hoped-for benefits of civic technology than on facilitating rigorous measurement and study of the phenomenon.

This study accordingly took some preliminary steps to refine our understanding of the concept. First, we clarified the common elements in previously observed definitions and practices to specify that civic technology denotes a set of collaborative practices that local governments and external stakeholders pursue in order to engage citizens and civic groups in jointly developing and using ICT tools and appli- cations that make use of open civic data to improve service delivery. Second we developed, tested, and refined a basic instrument for measuring civic technology through content analysis of local-government websites. Third, we used that definition and instrument to assess and compare the adoption of civic tech- nology in two populations of U.S. municipal governments. The results indicated that civic technology is an identifiable and measurable phenomenon, but not yet extensively adopted or institutionalized among U.S. local governments.

Future research is needed to refine the conceptual definition and measurement approach developed by this exploratory research for application in the U.S. and elsewhere, and then, presuming the contin- ued diffusion of the innovation, to begin to test some of the normative claims advanced by proponents (transformative improvements in service delivery, citizen engagement, equity, and so on). Future studies should look at some of the innovative organizational models (such as hackathons, crowdsourcing and other approaches) and evaluate their potential for creating innovative approaches to e-government.

These are exciting times for both practitioners and scholars. In many cases, what we do with civic technology and how we will do it is being created before our very eyes.

References

[1] Living Cities (2012). Field scan for civic technology. Available at https://www.livingcities.org/resources/131-field-scan- of-civic-technology.

[2] Baraniuk, C. (2013). The civic hackers reshaping your government. New Scientist, 218(2923), 36-39. [3] Goldstein, B. & Dyson, L. eds. (2013) Beyond Transparency. Washington, DC: Code for America. [4] Hébert, M. K. (2014). Come Hack with Me: Adapting Anthropological Training to Work in Civic Innovation. Practicing

Anthropology, 36(2), 32-36. [5] Harris, M. (2012). Nonprofits and Business: toward a Subfield of Nonprofit Studies. Nonprofit and voluntary sector

quarterly, 41(5), 892-902. [6] LeRoux, K., & Feeney, M. K. (2014). Nonprofit Organizations and Civil Society in the United States. Routledge. [7] Ott, J. S., & Dicke, L. (Eds.). (2011). Understanding nonprofit organizations: Governance, leadership, and management.

Westview Press. [8] Patel, M., Sotsky, J., Gourley, S., & Houghton, D. (2013). The emergence of civic tech: Investments in a growing field.

The Knight Foundation. [9] West, D. M. (2005). Digital government: Technology and public sector performance. Princeton: Princeton University

Press. [10] Layne, K. & Lee, J. (2001). Developing fully functional E-government: A four stage model. Government Information

Quarterly, 18(2), 122. [11] Moon, M. J. (2002). The evolution of e-government among municipalities: rhetoric or reality? Public Administration

Review, 62(4), 424-433. [12] Lips, M. (2012). E-government is dead: Long live public administration 2.0. Information Polity, 17(3), 239-250. [13] Durose C, Justice JB, Skelcher C. Governing at arm’s length: eroding or enhancing democracy? Policy & Politics. 2015

January; 43(1), 137-53. doi: 10.1332/030557314X14029325020059. [14] Justice, J. B., & Dülger, C. (2009). Fiscal transparency and authentic citizen participation in public budgeting: The role

of third-party intermediation. Journal of Public Budgeting, Accounting & Financial Management, 21(2), 254-288. [15] Justice, J. B. & McNutt, J. G. (2013-2014). Social capital, e-government and fiscal transparency in the states. Public

Integrity, 16(1), 5-24.

J.G. McNutt et al. / The diffusion of civic technology and open government in the United States 169

[16] Justice, J., McNutt, J. & Smith, E. (2011). Holding state governments accountable for the public purse: What institutional, organizational and environmental factors promote fiscal transparency? Paper read at the First Global Conference on Transparency Research. May 19-20, Newark, NJ.

[17] Justice, J. B., & Tarimo, F. (2012). NGOs holding governments accountable: Civil-society budget work. Public Finance and Management, 21(3), 204-236.

[18] Lathrop, D. & Ruma, L. (2010). (ed.) Open Government: Collaboration, Transparency, and Participation in Practice. Sevastopol, CA: O’Reilly.

[19] Davis, B., Baxandall, P., & Pierannunzi, R. (2012, March). Following the money 2012: How the 50 states rate in providing online access to government spending data. Retrieved June 5, 2012, from http://uspirg.org/reports/ usp/following-money- 2012.

[20] U.S. City Open Data Census [Internet]. Cambridge, United Kingdom: Open Knowledge Foundation; 2015 [cited 2015 July 1]. Available from: http://us-city.census.okfn.org/.

[21] Germany, J. B. (ed.) (2006). Person to person to person: Harnessing the political power of on-line social networks and user generated content. Washington, DC. The Institute for politics, democracy and the Internet, George Washington University.

[22] O’Reilly, T. (2005). Web 2.0: compact definition. Message posted to http://radar.oreilly.com/archives/2005/10/web_20_ compact_definition.html.

[23] O’Reilly, T. (2007). What is Web 2.0: Design patterns and business models for the next generation of software. Commu- nications and Strategies, 65(1), 17-37.

[24] Hendler, J. (2009). Web 3.0 Emerging. Computer, 42(1), 111-113. Hoffmann, L. (2012). Data mining meets city hall. Communications of the ACM, 55(6), 19-21.

