Adoption of Electronic Technologies in Local U.S. Governments

Introduction

E-government initiatives, or digital interactions between government and citizens, businesses, employees, and other government entities, are often pursued by local governments with the promise of improving the delivery of services to citizens and encouraging civic engagement and participation (Holzer & Manoharan, 2008). While it is important to understand the use of technology to advance efficiency in public organizations, it is increasingly important to understand the ways in which different types of digital technologies are shaping government work and why some governments adopt some technologies and not others.

Previous research has sought to understand how e-government approaches advance the goals of efficiency and community engagement and civic participation at federal, state, and local levels (Edmiston, 2003). The adoption of e-services, such as online job applications or online payment for services, provides local governments with potential savings through the reduction of implementation costs and the downsizing of human resources (Edmiston, 2003; Moon, 2002; Moon & Bretschneider, 2002). In comparison, the adoption of communication technologies, such as social networking tools, allow local governments to engage in one- and two-way communication with citizens, businesses, and other governments (Ahn, 2011; Moon, 2002). A large number of local governments have adopted and implemented a mixture of e-services and communication technologies as a means to improve managerial efficiency, create democratic potential, and transform government activities and interactions with citizens (Ho, 2002; Holzer & Manoharan, 2008; Moon & Norris, 2005). In much of the e-government literature, governments are expected to offer both e-services (digital government) and communication technologies (for enhancing communication and citizen participation; Holzer & Manoharan, 2008), however, in general, governments invest more of their efforts in providing e-services than communication technologies. One reason for the imbalance of providing these two types of e-government is the ways in which e-services and communication technologies are differently connected with organizational performance. In general, e-services, online services intended to serve or deliver services to citizens, have unambiguous goals and are clear mechanisms for improving efficiencies. In addition, the cost savings associated with providing services online are more easily quantified than the intended and unintended outcomes associated with communication technologies.

Fountain (2001) argues that “Cost savings and the benefits of increased public access to information and services, however, represent only a small subset of the promise of digital government. More important, however, is public dialogue about how digital government will be designed and implemented. The central issues are democratic in nature, rather than simply economic.” Essentially, researchers need to look at the factors that influence local government to adopt e-government technologies for promoting not just efficiency, but also civic engagement (Kakabadse, Kakabadse, & Kouzmin, 2003). We agree with Fountain and Kakabadse et al.; research needs to focus on the extent to which local governments adopt e-services for improving service provision and communication technologies for enhancing civic engagement. Specifically, it is increasingly important to separate these two types of e-government initiatives: e-services and communication technologies and understand what drives the adoption of these types of technologies. Unfortunately, there has been little empirical research investigating the determinants of adopting these two different types of e-government technologies.

Many e-government studies emphasize the organizational factors that enable or prevent the adoption of e-government initiatives in public organizations. For example, understanding the motivation and capacity required for an organization to adopt an intranet system in order to improve work efficiency or effectiveness (Moon & Bretschneider, 2002; Pandey & Bretschneider, 1997; Welch & Pandey, 2007). Typically, these studies focus on e-services which have more tangible and easily measured outcomes such as decreased time in processing forms, increased efficiencies, and less waste.

In comparison, little academic research has focused on understanding local governments’ adoption of communication technologies such as social networking tools. Communication technologies may enable local governments to better interact with one another and citizens, but those interactions may also increase costs and reveal more ambiguous and complex challenges for local governments and managers. For instance, a government might register a Facebook page or a Twitter account to announce policies and allow citizens to discuss policy issues but this two-way communication may create management difficulties, by slowing decision-making processes, muddling the process, or making it difficult for managers to assess the importance and voracity of online comments. Since the adoption of communication technologies requires more advanced understanding of e-government, more sophisticated information technology capacity, and a concerted effort to enhance interactions with the public, it is increasingly important to understand the determinants of government adoption of these types of communication technologies. As the array of e-government technologies have grown to include e-services and communication technologies, it is important to understand how different types of technologies might be motivated by different organizational characteristics and incentives. We start with the following question: Is there a distinction between the determinants for the adoption of e-services and the adoption of communication technologies in local governments? And if so, what organizational and managerial factors explain local government adoption of e-services and the adoption of communication technologies?

