Electronic Government and Civic Engagement: Citizen Interaction with Government Via Web Portals Y

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Digital Repository @ Iowa State University

Journalism Publications Greenlee School of Journalism and Communication

2006 Electronic government and civic engagement: Citizen interaction with government via Web portals Y. C. Chen Iowa State University

Daniela Dimitrova Iowa State University, [email protected]

Follow this and additional works at: https://lib.dr.iastate.edu/jlmc_pubs Part of the Journalism Studies Commons, and the Social Influence and Political Communication Commons The ompc lete bibliographic information for this item can be found at https://lib.dr.iastate.edu/ jlmc_pubs/7. For information on how to cite this item, please visit http://lib.dr.iastate.edu/ howtocite.html.

This Article is brought to you for free and open access by the Greenlee School of Journalism and Communication at Iowa State University Digital Repository. It has been accepted for inclusion in Journalism Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Electronic government and civic engagement: Citizen interaction with government via Web portals

Abstract This exploratory study examines civic engagement with e-government via Web sites. It provides an analytical framework that integrates both the supply and the demand sides of citizen interaction with e-government. In modeling three dimensions of online civic engagement (government information access, service transactions, and contributing to government policy-making processes), the study framework incorporates a number of variables, including political activism, civic involvement, perceived benefits nda difficulties, information channels, and demographic characteristics. Based on a national sample of Internet users, the study highlights the importance of the supply side (availability of e-government) for promoting civic engagement. Furthermore, political activism is found to be related positively to accessing government policy information and contributing to policy-making processes. The tudys results also confirm the significant impact of perceived benefits in fostering online civic engagement. Future research can benefit from this study by utilizing a more comprehensive model, treating various dimensions of online engagement separately, and conducting an in-depth analysis of the elements of perceived benefits.

Disciplines Journalism Studies | Social Influence and Political Communication

Comments This article is published as Chen, Y. C., & Dimitrova, D. V. (2006). Electronic government and civic engagement: Citizen interaction with government via Web portals, International Journal of Electronic Government Research, 2(1), 54-76. doi: 10.4018/jegr.2006010104. Posted with permission.

This article is available at Iowa State University Digital Repository: https://lib.dr.iastate.edu/jlmc_pubs/7 INTERNATIONAL JOURNAL OF ELECTRONIC GOVERNMENT RESEARCH

January - March 2006, Vol. 2, No. 1 Table of Contents

EDITORIAL PREFACE

IJEGR 2(1) i Donald F. Norris, University of Maryland, Baltimore County, USA

RESEARCH PAPERS

Information Technology and Administrative Reform: Will E-Government be Different? 1 Kenneth Kraemer, University of California, Irvine, USA John Leslie King, University of Michigan, Ann Arbor, USA This article examines the theoretical ideal of information technology as an instrument of administrative reform and examines the extent to which that ideal has been achieved in the United States. It then concludes that information technology has never been an instrument of administrative reform; rather, it has been used to reinforce existing administrative and political arrangements.

The Evolution of Web Governance in the Federal Government 21 Julianne Mahler, George Mason University, USA Priscilla M. Regan, George Mason University, USA This article seeks to describe and to account for the emergence of what is now being termed Web governance. It explores the evolution of the process by which Web governance decisions are being made government-wide and at individual federal agencies.

Generational Differences in Information Technology Use and Political Involvement 36 Mack C. Shelley, II, Iowa State University, USA Lisa E. Thrane, Wichita State University, USA Stuart W. Shulman, University of Pittsburgh, USA A structural equation model analysis of data from a 2003 national random sample survey on informational technology (IT) reveals important direct and indirect effects of generational demographic and attitudinal differences on electronic forms of political participation. Our model suggests that e- citizenry will compound existing social divisions, as non-electronic voices are marginalized and electronic voices are amplified.

RESEARCH NOTE

Electronic Government and Online Engagement: Citizen Interaction with Government 54 via Web Portals Yu-Che Chen, Iowa State University, USA Daniela V. Dimitrova, Iowa State University, USA This exploratory study examines civic engagement with e-government via Web sites. It provides an analytical framework that integrates both the supply and the demand sides of citizen interaction with e- government. 54 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

RESEARCH NOTE Electronic Government and Online Engagement: Citizen Interaction with Government via Web Portals* Yu-Che Chen, Iowa State University, USA Daniela V. Dimitrova, Iowa State University, USA

ABSTRACT

This exploratory study examines civic engagement with e-government via Web sites. It provides an analytical framework that integrates both the supply and the demand sides of citizen interaction with e-government. In modeling three dimensions of online civic engagement (government information access, service transactions, and contributing to government policy-making processes), the study framework incorporates a number of variables, including political activism, civic involvement, perceived benefits and difficulties, information channels, and demographic characteristics. Based on a national sample of Internet users, the study highlights the importance of the supply side (availability of e-government) for promoting civic engagement. Furthermore, political activism is found to be related positively to accessing government policy information and contributing to policy-making processes. The study results also confirm the significant impact of perceived benefits in fostering online civic engagement. Future research can benefit from this study by utilizing a more comprehensive model, treating various dimensions of online engagement separately, and conducting an in-depth analysis of the elements of perceived benefits.

Keywords: e-citizen; e-democracy; e-government; IT adoption; IT in public administration; online civic engagement

INTRODUCTION democracy represents one of the key chal- Engaging citizens online in order to lenges of the next generation of electronic improve governance and to facilitate e- government (Pratchett & Krimmer, 2005).

The first generation of electronic govern- parency (Pandey & Bretschneider, 1997), ment was focused more on providing ac- information and service delivery (Holden cessible information and less focused on et al, 2003), and reforms (Ho, 2002) making transactions available online. A mostly are based on surveys of either gov- critical examination of first-generation e- ernment Web sites or government officials. government projects can highlight some Although some popular studies take a opportunities for improving the next gen- more citizen-centric approach in order to eration of e-government. Surveys of gov- try to understand the demand side ernment Web sites indicate limited use of (Horrigan, 2004; Graafland-Essers & interactive features to engage citizens in for- Ettedgui, 2003), such studies have limited mulating public policy (West, 2004). The analytical methodologies for understand- progress made at the local level in adopting the impact of individual factors. Finally, ing interactive services is slow, based on current e-government studies tend to ig- the analysis of electronic government sur- nore opportunities for integrating insights veys conducted in 2000 and 2002 (Norris from relevant bodies of literature. For ex- & Moon, 2005). Although with the use of ample, public administration and political Internet technologies governments are shift- science researchers often overlook litera- ing away from the traditional bureaucratic ture on information management systems paradigm, providing customer services and rarely integrate social and network rather than enhancing citizen participation dimensions into the study of e-government in policymaking is the focus (Ho, 2002). (Norris & Lloyd, 2004). A study of local government officials also To fill these gaps in existing research, suggests that e-democracy is not high on this exploratory study aims to establish a their agenda for future deployment of elec- framework that takes into account both tronic government (Norris, 2005). the supply and the demand sides of civic Learning from experience and build- involvement in e-government. This study ing on the information and communication sheds light on how making online services infrastructure, governments can introduce available impacts the willingness of citi- meaningful ways to engage citizens in the zens to use e-government and on their policymaking process. One of the main actual utilization of e-government informa- criticisms of current e-government con- tion and services. Moreover, the study in- cerns the top-down bias that impacts de- tegrates insights from allied fields of re- cisions on what type and nature of infor- search, examining the social aspects of mation and services are to be provided civic engagement and analyzing the impor- (McNeal et al., 2003). Thus, e-govern- tance of perceived benefits. Finally, this ment tends either to ignore the citizen’s study employs regression analysis to as- perspective or to misunderstand it. More- certain both the direction and significance over, existing scholarly literature on e-gov- of various factors impacting online civic ernment seems to pay limited attention to engagement, thus moving beyond mere the citizen’s perspective. Studies of trans- descriptive studies.

