An empirical study of e-government adoption in the United Arab Emirates: A social cognitive perspective

Abstract

This study seeks to improve the current understanding of e-government adoption by taking a novel approach. Drawing on the social cognitive theory and its triadic reciprocal model, we developed and tested a new research model in the context of e-government adoption in the United Arab Emirates. The overall findings suggest that e-government adoption is influenced by the reciprocal interactions of personal, behavioural, and environmental factors. The findings imply that when formulating e-government strategy, policy makers need to address in a holistic and integrative way the issue of e-government environment and its alignment with citizens’ Internet use preferences and personal factors. This study opens a new lens for e-government adoption research and offers practical value for governments to develop effective e-government strategy.

Keywords

E-government adoption social cognitive theory triadic reciprocal causation model Internet self-efficacy United Arab Emirates

1. Introduction

Research on e-government (i.e. electronic government) has taken place for decades and generated a plethora of literature which takes a wide range of approaches and explores ‘everything from technology adoption to government-to-government information sharing’ (Bélanger & Carter, 2012). However, the recent literature reviews published on e-government research (e.g. Alcaide-Muñoz et al., 2015; Bélanger & Carter, 2005, 2012) noticed that there is a strong need to develop a more sophisticated theoretical foundation for e-government research and a native e-government adoption theory. Most of the existing theoretical-based e-government adoption research was built largely on diffusion of innovation (DOI) theories (Rogers, 1995; Legris et al., 2003), the various technology acceptance models (Davis, 1989; Venkatesh, 1999), the task-technology fit (TTF) model (Goodhue & Thompson, 1995) and socio-technical theories such as Fountain’s technology enactment framework (e.g. Al-Hujran et al., 2015). Studies based on these theories have advanced e-government adoption research from various perspectives. However, it is noted that the theories tend to take on a unidirectional causation method to explain and predict adoption behavior, which is limited to one-sided determinism. Studies (Doong et al., 2010; Zhao & Khan, 2013) suggest that e-government adoption by citizens involves dynamic interplays between various factors at both micro (individual) and macro (environmental) levels. In this regard, we argue that unidirectional research models are not sufficient to study the complex relationships affecting e-government adoption. For example, on the one hand, environmental factors such as government policies may affect citizens’ adoption of e-government services. On the other, citizens’ participation in e-government could help improve e-government policies and services, which may lead to greater e-government adoption. In this regard, a two-way interaction exists which warrants empirical studies to advance our understanding of the interplays. For that purpose, this study seeks to explore and test a reciprocal interaction model based on the social cognitive theory (SCT) of Bandura (1986). Thus, the primary research questions of this study are whether and how personal factors (e.g. Internet self-efficacy), behavioral factors (e.g. users’ Internet usage pattern) and environmental factors (e.g. accessibility of e-government services) interact and influence e-government adoption. By answering the questions, the findings of the study will shed new insight into the complex relationships affecting e-government adoption. In this regard, the study contributes to the theoretical advancement of current e-government adoption research.

This study also offers practical value. Expenditure in e-government projects has been increasing significantly across the globe over the past decade or so, according to United Nations’ survey (United Nations, 2014). However, the failure rate is also high in both developed and developing countries (Gunawong & Gao, 2017). This study aims to help policy makers deal with the low uptake of e-government through raising their awareness of the interplays between personal, behavioral and environmental factors and offering valuable insights for developing effective e-government strategies to optimize e-government resources.

The next section of this article focuses on a critical review of the social cognitive theory (SCT) and comparing it with technology acceptance model (TAM) to justify the advantage of adopting the SCT for this study. This is followed by hypothesis development drawing on the SCT to illustrate how personal, environmental and behavioral factors interact with each other and influence citizens’ take-up of e-government services. The paper then discusses the research methodology used in the data collection and analysis, and presents and discusses the research findings. The final sections discuss the implications of our findings for advancing e-government adoption theory and for developing e-government strategy, and consider the study’s limitations and areas for future research.

2. Social cognitive theory and technology adoption model

Literature review shows that many theoretical models, such as technology acceptance models (Davis, 1989; Venkatesh, 1999), the task-technology fit (TTF) model (Goodhue & Thompson, 1995) and socio-technical theories (e.g. Fountain, 2001), have been adopted to study e-government adoption (Bélanger & Carter, 2012). Given the limited space of this paper, we discuss the most popular technology adoption model (TAM) only and the reasons that we chose the SCT for this study.

The TAM has been widely adopted in e-government research to explain e-government use by identifying citizens’ adoption behavior. For example, Carter and Bélanger (2005) are among the first few pioneering e-government researchers to apply the TAM to the study of citizens’ behavioral intentions toward e-government services. There are two main constructs of the TAM that influence the adoption of e-government: (1) perceived usefulness (PU), and (2) perceived ease of use (PEU), both of which have been tested by many authors in different geographical settings (e.g. Sang et al., 2009; Alomari et al., 2012; Al-Hujran et al., 2015). However, research also shows that the TAM model needs to be integrated into a broader context and must include factors related to both human and social change processes (Straub, 2009; Bandura, 2012).