In the following section, we draw from the literature to introduce five hypotheses about the relationships between personnel constraints, work routineness, organizational centralization, and external pressures and the adoption of e-services and communication technologies in local governments. Second, we describe the data and methods used to test the hypotheses. Third, we present the results from four negative binominal regression models. We find that the predictors of the adoption of e-services and communication technologies are significantly different. The determinants of e-services include citizen demands and organizational factors such as personnel constraints, work routineness, and organizational centralization. In comparison, the determinants of adopting communication technologies are external to the organization: citizen demands and external influences. We conclude with a discussion of the results.

Literature and Hypotheses

Studies of the adoption and use of e-government technologies are rooted in the policy innovation adoption and diffusion literature (Berry & Berry, 1999; Damanpour & Schneider, 2009; Hage, 1999; Rogers, 2003) and the technology innovation literature (Tornatzky & Fleischer, 1990). Most scholars interested in the adoption of e-government technologies combine the two literatures (Ahn, 2011; Ho & Ni, 2004; Moon & Bretschneider, 2002; Moon & Norris, 2005; Nelson & Svara, 2011; Welch & Rainey, 2007) and typically focus a single innovation adoption, such as a system, a website (La Porte, Demchak, & Weare, 2005; Weare, Musso, & Hale, 1999), or the number of e-government web functions (Ho & Ni, 2004; Moon & Norris, 2005; Tolbert, Mossberger, & McNeal, 2008; Nelson & Svara, 2011). In addition, most of the research on innovation adoption is sensitive to the matter of how innovations are operationalized (Damanpour & Schneider, 2009; Downs & Mohr, 1976) and investigates (1) the relative financial expenditure associated with each new practice and (2) the relative impact of each practice on local government performance (Damanpour & Schneider, 2009).

More recently, however, Ahn (2011) has argued that it is increasingly important to distinguish between e-services and communication technologies. Ahn argues that the goal of e-services adoption is often for cost savings but the adoption of communication technologies aims to respond to public needs. The focus on costs and performance is especially relevant when considering the adoption of e-services that seek to enable government to do more with less and use technology to increase efficiency, and thus are often a response to reinventing government and other reform efforts. In comparison, the adoption of communication technologies may aim to increase information dissemination, communication with stakeholders, and public input into government activities. Communication technologies ask government employees and managers to do more by using technology, rather than doing more with less. Thus, the relative advantages, or characteristics of innovations, may be one of the vital factors that differentiate between the decisions to adopt e-services as compared to communication technologies (Berry & Berry, 1999; Damanpour & Schneider, 2009; Rogers, 2003).

Unfortunately, previous e-government research pays little attention to distinguishing between the adoption of communication technologies and e-services in local governments. Based on previous research (Ahn, 2011) and the understanding that the motivations for adopting e-services are distinct from those of communication technologies, we expect to find a significant distinction between the adoption of e-services and the adoption of communication technologies in local governments.

In addition, we investigate the relationships between organizational characteristics and constraints and the adoption of e-government technologies in local governments. Organizations can constrain innovation adoption in many ways; in particular, when an organization’s capacity is structurally constrained it is inert and slow to adopt changes (Cohen, Olson, & March, 1972; Hannan & Freeman, 1989). Structural inertia impedes an organization from adopting innovation because employees may resist change and inhibit fundamental change over time in terms of technology, consumers, or goals (Hannan & Freeman, 1989). Hence the barrier between structural inertia and innovation adoption may be mediated by implementation costs and decision-making. Thus, when managers make decisions about innovation adoption, managerial perceptions of the implementation costs of adopting innovations are critical to determining whether or not they resist and overlook or promote and adopt innovations (Damanpour & Schneider, 2009).

Previous studies have shown that many internal factors may influence the adoption of innovations including organizational centralization, work routineness, personnel constraints, organizational formalization, organizational complexity, and organizational culture (Ahn, 2011; Pandey & Bretschneider, 1997; Tornatzky & Fleischer, 1990). In this article, we explore the extent to which personnel red tape (Moon & Bretschneider, 2002; Tornatzky & Fleischer, 1990), work routineness (Aiken & Hage, 1971), organizational centralization (Aiken & Hage, 1971; Rogers, 2003; Tornatzky & Fleischer, 1990), and external pressures (Cohen & Sauermann, 2007; Rogers, 2003; Yang & Callahan, 2007) are related to the adoption of e-services and communication technologies.