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 56 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

Figure 1. A conceptual framework for online civic engagement

Availability and competition effect of multiple means of Exposure to information and information channels service delivery Civic engagement on government Social and political Web sites engagement

Citizen willingness Perceived benefits and to engage with difficulties government online

Demographic characteristics

We begin by introducing the pro- online. This supply-demand approach is posed analytical framework and by for- analogous to Pippa Norris’ (2005) politi- mulating hypotheses based on relevant cal market model. Graafland-Essers and theoretical insights. Next, we discuss the Ettedgui (2003) also provide some impor- models and measures used and explain the tant insights into the dynamics of supply data collection and data analysis process. and demand. Supply and demand are seen The discussion of the results includes as jointly determining the levels of engage- theoretical and policy implications of the ment and satisfaction. The framework is findings. Finally, the article concludes with illustrated in Figure 1. key findings and suggestions for future The other main goal of the frame- research. work is to make a clear distinction between willingness to adopt and actual uti- ANALYTICAL FRAMEWORK lization of electronic government informa- The proposed conceptual frame- tion and services. Willingness to adopt does work aims for comprehensiveness by ad- not translate necessarily into actual adop- dressing both the supply and the demand tion. It is particularly the case when the sides of e-government as they impact the interactive features that citizens may de- level of civic engagement. Supply side re- mand are not made available to them. As fers to the availability of e-government a result, the percentage of citizens using services. Demand side refers to citizen transactional services is relatively low. For willingness to interact with government example, online renewal of drivers’ licenses

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006 57 and recreational licenses are at 12% and ernment Web sites, particularly demo- 4%, respectively (Larsen & Rainie, 2002). graphic characteristics emphasized in stud- Several theoretical models are rel- ies on the “digital divide” (Mossberger et evant for understanding how citizens be- al., 2003; Norris, 2005). These studies have when they interact with government also provide some potential classifications online. The Technology Acceptance of dimensions of citizen online engagement Model (TAM) developed and modified (Ho, 2002; West, 2004). in Davis (1986, 1989) and Venkatesh and In the following, we incorporate Davis (2000) shows the importance of the these theoretical insights into the key com- perception of benefits and the perception ponents of the framework while hypoth- of task complexity in determining the ac- esizing the relationships among the com- tual use of information technology. Rogers’ ponents. We shall begin by defining online (2003) diffusion of innovations approach civic engagement as the main dependent puts emphasis on how social systems im- variable in the conceptual framework. pact diffusion and adoption decisions. These models are relevant, because en- Defining Online Civic gaging government online can be concep- Engagement with Government tualized either as acceptance of new tech- In a broad sense, civic engagement nology (i.e., Web-based technologies) or refers to “the participation of individual as innovation adoption (using online func- citizens in the association of civil and po- tions as adopting innovation). litical society” (Brint & Levy, 1999, p. Literature on social capital also pro- 164). The present study focuses on gov- vides certain insights (Lin, 2001). Social ernment Web sites as the vehicle for civic capital is conducive to traditional civic engagement. This approach complements engagement (Skocpol & Fiorina, 1999), without duplicating the efforts of political and it impacts cyber engagement, as well scientists to understand political mobiliza- (Brainard et al., 2003). Literature from tion via the Internet in elections.1 This study mass communication also proves relevant, aims to analyze the factors affecting the reinforcing the notion that initial exposure willingness and actual utilization of e-gov- to multiple information and communica- ernment services that help to engage citi- tion channels creates awareness and may zens in public governance. Thus, in this lead to shifting opinions about an issue study, civic engagement is limited to inter- (Andreasen, 1995; Rice & Atkin, 2001). actions with government online via Web Individuals who are exposed to multiple portals. information and communication channels Further, this study draws from Ho’s are more likely to learn about innovations (2002) scheme that classifies various ac- such as new e-government functions tivities in online civic engagement. Ho (Rogers, 2003; Whitehead, 2000). Re- (2002) argues that there is a hierarchy of viewing e-government research, we can activities. At the most basic level, citizens identify factors affecting the use of gov- conduct passive search for government

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 58 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

information, such as information on coun- voting patterns (Gibson, 2002), and as cil meetings, community calendars, and suggested by patterns in searches for gov- schedules for rule making. This basic level ernment information and getting services of engagement is important, because citi- online (GAO, 2001; Hart-Teeter, 2003; zens can use the information as resources. Horrigan, 2004; Mossberger et al., 2003). Above this level of interaction, citizens Third, education is positively correlated conduct relevant transactions with the gov- to sending comments to government ernment, such as downloading forms, pay- (Larsen & Rainie, 2002) and holding a ing taxes, and so forth. At yet another positive view of engaging government via higher level, citizens participate in govern- the Internet (Shelley et al., 2004). A fourth ment policymaking, providing online com- factor involves race. Minorities are ments and using other types of input underrepresented in Internet voting mechanisms. At the highest level, accord- (Gibson, 2002), and they conduct less ing to Ho (2002), citizens participate in online searches for government informa- government in real time. tion (Mossberger et al., 2003). The present study views such online The relationships suggested by these activities as interlocked dimensions, not studies are based on analyses of both as a hierarchy of levels. Online transac- Internet and non-Internet users. Follow- tions such as paying taxes and processing ing the insights of these studies, we for- permits are rather different from getting mulate Hypothesis 1: public policy information and providing input. There is probably no hierarchical Hypothesis 1 progression in such participation. H1a: A high level of education and income Nevertheless, we must make an im- is positively related to engagement portant distinction, differentiating between with government online. the mere willingness to interact and the H1b: Middle-age and non-minority citi- actual level of interaction with government zens are more likely to exhibit a will- online. This distinction may prove particu- ingness to interact with government larly critical when what government offers online. through its Web portal lags behind citizen demands and expectations. It should be noted that the present study is focused on a broad set of factors Demographic Characteristics determining online engagement with gov- Citizens who tend to interact with ernment, and the inclusion of demographic government online share several common characteristics is mostly for control. There demographic characteristics. First, users is already extensive literature focusing on of government Web sites tend to be middle these variables. Instead, we will concen- age (GAO, 2001; Horrigan, 2004; Tho- trate on less-researched, non-demo- mas & Streib, 2003). Second, these us- graphic factors. ers tend to be affluent, as evident in Internet