[25] Atzori, L., Lera, A., & Morabito, G. (2010). The internet of things: A survey. Computer networks, 54(15), 2787-2805. [26] Cook, D. J., & Das, S. K. (2012). Pervasive computing at scale: Transforming the state of the art. Pervasive and Mobile

Computing, 8(1), 22-35. [27] Desouza, K. C., & Bhagwatwar, A. (2012). Citizen apps to solve complex urban problems. Journal of Urban Technology,

19(3), 107-136. [28] King, S. F., & Brown, P. (2007, December). Fix my street or else: using the internet to voice local public service concerns.

In Proceedings of the 1st international conference on Theory and practice of electronic governance (pp. 72-80). ACM. [29] Clark, B. Y., & Brudney, J. L. (2014). Representation and Beyond: Coproduction and the New Information Technology.

Available at SSRN 2473543. [30] Clark, B. Y., Brudney, J. L., & Jang, S. G. (2013). Coproduction of government services and the new information

technology: Investigating the distributional biases. Public Administration Review, 73(5), 687-701. [31] Offenhuber, D. (2015). Infrastructure legibility – a comparative analysis of open311-based citizen feedback systems.

Cambridge Journal of Regions, Economy and Society, 8, 93-112. [32] Suri, M. V. (2013). From Crowdsourcing Potholes to Community Policing: Applying Interoperability Theory to Analyze

the Expansion of “Open311”. Berkman Center for Internet & Society at Harvard University. [33] Mills, C. W. (1959). The sociological imagination. Oxford: Oxford University Press. [34] Desouza, K. C., & Bhagwatwar, A. (2014). Technology-Enabled Participatory Platforms for Civic Engagement: The

Case of US Cities. Journal of Urban Technology, 21(4), 25-50. [35] Budhathoki, N. R., & Haythornthwaite, C. (2013). Motivation for Open Collaboration Crowd and Community Models

and the Case of OpenStreetMap. American Behavioral Scientist, 57(5), 548-575. [36] Goldstein, J., & Rotich, J. (2008). Digitally networked technology in Kenya’s 2007–2008 post-election crisis. Berkman

Center Research Publication, 9. [37] Okolloh, O. (2009). Ushahidi, or ‘testimony’: Web 2.0 tools for crowdsourcing crisis information. Participatory Learning

and Action 59(1), 65-70. [38] Davies, R. (2015). Three provocations for civic crowdfunding. Information, Communication & Society, 18(3), 342-355. [39] Johnson, P., & Robinson, P. (2014). Civic Hackathons: Innovation, Procurement, or Civic Engagement? Review of Policy

Research, 31(4), 349-357. [40] Popyack, J. L. (2014). Prohacktivity, or one giant hack for mankind. ACM Inroads, 5(2), 40-52. [41] Stepasiuk, T. (2014). Civic hacking: A Motivational Perspective. New Visions in Public Affairs. 6, 21-30. [42] Verba, S., Scholzman, K., Brody, H. (1995). Voice and Equity: Civil Volunteerism in American Politics. Cambridge:

Harvard University Press. [43] Berry, J. M. (1977). Lobbying for the people. Princeton: Princeton University Press. [44] Putnam, R. D. (2000). Bowling alone: The Collapse and Revival of American Community. New York: Simon & Schuster. [45] Skocpol, T. (2003). Diminished Democracy: From Membership to Management in American Civic Life. Oklahoma City:

University of Oklahoma Press. [46] Norris, P. (2002). Democratic phoenix: Reinventing political activism. Cambridge: Cambridge University Press.

170 J.G. McNutt et al. / The diffusion of civic technology and open government in the United States

[47] Smith, A. (2008). From BarackObama.com to Change.gov: those active in the Obama campaign expect to be involved in promoting the administration. Retrieved from http://pewresearch.org/pubs/1068/post-election-voter-engagement.

[48] Gibson, R., & Cantijoch, M. (2013). Conceptualizing and measuring participation in the age of the internet: Is online political engagement really different to offline? The Journal of Politics, 75(03), 701-716.

[49] Gordon, E., Baldwin-Philippi, J., & Balestra, M. (2013). Why we engage: How theories of human behavior contribute to our understanding of civic engagement in a digital era. Berkman Center Research Publication, (21), 2008-09.

[50] Guo, C., & Saxton, G. (2013). Tweeting Social Change: How Social Media Are Changing Nonprofit Advocacy, Nonprofit and Voluntary Sector Quarterly 43(1), 57-79.

[51] Smith, A, Schlozman, K., Verba, S. & Brady, H. (2009). The internet and civic engagement. Washington, DC. PEW Inter- net and American Life Projects. Retrieved from http://www.pewinternet.org/Reports/2009/15–The-Internet-and-Civic- Engagement.aspx.

[52] Rogers, E. (2003). Diffusion of innovations. [5th Edition] New York: Free Press. [53] Dillon, A. & Morris, M. G. (1996). User acceptance of information technology: Theories and models. Annual Review of

information science and technology, 31, 3-32. [54] Robinson, M., Swan, J., & Newell, S. (1996). The role of networks in the diffusion of technological innovation. Journal

of Management Studies, 33(3), 333-359. [55] Strang, D., & Soule, S. A. (1998). Diffusion in organizations and social movements: From hybrid corn to poison pills.

Annual Review of Sociology, 22(1), 265-91.

Copyright of Information Polity: The International Journal of Government & Democracy in the Information Age is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.