Public administration research often notes that one of the most important distinctions between public and private organizations is that governments tend to have higher levels of rules, red tape, and personnel constraints (Baldwin, 1990; Feeney & Rainey, 2010; Rainey, 1983; Wilson, 1989). Despite numerous reforms to pay structures, personnel rules continue to constrain public organizations from quickly hiring, firing, and promoting employees. Research overwhelmingly indicates that public managers report higher levels of personnel constraints than those working in the private and nonprofit sectors (Baldwin, 1990; Bozeman & Bretschneider, 1994; DeHart-Davis & Pandey, 2005; Feeney & Rainey, 2010).

Because personnel constraints are ubiquitous in government and managers repeatedly report that personnel constraints (or red tape) prevent them from acting efficiently and effectively, it follows that personnel constraints, which are often associated with bureaucratic control, delays, and extensive paperwork, would be negatively related to a local government organization’s ability to adopt innovations. In contrast, some research has found that perceptions of red tape are positively related to information technology innovations (Moon & Bretschneider, 2002) and an organization’s interest in new technology (Pandey & Bretschneider, 1997). Moon and Bretschneider (2002) argue that these findings might be explained by the fact that red tape encourages an organization to solve problems and reduce transaction costs by seeking innovative solutions such as new technology. Thus, it is unclear whether personnel constraints inhibit or motivate the adoption of e-government technologies.

As noted earlier, the adoption of e-government technologies may enhance work processes, as in the case of e-services, but might also increase contact and communication with multiple stakeholders and slow decision-making, in the case of e-communications. Because the adoption and use of e-government technologies requires organizational and managerial flexibility, we expect that personnel flexibility and a low emphasis on work rules will facilitate the adoption of new ideas and innovations of all types (Damanpour, 1991; Hage, 1999; Tornatzky & Fleischer, 1990). Thus, we expect that increased perceptions of personnel constraints will be negatively associated with the adoption and use of electronic technologies.

The literature indicates that work environment and work tasks are related to innovation adoption and change in organizations (Aiken & Hage, 1971). In particular, work routineness, defined as “the level of variety in an organization’s work task” (DeHart-Davis & Pandey, 2005, p. 140), can create obstacles to innovation. In an organization with a diversity of task structures, or lower levels of work routineness, innovations or new ideas have a higher propensity to be triggered since individuals have increased opportunities to meet someone with heterogeneous expertise (Aiken & Hage, 1971). The robust findings on heterogeneous knowledge indicate that individuals in contact with more heterogeneous knowledge will be more innovative (Powell & Grodal, 2006). Similarly, individuals would have lower opportunities to innovate or adopt new ideas when they are assigned routine tasks and deal with the same affairs every day. Thus, we expect that public managers with a high level of work routineness will report lower adoption and use of e-government technologies, while those with a low level of work routineness will report higher adoption and use of e-government technologies.

We also expect that organizational centralization will be related to the adoption of e-government technologies in local governments. Organizational centralization refers to the centralized decision-making structure in an organization (Rogers, 2003). In a highly centralization environment, organization members have less power to make decisions or participate in decision-making processes. A highly centralized decision-making structure will constrain innovations because it creates an environment with a low level of participation among organizational members, thus reducing the amount of heterogeneous information available to individuals. In comparison, a disperse decision-making structure may facilitate innovations because the structure engages individuals in decision-making, increases knowledge and information sharing, and allows flexibility to achieve individual and organizational goals (Aiken & Hage, 1971; Damanpour, 1991, 1996; Rogers, 2003; Tornatzky & Fleischer, 1990). The literature overwhelmingly shows that organizational centralization has a robust and negative relationship with innovation adoption (Damanpour, 1991; Hage, 1999). Therefore, we expect that public managers in more centralized decision-making environments will report decreased adoption and use of e-government technologies.