Social and Political Involvement stems from two complementary factors: All else being equal, one might ex- the positive role of social influence in shap- pect politically active citizens to be more ing innovation adoption decisions (Lin, likely to engage government online. In- 2003; Lynch et al., 2001; Rogers, 2003) deed, there is a positive correlation be- and the need for involved citizens to en- tween Internet use and political participa- gage government to foster communities. tion (Weber et al., 2003). Such political In terms of the first factor, when citizens participation would include attending pub- are active in a community, they are more lic meetings, writing letters to elected offi- likely to be exposed to innovative ideas cials, and participating in political rallies or are more likely to be innovators them- and speeches (Weber et al., 2003). selves. As a result, these community ac- Internet use includes, for example, surfing tivists are more likely to utilize online in- the Internet for recreation and accessing formation and services that help to further digital libraries, newspapers, and maga- build a sense of community. For instance, zines. Moreover, based on European So- the electronic village at Blacksburg, Vir- cial Survey data, Pippa Norris (2005) has ginia, has seen church groups and clubs shown the importance of prior political actively utilize the online community net- orientation in shaping the demand for elec- work to facilitate communication and col- tronic information on government and poli- laboration with their existing and potential tics. Dimitrova and Chen (in press) also members (Carrroll & Rosson, 1996). found that prior interest in government is In terms of the second factor, en- a significant predictor of e-government gaging government to foster community adoption. Thus, we can expect that a citi- may include, for example, applying for a zen who is more active in politics is more community development grant to construct likely to engage government online. a community center or participating in public meetings. Online networks can facili- Hypothesis 2 tate citizens’ involvement further in dis- A higher level of political activism is cussing about and mobilizing around com- positively related to engagement munity issues (Rogers et al., 1994). The with government online. use of the Public Electronic Network (PEN), an interactive communication sys- In addition to political activism, com- tem, in Santa Monica facilitates the in- munity involvement is also likely to be volvement of disadvantaged segments of positively associated with engaging gov- the population (i.e., the homeless) in de- ernment online. Community involvement liberation of public policy. The experience refers to the level of citizen participation of PEN demonstrates the association be- in community groups such as church or tween community involvement and the use school organizations. A relationship be- of online government services. As govern- tween community involvement and online ment puts more and more information and engagement with government probably services online, the citizens with high com-

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 60 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

munity involvement will be more likely to formation and service offerings access these services. Hence, we arrive (Andreasen, 1995; Rogers, 2003). As a at Hypothesis 3. result, we expect that those who are exposed to multiple information and com- Hypothesis 3 munication channels would be more likely A higher level of community involve- to show a willingness to engage government is positively related to engagement online. ment with government online. Hypothesis 4 Information Channels High-level utilization of information One basic theme of electronic gov- channels is positively related to en- ernment is the utilization of information and gagement with government online. communication channels. As posited by the literature on innovation diffusion, early Perceived Usefulness adopters of innovations tend to utilize more and Complexity information channels than non-adopters Perceived usefulness is important for (Rogers, 2003). Mass media channels are the adoption of new technologies. Based particularly important in creating aware- on the Technology Acceptance Model ness of new services such as e-govern- (TAM) (Davis, 1986, 1989; Venkatesh ment transactions (Andreasen, 1995; Rice & Davis, 2000), perceived usefulness is & Atkin, 2001). These channels include asserted empirically to be an important newspapers, TV, radios, billboards, determinant of adoption. Perceived use- printed materials such as government fulness refers to “the extent to which a newsletters, electronic mails, and Web person believes that using the system will sites. An individual who utilizes many chan- enhance his or her job performance” nels of communication is probably more (Venkatesh & Davis, 2000, p. 187). likely to use government Web portals. Perceived usefulness is a concept Mass media information channels are ef- broad enough that it can be applied to citi- fective not only in creating awareness, but zen interaction with government. Nedovic- also in changing the attitudes of potential Budic and Godschalk (1996), utilizing adopters (ITPC, 2002; Whitehead, TAM, studied the willingness to adopt 2000). geographic information systems at govern- Existing e-government research ne- ment agencies. For individuals, perceived glects to assess the relationship between usefulness in comparison with the old sys- the use of multiple information and com- tem is an important factor in their willing- munication channels and the propensity for ness to use that system. Perceived ben- engaging government online. Diffusion lit- efits could include improvement in job erature shows that individuals with a high performance and convenience. Extending level of utilization are more likely to re- the argument to the context of online civic ceive information about e-government in- engagement, perceived usefulness refers

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006 61 to the belief that interacting with govern- the computer. If software is easy to use, ment online benefits participating citizens. users are more likely to adopt the appli- These benefits may include the conve- cation. Previous computer experience and nience of no waiting in line, avoidance of exposure to technology are also favorable mail delay, and the convenience of access- conditions for individual adoption of geo- ing information and services 24/7. If citi- graphic information systems in a local gov- zens believe they can benefit significantly ernment (Nedovic-Budick & Godshalk, from obtaining online information or from 1996). conducting online transactions on govern- In terms of engaging government ment Web sites, they will be more likely online, there are several sources of po- to use them. Therefore, we expect that tential problems, such as the quality and perceived usefulness is positively associ- accessibility of information (Horrigan, ated with a willingness to interact with 2004) and privacy and security (West, government online. 2004). Usability issues such as navigation and help functions are also critical for the Hypothesis 5 use of e-government services. These can Perceived usefulness is positively be seen as barriers to using e-government related to engagement with govern- information and services. In sum, citizens ment online. would be less willing to engage government online if they perceive problems. On the other hand, perceived problems are likely to reduce the willingness of Hypothesis 6 citizens to interact with government online. Perceptions of problems with e-gov- Individuals may lack the needed comfort ernment information and services level to conduct online transactions. This are negatively associated with en- may stem from personal attitudes or be- gaging government online. liefs (Ajzen & Fishbein, 1980; Lin, 2003). For example, citizens may not wish to Availability and Competition Effects conduct transactions with government Availability of government informa- because of a lack of trust due to perceived tion and services is likely to shape the ex- security problems or privacy concerns. tent to which citizens engage government Moreover, it may stem from a lack of tech- online. At the local level, e-government nological expertise or perceptions of low Web sites still are focused mostly on in- self-efficacy (Bandura, 1997). formation (Norris & Moon, 2005). Online Technical expertise of individuals transactions are rather rare. There are lim- shapes their perception of the severity of ited interactive services on government the problems and difficulties associated Web sites (Hart-Teeter, 2003; Thomas & with adopting a new piece of technology. Streib, 2003). Even the creators of gov- Northrop et al. (1994) show that people’s ernment Web portals do not perceive the computer background affects their use of portals as a vehicle for interaction and

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 62 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

transactions with government but simply teract with government via traditional as a channel for information delivery channels (Horrigan, 2004). When survey (Norris, 2005). As for specific interactive respondents were asked to rank methods features, about 15% of Web sites offer of interaction, communication by tele- areas for posting comments or complaints phone ranked the highest (42%) among (West, 2004). About 7% of Web sites other options such as Web site visit, in- offer broadcasting of government events, person visit, e-mail correspondence, and and about 1% of Web sites offer person- letter writing. The use of government Web alized information gathering targeted di- sites ranked second (29%). Nevertheless, rectly to the attention of the citizen (West, in-person visits and e-mail correspon- 2004). Thus, it is important to qualify the dence remained viable options (20% and limited use of interactive functions accord- 18%, respectively). We will use willing- ing to actual availability. ness to engage government via traditional Availability is the first barrier to ac- channels as a control. tual utilization of government information and interactive services. However, avail- METHODS AND DATA ability is also a matter of awareness. If the user is not aware of the services, he or Models and Measures she cannot utilize them. Some government This study analyzes three dimensions officials have admitted that making the of citizen engagement with government public aware of available services is some- online: accessing information, conducting times a challenging task. We can expect transactions, or providing public policy that perceived availability is positively as- input. Each of the hypotheses proposed sociated with the actual online engagement in the conceptual framework will be tested with governments. against these three dimensions. Further, each dimension then will be tested in two Hypothesis 7 models, using willingness and actual utili- Awareness of information and ser- zation to differentiate between intentions vice availability is positively related and use of e-government information and to engagement with government services. As a result, we have six models: online. the first three models capture willingness to engage with e-government and the next Engaging government online should three models capture actual engagement be put in its proper context. Online en- with government online (see Table 1). gagement is only one of several ways for Willingness is measured by a scale- citizens to interact with government. Citi- based query. Respondents are asked to zens can visit government offices in per- rate the extent to which they agree with a son. They also have the option of writing statement, using a scale of strongly dis- letters or making phone calls. A 2003 na- agree (value = 1) to strongly agree (value tional survey showed that citizens still in- = 5).2 For the context of accessing infor-

Table 1. Operationalization of citizen online engagement with government

Dimensions of Online Matched Survey Items Engagement Getting government information A summation of scores for the following five survey items on online the question, “How often do you look for the following information on government Web sites?” (Scale of 1-5; 1 = never, 5 = very often): (a) Look up contact information (e- mail, phone number) for government offices or officials; (b) seek information about public policies or issues (environment, safety, etc.); (c) get information about elections or voting; (d) get information about public hearings or other policy forums; and (e) get information about the impact of government decisions on your community. Conducting transactions with A summation of scores for the following three survey items government online on the question, “How often do you conduct the following activities on government Web sites?” (Scale of 1-5; 1 = never, 5 = very often): (a) File taxes (state or federal income taxes); (b) purchase or renew licenses (driver’s, business, hunting, fishing, and other); and (c) pay bills or tickets (i.e., utility bills, parking tickets). Providing public policy input to A summation of scores for the following four survey items on government online the question, “How often do you conduct the following activities on government Web sites?” (Scale of 1-5; 1 = never, 5 = very often): (a) Give comments to government officials; (b) express a position on a government policy or initiative online; (c) submit information to assist in ensuring public safety, protecting the environment, and so forth; and (d) file complaints.

mation, the survey statement reads, “I example, for the public policy input vari- would like to search for information on a able, this question is posed: How often government Web site.” For the context of do you conduct the following activities on conducting transactions and providing government web sites? (a) give comments public policy input, the survey statement to government officials; (b) express a po- will begin similarly with the phrase, “I sition on a government policy or initiative would like to,” but each statement uses a online; (c) submit information to assist in different inserted phrase, such as “con- ensuring public safety, protecting the en- duct transactions with government agen- vironment, and so forth; and (d) file com- cies online,” or “participate in government plaints. Each activity is measured on a policy making by providing my opinions scale of 1 to 5 (1 = never, 5 = very often), and comments online.” and the total sum is the measurement of The degree of engagement is mea- degree of engagement. Table 1 provides sured by a score that evaluates both the more details about the other two depen- number of activities and their intensity. For dent variables.

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 64 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

The key independent variables iden- both are critical factors in choosing one tified in the conceptual framework and their channel of communication over another. measures are presented in Table 2. The Standard demographic variables are used measure of political involvement focuses for control. Operationalization of each on both willingness and actual action. variable is provided in Table 2. Community involvement is a measure of the extent to which a citizen is involved in Research Design and Data Collection the community, including number of mem- The target population of this study berships in various organizations such as was Internet users in the United States. fraternities, labor unions, and youth The unit of analysis was the individual groups. The questions are modeled after Internet user. We designed and imple- the U.S. General Social Survey in order mented a two-stage data collection in or- to ensure comprehensiveness in naming der to achieve a sample of these Internet community groups and to take into ac- users and to collect relevant data on the count intensity of involvement. The utili- conceptual constructs operationalized in zation of information channels construct is Tables 1 and 2. Since our target popula- specific to learning about information and tion was Internet users, we felt that it was services accessible on government Web reasonable to reach respondents via the sites. This degree of specificity will let us Internet. Moreover, despite their short- pinpoint the government Web site activi- comings, Internet surveys have become ties as the foci of the framework. an established data collection method and The perceived difficulties construct are considered a viable alternative to tra- measures the degree of perceived ob- ditional survey methods (Dillman, 2000; struction preventing online interaction with Schonlau et al. 2001). the government, as identified by the re- We took several precautions in or- spondents. Similarly, the perceived ben- der to avoid the common pitfalls of Internet efits construct is specific to each dimen- surveys, such as lack of control over the sion of civic engagement. This level of number of times an individual can take the specificity matches the three dimensions survey, whether the survey respondent is of civic engagement. Likewise, the avail- a legitimate individual, and the overall qual- ability construct is specific to each dimen- ity of responses. First, we worked with a sion of civic engagement, and each is mea- professional survey company, Survey sured according to the rating of availabil- Sampling International (SSI), Inc., in or- ity of specific online activities on a scale der to generate a panel of legitimate U.S. of 1 to 5. The competition construct cap- Internet users for the survey. Each person tures the competition effect of alternative in the panel had to be registered with the channels of communication (i.e., personal survey company. We further employed visits to government offices or writing let- commercial online survey software called ters). This is a composite index that takes Opinio to restrict the same individual from into account use and level of satisfaction; taking the survey twice, based on their IP

Table 2. Conceptual constructs of determinants for engaging government online and their measures