While we expect organizational factors such as personnel constraints, work routineness, and centralization to be related to innovation adoption, it is also important to consider factors external to the organization that may be related to the adoption of e-government technologies. As noted earlier, governments typically use e-services to deliver online services to the public and utilize communication technologies to communicate with the public. Both functions aim at connecting governments with the interests of external stakeholders and the community. The literature notes that stakeholders can affect public managers’ decision-making in two ways. First, public managers are assumed to be passive or neutral, with managers seeking to maximize personal benefits, making stakeholder influences critical for pushing public managers to respond to community needs (Yang & Callahan, 2007). These influences might come from nongovernment or political actors or state, court, and federal agencies (Welch & Pandey, 2007). In the second case, public managers are assumed to have strong internal professional motivations to fulfill community demands. These internal incentives drive public managers to develop an active propensity to identify community problems and make decisions based on professional assessments and expertise (King & Stivers, 1998; Selden, Brewer, & Brudney, 1999). Of course, it is also possible that local government managers balance the pressure from external stakeholders with their own propensity for identifying, assessing, and responding to community problems.

Some researchers note that e-government innovations are developed in response to external pressures from businesses, citizens, lobbyists, and interest groups. For example, Ho and Ni (2004, p. 166) argue that “Many innovative practices in government, especially in areas of technology and computer usage, are the result of demand from business leaders and citizens, who see demonstrated successes in the business sector and challenge the government to change and adopt them.” External pressures might also encourage a government to adopt a particular type of technology (Ahn, 2011; Cohen & Sauermann, 2007; Ho & Ni, 2004). For example, stakeholders might demand increased information about government activities, thus necessitating the development of one-way e-communications and the posting of information. Other stakeholders might demand increased engagement and participation via electronic technologies, resulting in increased two-way communications. Since electronic technologies are often developed in response to stakeholder demands, we expect that public managers under higher levels of pressure from external stakeholders will report increased adoption and use of e-government technologies.

Data and Method

This research tests the proposed hypotheses using data from a 2010 national web survey on e-government technology and civic engagement sponsored by the Institute for Policy and Civic Engagement at the University of Illinois at Chicago. The survey was administered as a two-stage cluster sample to managers in 500 randomly selected local governments with populations ranging from 25,000 to 250,000. Because larger cities often have greater financial and technical capacity for e-government, all 184 cities with a population over 100,000 were selected while a proportionate random sample of 316 was drawn from 1,002 cities with populations under 100,000. ¹ The data are weighted to reflect this sampling procedure. ² For each city, lead managers were identified in each of the following five departments, which are present in all 500 cities: mayor’s office (city manager or equivalent—not the mayor), community development (director), finance (director), police (deputy police chief), and parks and recreation (director). A total of 2,500 managers were invited to take part in the survey, which began on August 2, 2010 and closed on October 11, 2010. The final response rate was 37.9%, with 902 responses and under a 2.6% margin of error at 95% confidence interval.

Independent variables

The model includes five independent variables. The variable, personnel constraints is defined as personnel rules that constrain performance, such as rigid promotion, reward, and dismissal (Feeney & Rainey, 2010; Pandey & Scott, 2002). The variable is an average of responses from the following two questionnaire items (Cronbach’s α = 0.652): (1) the formal pay structures and rules make it hard to reward a good employee with higher pay here; and (2) even if a manager is a poor performer, formal rules make it hard to remove him or her from the organization. Response categories are a 5-point Likert-type scale of agreement, ranging from 1 = strongly disagree to 5 = strongly agree.

Work routineness is a scale measure developed by Hage and Aiken (1969). This questionnaire included three of Hage and Aiken’s original four items: (1) people here do the same job in the same way every day, (2) one thing people like around here is the variety of work, and (3) most jobs have something new happening every day. Response categories are a 5-point Likert-type scale of agreement, ranging from 1 = strongly disagree to 5 = strongly agree. Work routineness as the average of responses from the first three items questionnaire items (Cronbach’s α = 0.595). Although the Cronbach’s α is low, 0.595, we retain the measure because these three items are commonly used in the public administration literature to capture work routineness (DeHart-Davis & Pandey, 2005; Pandey & Wright, 2006; Yang & Pandey, 2009).