Conceptual Survey Items and Operationalization Constructs Political A summation of scores for the following two survey items on the question, Involvement “How much do you agree or disagree with the statement?” (scale of 1-5, 1 = strongly disagree, and 5 = strongly agree): (a) I am interested in politics; (b) I am active in lobbying, elections, and other political activities. Community The product of the total number of membership in the community and the Involvement level of activity. Total number of membership includes those in the following 15 types of organizations: fraternal group, service group, veterans group, political club, labor union, sports club, youth group, school organization (PTA), hobby club, school fraternity/sorority, nationality group, farm organization, literary or arts group, professional society, church organization. The level of activities is measured on a scale of 1 to 5, where 5 means very active, and 1 means very inactive. Utilization of A count of the number of information channels being utilized to learn about Information the information and services provided on government Web sites. The list of Channels channels include the following: newspaper advertisement; TV advertisement; radio advertisement; billboard; printed materials from government (newsletter, payment notice, postcard, etc.); e-mail; ad on a search engine (e.g., Google or Yahoo); online advertisement on a general Web site; online advertisement on a government Web site; mobile advertisement (e.g., putting the government URL on a license plate). Perceived A rating of the following statements (scale of 1-5, 5 = strongly agree, and 1 Benefits = strongly disagree). For information-related models (models 1 and 4), the rating for question (a) is used. For online transaction models (model 2 and 5), the rating for question (b) is used. For public policy input, question (c) is used. Question (a): I could benefit significantly from obtaining online information from government Web sites. Question (b): I could benefit significantly from conducting online transactions with government Web sites. Question (c): I could benefit significantly from submitting my opinions or comments on public policy issues online at government Web sites. Perceived A count of the number of the following barriers identified by the respondent: Difficulties quality of information; accessibility of information; privacy, security (e.g., identity theft); too hard to obtain information; no person-to-person contact; difficult to follow online instructions. Availability and Competition Availability This is a rating of the availability of the information and services relevant to a particular dimension of online engagement (scale of 0-4, 4 = always available, and 0 = not available). For information-related models (models 1 and 4), the score is the rating of availability of public information, such as recreation, businesses, health, and business opportunities. For online transaction models (models 2 and 5), the score is the rating of availability of transactions with government agencies, such as paying taxes, getting driver or professional licenses, getting permits. For public policy input models (models 3 and 6), the score is the rating of the availability of online features in order to voice your opinion on public policy.

(continued on next page)

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 66 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

Table 2. Conceptual constructs of determinants for engaging government online and their measures (cont.)

Availability Survey Items and Operationalization and Competition Competition This is the ranking of various channels that citizens can use to interact with Ranking government. These channels include (a) in person; (b) letter; (c) telephone; (d) e-mail, and (e) government Web site. The ranking is based on a composite index of frequency of use and satisfaction. A ranking score of 5 means the best option, and a score of 1 means the least preferred option. Demographics Income level This is done through ranking: (1) below $20,000; (2) $20,000 – $29,999; (3) $30,000 – $39,999; (4) $40,000 – $49,999; (5) $50,000 – $59,999; (6) $60,000 – $74,999; (7) $75,000 – $99,999; (8) $100,000 – $149,999; (9) over $150,000 Race 1 = Caucasian and 0 = other Gender 1 = male and 0 = female Age Actual age at the time of the survey Level of This is done through ranking: (1) completed some high school; (2) high Education school graduate; (3) completed some college; (4) college degree; (5) completed some postgraduate; (6) master’s degree; (7) doctorate, law, or professional degree

address. Further, the quality of responses achieved two objectives at the first stage. was carefully assessed, based on the time First, we collected additional demographic used to fill out the survey, the required information, such as race, age, income, fields, and the demographic data from the and education, to include in the data analy- survey company to validate with the same sis. Second, we were able to identify those information collection through our online individuals who had previous experience survey. Despite these efforts, it is impor- using online government information and tant to recognize that online surveys, as services. At the second stage, we con- well as traditional surveys, are subject to tacted those individuals to ask more de- non-response error (i.e., potential bias in tailed questions about their online civic the data, if the non-respondents are sig- engagement. The survey invitations were nificantly different from those who com- sent out on February 1, 2005. Two e-mail pleted the survey). reminders were sent out on February 7 On July 11, 2004, SSI sent out via and 10, 2005. A total of 143 valid re- e-mail a total of 5,000 survey invitations. sponses were generated at this stage, The first stage of data collection ended which translates into a response rate of within 10 days. This yielded 447 valid re- approximately 3%. These responses form sponses leading to a response rate of 9%, the basis of our analysis. which is comparable with similar Web surveys (Comley, 1996; Smith, 1997). We

Table 3. Age group breakdown by gender for creating a weighted sample to match Pew data on Internet users

Pew Pew Sample Distribution Data Sample Distribution Data Before After Before After Female Weighting Weighting Male Weighting Weighting 18-24 3% 6% 6% 18-24 1% 8% 8% 25-34 17% 10% 10% 25-34 2% 10% 10% 35-44 9% 13% 13% 35-44 7% 11% 11% 45-54 20% 11% 11% 45-54 9% 10% 10% 55-64 10% 7% 7% 55-64 15% 7% 7% 65+ 3% 3% 3% 65+ 6% 4% 4% 50% 50% 50% 50%

Data Analysis cent distribution of the sample respondents The first analysis we conducted was that belonged to various combinations of to assess the extent to which the resulting gender and age groups before and after sample after the two-stage data collec- weighting. For example, 9% of the sample tion was representative of the U.S. Internet respondents belonged to the female, 35 user population. We benchmarked our to 44 age group before weighting, as op- sample against a national random sample posed to 13% after weighting. After of the U.S. population obtained via the weighting (see after weighting and Pew Pew Internet and American Life Project, columns), the sample distribution be- which included 2,925 Americans age 18 comes representative of Pew data on and over. For the comparison of demo- Internet users for each gender and age graphic characteristics between our combination. sample and the Pew sample, we included Nonetheless, we recognize that our only those Pew respondents who had ac- small sample size as well as sampling process to the Internet. This yielded a sample cess may introduce some biases. Even after of Internet users (n = 1,899). Given the weighting according to age groups and random sampling used, the Internet user gender, the sample data may not be able sample from the Pew study should be rep- to achieve full representation on all de- resentative of the national Internet user mographic characteristics. As a result, this population and can serve as a benchmark. study should be viewed as exploratory in As Table 3 shows, we achieved a nature, and generalizations of the findings representative sample of Internet user to all Internet users should be treated with population by matching our sample to that caution. of the Pew study. In particular, we We then employed OLS regression matched gender and age breakdowns of analysis to test the hypotheses generated the Pew sample. Table 3 includes the per- in the analytical framework for three di-

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 68 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

mensions of online civic engagement both Models 5 and 6 measure the effects of in terms of willingness and actual utiliza- the same set of variables on actual utilization. To address potential multicollinearity tion of online transactions and on provid- problems, we ran collinearity diagnoses ing public policy input, respectively. The for all models. Since all tolerance collinear- adjusted R-squares show substantial ex- ity statistics are above 0.5, we concluded planatory power for each of the six mod- that multicollinearity is not an issue.3 Mul- els. The following discussions present the tivariable regression analysis allows us to general patterns observed in all six mod- assess the significance as well as the di- els and their theoretical and policy impli- rection of each hypothesized relationship. cations. Thus, we are able to identify the independent effect of factors such as political in- Engaging Government Online is volvement while controlling for age, edu- a Multifaceted Phenomenon cation, and other demographics. The first important finding suggests that engaging citizens with government RESULTS AND DISCUSSION online is a multifaceted phenomenon. Each The study results are summarized in dimension of engagement has its own dy- six regression models presented in Table namics. In particular, engaging citizens in 4. The first three focus on willingness, and online transactions with government (i.e., the remaining three capture actual engage- paying taxes) differs from citizens access- ment with e-government. Model 1 exam- ing online information and providing input ines the influence of political involvement, on public policy. Political involvement as community involvement, information chan- measured by interest and level of activi- nels, perceived benefits, perceived diffi- ties in politics is not significantly associ- culties, and demographic characteristics ated with transactions (see Models 2 and on the willingness to conduct information 5). On the other hand, political involve- searches on government Web portals (see ment is strongly associated with both the Table 4). Models 2 and 3 measure the willingness and actual online engagement influence of the same variables on willing- with government to access public policy ness to conduct online transactions and information and provide input (see Mod- willingness to provide public policy input, els 1, 3, and 4). respectively. This finding has both theoretical and Model 4 measures the effects of policy implications. First, it seems impor- political involvement, community involve- tant to differentiate between these two ment, information channels, perceived activities. Public administrators should benefits, perceived difficulties, demo- examine how political involvement may graphic characteristics, availability, and create two different kinds of groups who competition from off-line government ser- interact with government online. Different vices on actual utilization of government strategies may need to be employed for Web portals for information searches. promoting engagement, when the target

Table 4. Models of various dimensions of engaging citizens online with government

Model # Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Dependent Variable Willingness Actual Utilization Public Online Public Public Online Public Info. Transactions Policy Info. Transactions Policy Search Input Search Input Independent Variables Constant 1.717 0.205 0.578 4.321* 1.882 0.521 (.427) (.457) (.427) (1.944) (1.607) (1.496) Political Involvement 0.080** 0.043 0.131* 0.612*** -0.173 0.247 (.032) (.038) (.038) (.147) (.133) (.135) Community -0.015 -0.019 -0.015 -0.037 0.030 0.099* Involvement (.010) (.011) (.011) (.044) (.039) (.039) Utilization of 0.007 0.047 0.009 0.536** 0.127 0.177 Information Channels (.039) (.044) (.044) (.178) (.156) (.155) Perceived Benefits 0.574*** 0.762*** 0.518*** 0.805* 1.242*** 0.568* (.082) (.075) (.080) (.371) (.264) (.279) Perceived Difficulties -0.029 -0.034 0.157*** 0.004 -0.154 0.049 (.041) (.044) (.045) (.185) (.156) (.157) Availability and Competition Availability -0.053 -0.039 0.156* 0.372 1.120*** 0.559* (.084) (.071) (.079) (.385) (.250) (.276) Competition 0.059*** 0.024 0.027 0.273*** 0.116 0.154* (.016) (.018) (.017) (.074) (.063) (.061) Demographics Income Level -0.016 0.081* 0.045 0.040 0.130 0.116 (.029) (.033) (.032) (.133) (.115) (.112) Race (white vs. non- 0.372 0.333 -0.107 0.136 -0.023 -1.387 white) (.223) (.252) (.248) (1.014) (.888) (.870) Gender 0.288* 0.102 -0.002 -0.586 1.057 0.632 (.141) (.156) (.161) (.641) (.549) (.563) Age -0.013** -0.003 0.002 -0.048* -0.025 0.032 (.005) (.005) (.005) (.022) (.019) (.019) Education Level -0.059 0.006 -0.099 -0.089 -0.447* -0.288 (.049) (.056) (.054) (.222) (.196) (.191) Model Fit R-Square 0.587 0.585 0.591 0.470 0.423 0.437 Adjusted R-Square 0.549 0.547 0.553 0.421 0.370 0.385 *, **, and *** denote a coefficient significant at the .05, .01, and .001 level, respectively. groups of online transactions and partici- input. For online transactions (see Model patory governance differ. 5), perceived availability is significantly and positively associated with actual transac- Availability is Critical in tions (e.g., e-filing, purchasing or renew- Determining Utilization ing licenses, paying bills or penalties, etc.). The results also confirm the impor- Similarly, perceived availability is signifi- tance of availability, particularly for online cantly and positively associated with pro- transactions and interactive public policy viding public policy input to government

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 70 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

online in terms of both willingness and ac- Perceived Benefits Are Conducive tual utilization (e.g., expressing a position to Online Engagement on a government policy or initiative, sub- Perceived benefits are a critical fac- mitting information to assist in ensuring tor in determining the willingness and utili- public safety and protection of the envi- zation of e-government. This is the only ronment, etc.) (see Models 4 and 6). factor that shows consistent significance The significance of availability as an in all six models. In almost all models, this independent determining factor is particu- factor has the highest level of significance larly notable when all other factors are compared to other variables. The results taken into account. Perceived availability, show that the more benefits citizens per- although positively related to accessing ceive, the more likely they will be to inter- government information online, is not sig- act with government online. nificant. Perhaps governments are already The importance of perceived ben- doing a fair job in providing some impor- efits lends support to the Technology Ac- tant information online. Thus, in this spe- ceptance Model (TAM). This seems to cific case, availability seems no longer to be a major factor in the adoption of new be an issue. e-government information access and ser- The question of availability is critical vices. When the transaction involves ob- in understanding both the supply and de- taining permits or filing taxes, the perceived mand for e-government Web sites. Fu- benefits may be convenience or a mon- ture researchers should consider treating etary advantage. For engaging citizens in supply and demand simultaneously rather public policy discussions, government than looking at supply or demand individu- must show how such discussions have a ally. If public administrators want to im- bearing on the content and direction of fi- prove their e-government efforts, perhaps nal policy decisions. Otherwise, citizens they may need to communicate better the will not see any benefit in providing policy availability of such services. input, and they will simply stop participat- Availability is an important prereq- ing. uisite in order for citizens to utilize e-government services. Another key consider- Political Activism Drives Online ation is perception. Governments some- Public-Policy Engagement times are being criticized for not being Political activism emerges as a sig- particularly effective in marketing their ser- nificant predictor for seeking policy infor- vices. At times, transactions and options mation and engaging in public governance are available online, but citizens are not online (Models 1, 3, and 4). The relation- aware of them. Governments need to con- ship is positive. In contrast, for conduct- sider using information channels more ef- ing online business transactions, political fectively to address the gap between the activism has no significant bearing (Mod- perception and reality of availability. els 2 and 5). An interpretation of these findings may be offered: political activism