The variable organizational centralization is an average of the following three items (Cronbach’s α = 0.750; Aiken & Hage, 1971): (1) there can be little action taken here until a supervisor approves a decision; (2) in general, a person who wants to make his own decisions would be quickly discouraged in this agency; and (3) even small matters have to be referred to someone higher up for a final answer. Response categories are a 5-point Likert-type scale of agreement, ranging from 1 = strongly disagree to 5 = strongly agree.

Pressures from external stakeholders are captured using two measures: external influence and citizen demands. The variable, external influence is the average of responses to a set of items that asked respondents to indicate the level of influence that the following actors have on their departments: business groups, advocacy groups, public opinion, and media (Cronbach’s α = 0.818) (Welch & Pandey, 2006). Response categories range from 1 = no influence to 5 = very strong influence. Citizen demands is comprised of responses to the following questionnaire item: In our view, residents in this city want more services online. Response categories range from 1 = strongly disagree to 5 = strongly agree.

We include the following individual-level control variables: male (male = 1; female = 0), college (college = 1 and not college = 0), White (White = 1; non-White = 0), and age and job tenure, continuous variables indicating the respondent’s age and the number of years the respondent has worked in the position, respectively. Organizational-level control variables include organization size, ⁴ the log of the number of full time equivalent employees in the organization, and a series of dummy variables for department type (mayor’s office, community development department, finance department, parks and recreation department, and police department). We also control for the level of technology in the department with three variables. Intranet is a dummy variable indicating if the department has an intranet (yes = 1; no = 0). % Internet for work indicates the percent of employees in the department that use the internet for work. Information technology dept is a dummy variable indicating whether or not the organization has a separate information technology department (yes = 1; no = 0). Mayor council is a dummy variable indicating whether the city’s political structure is mayor–council (=1) or council–manager (=0) form of government. Descriptive statistics for all variables are listed in Table 1.

OPEN IN VIEWER

Table 1.

Descriptive Statistics for Study Variables.

Variable	N	Min	Max	Mean	SD
Dependent variables
Adoption of E-services	727	0	4	2.40	1.340
Adoption of communication technologies	808	0	5	3.09	1.533
Independent variables
Personnel constraints	850	1	5	3.66	0.991
Work routineness	843	1	5	2.33	0.644
Organizational centralization	839	1	5	2.32	0.745
External influence	837	1	4.75	2.37	0.735
Citizen demands	882	1	5	3.92	0.776
Controls
Male	841	0	1	0.77	0.423
Age	803	25	75	50.96	8.519
College	838	0	1	0.94	0.231
White	860	0	1	0.85	0.360
Job tenure	840	0	34	6.69	6.276
Organization size (logged)	801	0	4.26	1.55	0.682
Mayor’s office	902	0	1	0.15	0.361
Community development dept	902	0	1	0.23	0.422
Finance department	902	0	1	0.17	0.377
Parks and recreation dept	902	0	1	0.23	0.421
Police department	902	0	1	0.21	0.409
Intranet	823	0	1	0.82	0.380
% Internet for work	840	0	100	73.28	30.543
Information technology dept	892	0	1	0.57	0.495
Mayor council	902	0	1	0.34	0.475

Results

The research question focuses on the determinants of the adoption of two types of e-government technologies: e-services and communication technologies. Since both dependent variables are count scale, we use Poisson regression models to estimate the goodness-of-fit. The goodness-of-fit for the adoption of e-services model is 1085.801 (Prob > χ²[547] = 0.0000 < 0.0001) and the adoption of communication technologies model is 1080.221 (Prob > χ²[601] = 0.0000 < 0.0001), indicating that the Poisson regression models are not appropriate. Because the dependent variables might be overdispersed, we estimate the models using negative binominal regression.