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006 71 gives citizens the basic motivation for seek- mining willingness or actual utilization of ing public policy information and engag- online civic engagement options. When ing government in public policy issues. other factors are taken into account, in- Government Web sites provide new ways come level does not impact searching for for doing so. In comparison, community online government information or provid- involvement does not establish itself as a ing online input into policy. Age plays a significant motivation for engaging govern- role only in information searches but not ment online. Having more and active mem- in transactions or in public policy input. bership in community organizations such Level of education plays a role only in as PTAs, hobby clubs, service groups, and conducting online transactions but not in so forth does not show any significant re- any other online activities. These findings lationship with engaging government suggest that, for our sample of Internet online; the only exception is utilizing online users, demographic characteristics have mechanisms to give comments to govern- rather limited influence on user willingness ment. This suggests that membership in and actual engagement with e-government. community groups, while controlling for Treating age, gender, race, and in- other factors, only plays a role in giving come level as control, the findings dem- government comments online. onstrated that those citizens who are po- The finding puts political activism at litically involved off-line and perceive e- center stage, suggesting that interest in government as personally beneficial are politics as well as actual participation in more likely to use online government. This election and lobbying activities play a finding should be interpreted with caution, much more important role than do gen- however, since it cannot be generalized to eral community involvement. Social capi- the general population. Our sample was tal, as defined by community involvement, limited to computer users who have ac- in general does not translate directly into cess to the Internet already; in other online civic engagement. Government words, higher income people who are not Web sites seem to serve more as exten- representative of the population at large. sions of traditional methods of political One possible explanation is that once a participation rather than extensions of com- citizen has become part of the online community building. If government aims to fos- munity, factors such as perceived benefits ter online public policy discussions, one and political activism seem to play impor- important way is to augment political ac- tant roles. At the same time, demographic tivities with electronic options. characteristics become less significant.

Demographic Characteristics CONCLUSION of Internet Users This exploratory study is an impor- Have Limited Relevance tant first step in understanding the supply Our findings show that demographic and demand sides of online civic engage- characteristics play a limited role in deter- ment. The proposed framework incorpo-

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 72 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

rates both the availability and competition portant to note that availability becomes effects of various communication channels, less of a factor for engagement when suf- aiming to provide a more comprehensive ficient amounts of information and types view of online civic engagement. The of services are made available. framework combines several dimensions Future research can build on the con- of online engagement — e-government ceptual framework proposed in this study, information searches, online business which integrates insights from various transactions, and contributing public policy fields. One fruitful area of research will be input. The study is limited by the small to explore the various elements that con- sample size and its focus only on Internet stitute perceived benefits. When govern- users. ments learn from citizens, realizing that the This study, nevertheless, highlights connection between policy input and final the importance of several key factors in policy decision is the most important ele- explaining both the willingness and actual ment, governments will need to make that utilization of various online civic engage- link a priority and communicate this to its ment options. Perceived benefits seem to citizens. Another avenue for future re- be a key factor in explaining the intensity search is to differentiate between the three of online civic engagement. This validates levels of government and to test the mod- the relevance of the Technology Accep- els with longitudinal data in order to see tance Model. In terms of policy recom- how perceived availability shapes actual mendations, public managers need to have utilization over time. This approach will a better understanding of the specific kinds provide insights into the evolution and dy- of benefits that citizens want. namics of supply and demand as they im- Political involvement off-line pact online civic engagement. emerged as a critical factor in explaining why citizens seek public policy informa- REFERENCES tion and provide public policy input online. Ajzen, I., & Fishbein, M. (1980). Under- In comparison, general community in- standing attitudes and predicting volvement does not play as significant a social behavior. Englewood Cliffs, NJ: role as originally expected. E-government Prentice-Hall. efforts need to take this into consideration Andreasen, A. R. (1995). Marketing so- in order to engage citizens effectively cial change. San Francisco: Jossey- online. Bass. Finally, perceived availability of e- Bandura, A. (1997). Self-efficacy: The government services seems to encourage exercise of control. New York: W.H. citizens to interact with government online. Freeman. Government can bridge the availability gap Brainard, L. (2003). Citizen organizing in by providing more online business trans- cyberspace: Illustrations from health actions and more opportunities for online care and implications for public admin- public policy input. In addition, it is im- istration. American Review of Public

Administration, 33(4), 384-406. Science Quarterly, 116(4), 561-583. Carroll, J. M., & Rosson, M. B. (1996). Graafland-Essers, I., & Ettedgui, E. Developing the Blacksburg electronic (2003). Benchmarking e-govern- village. Communication of the ACM, ment in Europe and the US: Statisti- 39(12), 69-74. cal indicators benchmarking the in- Comley, P. (1996). The use of Internet formation society (SIBIS). Santa as a data collection method. Paper Monica, CA: RAND. presented at the ESOMAR, Edinburg. Hart-Teeter. (2003). The new e-govern- Retrieved August 20, 2004, from http:/ ment equation: Ease, engagement, /www.virtualsurveys.com/news/papers/ privacy and protection. Retrieved Au- paper_9.asp gust 15, 2004, from http://www.cio.gov/ Davis, F. D. (1986). A technology ac- documents/egovpoll2003.pdf ceptance model for empirically test- Ho, A. T. (2002). Reinventing local gov- ing new end-user information sys- ernments and the e-government initia- tems: Theory and results. Doctoral tive. Public Administration Review, dissertation, MIT, Sloan School of 62(4), 434-444. Management, Cambridge, MA. Horrigan, J. (2004). How Americans get Davis, F. D. (1989). Perceived usefulness, in touch with government: Internet perceived ease of use, and user accep- users benefits from the efficiency of tance of information technology. MIS e-government, but multiple channels Quarterly, 13(3), 319-339. are still needed for citizens to reach Dillman, D. (2000). Mail and Internet agencies and solve problems. Wash- surveys (2nd ed.). New York: John ington, DC: Pew Internet & American Wiley and Son. Life Project. Dimitrova, D. V., & Chen, Y. C. (in press). Indiana Tobacco Prevention and Cessa- Profiling the adopters of e-government tion (ITPC) Board. (2002). Seeing is information and services: The influence believing: How exposure to ITPC’s of psychological characteristics, civic media campaign affects tobacco mindedness, and information channels. knowledge, attitude, and beliefs Social Science Computer Review, among Hoosier adults. Retrieved Feb- 24(2). ruary 25, 2004, from http:// General Accounting Office (GAO). www.in.gov/itpc/files/research_80.pdf (2001). Characteristics and choices Larsen, E., & Rainie, L. (2002). The rise of Internet users. Washington, DC. of the e-citizen: How people use gov- Retrieved September 24, 2005, from ernment agencies’ Web site. Washing- http://www.gao.gov/new.items/ ton, DC: Pew Charitable Trust. Retrieved d01345.pdf June 1, 2004, from http:// Gibson, R. (2002). Elections online: As- www.pweinternet.org sessing Internet voting in light of the Lin, C. A. (2003). An interactive commu- Arizona democratic primary. Political nication technology adoption model.

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 74 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

Communication Theory, 13(4), 345- cember 12-15). The scholarly litera- 365. ture on e-government: Characteriz- Lin, N. (2001). Social capital: A theory ing a nascent field. Paper presented of social structure and action. Cam- at the Pre-ICIS (International Confer- bridge: Cambridge University Press. ence on Information Systems) Work- Lynch, P. D., Kent, R. J., & Srinivasan, shop, Washington, DC. S. S. (2001). The global Internet shop- Norris, P. (2001). Digital divide: Civic per: Evidence from shopping tasks in engagement, information poverty, twelve countries. Journal of Advertis- and the Internet worldwide. New ing Research, 41(3), 15-23. York: Cambridge University Press. McNeal, R. S., & Tolbert, C. J., Norris, P. (2005). The impact of the Mossberger, K., & Dotterweich, L. J. Internet on political activism: Evidence (2003). Innovating in digital government from Europe. International Journal of in the American states. Social Science Electronic Government Research, Quarterly, 84(1), 54-70. 1(1), 20-39. Menard, S. (1995). Applied logistic re- Northrop, A., Dunkle, D., Kraemer, K., gression analysis. Thousand Oaks, & King, J. (1994). Computer, police CA: Sage. and the fight against crime: An ecology Moon, M. J. (2002). The evolution of e- of technology, training, and use. Infor- government among municipalities: mation and the Public Sector, 3(1), Rhetoric or reality. Public Administra- 21-45. tion Review, 62(4), 424-433. Pandey, S., & Bretschneider, S. (1997). Mossberger, K., Tolbert, C., & Stansbury, The impact of red tape’s administrative M. (2003). Virtual inequality: Be- delay on public organizations’ interest yond the digital divide. Washington, in new information technologies. Jour- DC: Georgetown University Press. nal of Public Administration Research Nedovic-Budic, Z., & Godschalk, D. R. and Theory, 7(1), 113-30. (1996). Human factors in adoption of Pew Internet & American Life Project. geographic information systems: A lo- (2004, May 24). How Americans get cal government case study. Public Ad- in touch with government. Retrieved ministration Review, 56(6), 554-567. July 22, 2004, from http:// Norris, D., & Moon, J. (2005). Advanc- www.pewinternet.org/pdfs/PIP_E- ing e-government at the grassroots: Gov_Report_0504.pdf Tortoise or hare? Public Administra- Pratchett, L., & Krimmer, R. (2005). The tion Review, 65(1), 64-74. coming of e-democracy. International Norris, D. F. (2005). Electronic democ- Journal of Electronic Government racy at the American grassroots. Inter- Research, 1(3), i-iii. national Journal of Electronic Gov- Rice, R. E., & Atkin, C. K. (Eds.). (2001). ernment Research, 1(3), 1-14. Public communication campaigns Norris, D. F., & Lloyd, B. (2004, De- (3rd ed.). London: Sage.

Rogers, E. (2003). Diffusion of innova- acceptance model: Four longitudinal tions (5th ed.). New York: Free Press. field studies. Management Science, Rogers, E., Collins-Jarvis, L., & Schmitz, 46(2), 186-204. J. (1994). The PEN project in Santa Weber, L., Loumakis, A., & Bergman, J. Monica: Interactive communication, (2003). Who participate and why? An equality, and political Action. Journal analysis of citizens on the Internet and of the American Society for Informa- the mass public. Social Science Com- tion Science, 45(6), 401-410. puter Review, 21(1), 26-42. Schonlau, M., Fricker, R. D., & Elliott, West, D. (2004). E-government and trans- M. N. (2001). Conducting research formation of service delivery and citi- surveys via e-mail and the Web. Re- zen attitudes. Public Administration trieved August 18, 2004, from http:// Review, 64(1), 15-27. www.rand.org/publications/MR/ Whitehead, D. (2000). Using mass me- MR1480/ dia within health-promoting practice: A Shelley, M., Thrane, L., Shulman, S., nursing perspective. Journal of Ad- Lang, E., Beiser, S., Larson, T., et al. vanced Nursing, 32(4), 807-816. (2004). Digital citizenship. Social Sci- Witte, J., Amoroso, L., & Howard, P. ence Computer Review, 22(2), 256- (2000). Method and representation in 269. Internet-based survey tools: Mobility, Skocpol, T., & Fiorina, M. (1999). Civic community, and cultural identity in Sur- engagement in American democracy. vey2000. Social Science Computer Washington, DC: Brookings Institution Review, 18(2), 179-195. Press. Smith, C. B. (1997). Casting the net: Sur- ENDNOTES veying an Internet population. Journal * This study was funded by an internal of Computer-Mediated Communica- grant from the Institute of Science and tion, 3(1). Retrieved August 20, 2004, Society (ISS) at Iowa State University. from http://www.ascusc.org/jcmc/vol3/ The authors would like to thank ISS issue1/smith.html for its support. Thomas, J. C., & Streib, G. (2003). The 1 For an illustrative example, see Brainard new face of government: Citizen-initi- et al. (2003). ated contacts in the era of e-govern- 2 The online survey did not differentiate ment. Journal of Public Administra- between the three levels of government. tion Research and Theory, 13(1), 83- 3 The tolerance statistics are in the range 102. of .5 and 1.0. They are well above .2, Venkatesh, V., & Davis, F. D. (2000). A a number below which would indicate theoretical extension of the technology a potential problem (Menard, 1995).

Copyright © 2006, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 76 International Journal of Electronic Government Research, 2(1), 54-76, January-March 2006

Yu-Che Chen is an assistant professor of eGovernment and public management at Iowa State University. He is the lead faculty member of the Public Policy and Administration Program’s eGovernment concentration. Dr. Chen’s current research interests include e-governance, management and implementation of e- government projects, and the role of information technology in collaborative public management networks. He has published in Public Performance and Management Review, Government Information Quarterly, Public Administration Quarterly, and International Journal of Electronic Government Research. In addition, he has published management reports on IT outsourcing issues with the IBM. His national and international service appointments include the National Association of Schools of Public Affairs and Administration’s Information Technology Committee and the International Advisory Board of the Encyclopedia of Digital Government.

Daniela V. Dimitrova received her PhD in mass communication from the University of Florida. Currently she is an assistant professor in the Greenlee School of Journalism and Communication, Iowa State University. Her research focuses on the role of new media technologies in political communication. Dr. Dimitrova’s research has been published in journals such as New Media & Society, Journalism Studies, Telecommunications Policy, and Gazette: The International Journal for Communication Studies.