Taken together, the models indicate support for the first hypothesis that the predictors of the adoption of e-services vary significantly from the predictors of the adoption of communication technologies, supporting Ahn’s (2011) research which argues that the motivations for adopting e-services are distinct from those for adopting communication technologies. This finding points to the importance of separating our discussion of e-services from that of communication technologies in e-government research and practice. Overall, we see that organizational constraints such as personnel constraints and organizational centralization are negatively related to the adoption of e-services but are not significantly related to the adoption of communication technologies. E-services, which aim to increase efficiencies in organizations, are less likely to occur in local government organizations that are centralized and constrained. In comparison, the adoption of communication technologies is positively related to pressure from external groups and citizen demands, which makes sense given the ways in which one- and two-way communication technologies can be used to provide external stakeholders with information and access to local governments.

Adoption of E-services

Table 2 shows the control model (model 1) and full model (model 2) predicting the adoption of e-services. Model 1 predicting the adoption of e-services is statistically significant (χ² = 130.71, p < .0001 with robust standard errors). Model 2 is also statistically significant (χ² = 205.33, p < .0001 with robust standard errors). The comparison between models 1 and 2 indicates statistically significant differences. Both the AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) values in model 2 are significantly lower than model 1, indicating that the full model, which includes the variables for personnel constraints, work routineness, external pressures, and citizen demands, is a significant improvement over the control model.

OPEN IN VIEWER

Table 2.

Adoption of E-Services Models.

	Adoption of E-services
	Model 1			Model 2
	Coefficient	Robust SE	IRR	Coefficient	Robust SE	IRR
Personnel constraints				−0.060	0.022	0.941**
Work routineness				−0.097	0.042	0.907*
Organizational centralization				−0.083	0.036	0.920*
External influence				0.058	0.032	1.059
Citizen demands				0.080	0.029	1.084**
Male	−0.007	0.056	0.993	0.006	0.052	1.006
Age	0.009	0.003	1.009*	0.007	0.003	1.007*
College	0.157	0.132	1.170	0.200	0.132	1.221
White	0.078	0.083	1.081	0.033	0.076	1.034
Job tenure	0.009	0.003	1.009**	0.010	0.003	1.011***
Organization size (logged)	0.129	0.043	1.138***	0.093	0.042	1.098*
Mayor’s office (reference)
Community development dept	−0.176	0.085	0.838*	−0.104	0.083	0.901
Finance dept	0.086	0.083	1.090	0.216	0.081	1.241**
Parks and recreation department	0.099	0.072	1.104	0.109	0.069	1.115
Police dept	−0.239	0.078	0.787**	−0.162	0.078	0.850*
Intranet	0.304	0.082	1.355***	0.273	0.077	1.313***
% Internet for work	0.002	0.001	1.002	0.001	0.001	1.001
Information technology dept	0.011	0.047	1.011	0.026	0.045	1.026
Mayor council	−0.196	0.057	0.822**	−0.188	0.056	0.828***
Constant	−0.376	0.263		−0.038	0.307
N	576			567
Log pseudo-likelihood	−2,369.99			−2,294.04
χ2	130.71			205.33
AIC	4,769.97			4,628.08
BIC	4,835.31			4,714.88

Note. IRR = incidence rate ratios (log of the ratio of expected counts).

*

p < .05. **p < .01. ***p < .001.

The results confirm hypotheses 2, 3, and 4 that personnel constraints, work routineness, and organizational centralization are negatively related to the adoption of e-services. Public managers that perceive higher levels of personnel constraints report fewer e-services; a one unit increase in managers’ perceptions of personnel constraints is related to a 5.9% decrease in the rate of e-services adoption. The IRRs also indicate that the percent change in the incident rate of the adoption of e-services is a decrease of 9.3% for every unit increase in managerial perceptions of work routineness and a decrease of 8% for every unit increase in managerial perceptions of organizational centralization. E-service adoption is higher in departments where public managers perceive fewer rigorous pay structures and difficulties for promotion, lower work routineness, and lower organizational centralization. Thus, local government managers with more flexibility in decision-making report increased adoption of e-services.

The results presented in Table 2 indicate that citizen demands are positively associated with the adoption of e-services, supporting hypothesis 5. Specifically, the percent change in the incident rate of the adoption of e-services is an increase of 8.4% for every unit increase in citizen demands. In addition, public managers who report that residents want more services online report increased rates of e-services adoption.

The control variables indicate that public manager job tenure and organizational size are positively related to the adoption of e-services. We find some differences in the adoption of e-services across city departments. Compared to the mayor’s office, public managers in finance departments report the adoption of more e-services and respondents in police departments report fewer e-services. In addition, mayor–council type governments report fewer e-services as compared to council–manager governments. One explanation for this finding might be the more professional approach taken in council–manager governments, where professionalized managers are working to improve efficiencies in local governments as compared to mayor–council governments that are subject to the leadership and decision-making of elected officials who may or may not have the time to implement e-government initiatives. Finally, public managers who work in departments with an intranet report more e-services in comparison with those in a nonintranet environment, indicating that overall technology capacity in departments is related to the provision of e-services.

Adoption of communication technologies

Table 3 shows the determinants of the adoption of communication technologies. Model 1, with only control variables (χ² = 61.79, p < .0001 with robust standard errors) and model 2, including all variables, are both statistically significant (χ² = 98.43, p < .0001 with robust standard errors). Both AIC and BIC values in model 2 are significantly reduced by including personnel constraints, work routineness, external pressures, and citizen demands variables indicating that the model fit is better in model 2.

OPEN IN VIEWER

Table 3.

Adoption of Communication Technologies Models.

	Adoption of communication technologies
	Model 1			Model 2
	Coefficient	Robust SE	IRR	Coefficient	Robust SE	IRR
Personnel constraints				−0.007	0.020	0.993
Work routineness				−0.086	0.037	0.917*
Organizational centralization				0.037	0.031	1.038
External influence				0.086	0.028	1.089**
Citizen demands				0.058	0.025	1.060*
Male	0.050	0.052	1.051	0.078	0.051	1.081
Age	0.009	0.003	1.009**	0.009	0.003	1.009***
College	0.075	0.103	1.077	0.112	0.101	1.118
White	−0.041	0.066	0.960	−0.063	0.065	0.939
Job tenure	0.000	0.003	1.000	0.002	0.003	1.002
Organization size (logged)	0.073	0.035	1.076*	0.046	0.035	1.047
Mayor’s office (reference)
Community development dept	−0.188	0.072	0.829**	−0.198	0.073	0.820**
Finance dept	0.014	0.072	0.986	0.052	0.074	1.054
Parks and recreation dept	−0.040	0.063	0.961	−0.098	0.064	0.906
Police dept	−0.133	0.064	0.876*	−0.099	0.065	0.906
Intranet	0.065	0.058	1.067	0.049	0.058	1.050
% Internet for work	−0.000	0.001	1.000	−0.001	0.001	0.999
Information technology dept	−0.001	0.040	0.999	−0.010	0.040	0.990
Mayor council	−0.120	0.216	0.887**	−0.124	0.047	0.883**
Constant	0.573	0.216	**	0.328	0.262
N	631			621
Log pseudo-likelihood	−2,832.61			−2,764.45
χ2	61.79			98.43
AIC	5695.22			5568.89
BIC	5761.93			5657.52

Note. IRR = incidence rate ratios (log of the ratio of expected counts).

*

p < .05. **p < .01. ***p < .001.

The full model indicates no support for hypotheses 2 and 4 that predicted relationships between personnel constraints and organizational centralization and the adoption of communication technologies. The full model indicates support for hypothesis 3; work routineness is negatively related to the adoption of communication technologies. Hypothesis 5 is confirmed: public managers perceiving higher levels of external pressure from stakeholder groups including business groups, advocacy groups, public opinion, and media, report increased adoption of communication technologies. Specifically, the percent change in the incident rate of the adoption of communication technologies is an increase of 8.9% for every unit increase in external influence. Moreover, public managers perceiving higher levels of citizen demands for online services report increased adoption of communication technologies, with the percent change in the incident rate of the adoption of communication technologies increasing 6% for every unit increase in citizen demands. Thus, we see that the most important predictors of adopting communication technologies are external stakeholders and citizen demands, and less so organizational characteristics such as personnel constraints, organizational centralization, or organizational size.

In addition, the adoption of communication technologies is only related to department type in the case of community development departments, with managers reporting lower adoption of communication technologies in comparison with the mayor’s office. Similar to the e-services model, communication technologies are more frequently reported in council–manager governments than in mayor–council governments. We expect this finding is related to the differences inherent in local governments operated by professional managers.

Discussion

It is important to note the limitations of this research. First, the data are cross-sectional. Thus, although we find significant relationships between the variables, we are limited in our ability to test causal relationships. Ideally, future research will use longitudinal data to better capture the causal mechanisms driving the adoption of e-government technologies. Second, the items used to operationalize the concepts in these models are limited to those that were included in the survey and the data are subject to the types of errors and biases that occur when asking respondents to report activities and perceptions. Finally, this research focuses on the use of e-services and communication technologies in local governments, but not the quality of those services. We are unable to draw conclusions about how effectively managers use these technologies and for what purposes.

The primary goal of this research is to explore the factors that determine the adoption of e-services and communication technologies in local governments. The results show that the predictors of the adoption of e-services are significantly different from the predictors of the adoption of communication technologies. The predictors of e-services in local governments can be best described as organizational type, capacity, and culture. E-services are adopted at a higher rate in organizations that are centralized, larger, and that have an overall technology capacity level that provides intranet. In addition, e-service adoption is related to the organizational type, with e-services being more frequent in finance and police departments, as compared to mayor’s offices. Thus, the use of e-services in local governments is as best understood as an outcome of organizational type and practices.

In comparison, the determinants of communication technologies are best described as external environment factors. Local government managers that report strong external influences and citizen demands on their organizations report more use of communication technologies. This finding makes sense because communication technologies are often used to disseminate information to the public or provide a means for communicating with and soliciting input from external stakeholders—which might be driven by citizen demands. Thus, we find that local governments that are adopting communication technologies, whether one- or two-way, are doing so in response to demands from citizens and external stakeholders.

These results indicate two focal contributions of this research: first, the importance of distinguishing between the adoption of e-services and communication technologies and second, the importance of understanding the relationships between different types of organizational constraints and external factors on innovation adoption.

First, the results show different estimation patterns between the adoption of e-services and the adoption of communication technologies. The negative relationships between personnel constraints, work routineness, and organizational centralization indicate that the e-services model is closely related to previous innovation studies in the private sector which assume that an organization adopts innovations for the purposes of increasing competence and performance (Nelson, 1990) and local governments adopt e-services when perceiving more advantages in comparison with other supplements (Bozeman & Kingsley, 1998). Local governments that are more flexible (e.g., have fewer personnel constraints and are less centralized), adopt more e-services as compared to more constrained local governments—which might also be an indication of organizations responding to new public management reforms and pressures to do more with less, since e-services, such as online payment, delivery of records, requests to citizens, and online job applications, aim to reduce costs and increase performance.

In comparison, organizational constraints are less important in explaining the adoption of communication technologies. The determinants for adopting more complex e-government technologies such as social networking tools, text messaging, e-mails, and webcasts include pressures from external stakeholders, indicating the important connection between communication technologies and stakeholder demands and calls for civic participation. Thus, while the outcomes or benefits of adopting communication technologies might be ambiguous and difficult to measure, their use is significantly related to connections to the public.

Finally, we argue that the distinction in the determinants for adopting e-services and communication technologies is important for understanding the complex relationships between internal and external constraints on local governments’ innovation adoption. By distinguishing between the types of e-government initiatives, researchers and practitioners can better understand the determinants of adoption and the potential outcomes of each. E-services are driven by organizational capacity and type and therefore, when comparing e-services across local governments it is important to consider the context and organizational environment in which those services are provided. In comparison, when investigating the prevalence and outcomes of communication technologies researchers and practitioners will want to focus on the role of external stakeholders. It is possible that stakeholder influences in the adoption of communication technologies models are related to coercive isomorphism (DiMaggio & Powell, 1983), where stakeholders are pressuring local governments to provide services and communication via web 2.0 applications and local governments are, in some cases, responding to these demands. In conclusion, distinguishing between e-services and communication technologies is important for understanding how the characteristics of innovations matter in local governments—with internal factors being related to e-services and external factors being related to communication technologies. Future research should focus on how these factors, and others, in particular political engagement and citizen participation, shape the outcomes of these two types of e-government initiatives.