Figure 1 illustrates that TAM studies the relationship between technology adoption and factors that influence users’ attitude and behavioral intention in a unidirectional causation mode but does not consider and explore two-way interactive relations between them.

Figure 1.

The technology acceptance model (Compeau & Higgins, 1995; Straub, 2009).

For e-government adoption research, the relationships are far more complicated and dynamic (e.g. Alcaide-Muñoz et al., 2015; Bélanger & Carter, 2005, 2012). We argue that although facilitating conditions (environmental factors) may affect e-government adoption, which is a one-way causation, the environmental factors may interact with personal factors such as Internet self-efficacy and influence e-government adoption. Behavior factors (i.e. Internet usage patterns or preferences) may change, or be re-shaped by, environmental and/or personal factors and influence e-government adoption. These relationships are no longer one-way causation and become two-way interactions. For example, according to the recent United Nations’ e-government survey (United Nations, 2014), with the easy access to social media, more countries are moving towards participatory decision making which improves e-government adoption. However, research also shows that the use of social media is affected by personal factors such as gender, age, education, social-economic status, ethnic background and geographic locations (e.g. rural or urban) (Perrin, 2015). In this regard, e-government adoption could be the result of the interaction between easy access to social media (i.e. an environmental factor) and personal factors such as age, gender and education. To better understand the dynamic relationships, we decided to consider social cognitive theory (SCT), arguing that e-government adoption could be affected by reciprocal interactions among various factors.

Social cognitive theory (SCT) has been widely cited in the literature of psychology, education, and, recently, the adoption of information technology (e.g. Deci & Ryan, 2010; Zhou, 2011; Wyer, 2014). It provides the theoretical framework for predicting human behavior, particularly in the context of social change and learning (Bélanger & Carter, 2012). According to SCT, human behavior is determined by the interaction between personal factors and environmental influences. The interaction effects among environment and behavior produce different situations that influence individual behavior. Individual behavior, in turn, influences the environment. Thus, the distinctive nature of SCT lies in its emphasis on the reciprocal nature of the determinants of human functioning. Unlike other theories that depict human behavior according to unidirectional causation, SCT is best manifested by a triadic reciprocal causation model. ‘In this model of triadic reciprocal causation, behavior, cognition and other personal factors, and environmental influences all operate as interacting determinants that influence each other bi-directionally’ (Bandura, 1989). It is also noted that some reciprocal influences could be stronger than others when interacting and that they do not necessarily occur simultaneously (Bandura, 1989). At the core of SCT is perceived self-efficacy: “The belief in one’s capability to organize and execute the courses of action required to manage prospective situations” (Bandura, 1989). Self-efficacy (a shortened term for perceived self-efficacy in this article) is viewed as a focal determinant because it affects behavior or action directly, and through its influence it also impacts on goals, outcome expectations, and perceived impediments (Bandura, 2012).

After searching extensively the current e-government literature published in indexed academic journals in English, we found that very few studies have used SCT in e-government research. Among the few, for example, Liang and Lu (2013) used SCT together with diffusion of innovations and contingency theory to study the adoption of an online tax filing system in Taiwan. This study chose self-efficacy as an individual (personal) factor and social norms as the environmental factor and found that social norm had a significant effect on people’s intention to use the online tax filing system. Although the study examined the interaction between personal and environmental factors, it did not consider whether adoption behavior could in turn influence environmental factors. Our study addresses the limitation and tests the triadic reciprocal relationships.

3. Hypothesis development

The conceptual model of triadic reciprocal causation posits that personal, behavioral (i.e. behavioral patterns), and environmental factors influence and interact with each other (Bandura, 1989). To conceptualize this in the context of e-government adoption, we develop three hypotheses to illustrate broadly how the three factors interact with each other and influence e-government adoption. Figure 2 presents our research model and the hypothetical relationships among the three factors.

Figure 2.

A triadic reciprocal model to study e-government adoption.

Table 1

Variables and measures

Variables	Measures
Personal factors (IV)	Internet self-efficacy
	Gender
	Age
	Education
	Occupation
	Income
Behavioral factors (IV)	Enjoy searching for information online
	Prefer using the Internet to find information related to work/job/study
	Enjoy using social networking tools
	Feel uncomfortable making payments online
	Feel uncomfortable providing personal information online
Environmental factors (IV)	Raising public awareness of online government services
(i.e. facilitating factors)	Keeping people informed of what governments are doing
	Improving accessibility of government web sites and online services
	Making government web sites and online services more user-friendly
	Making government web sites and online services more secure
	Using social media to offer online services to the public
E-government adoption (DV)	Looking for information about a policy/regulation
	Investigating what services governments are offering
	Paying a fine
	Making a payment other than a fine
	Applying for and/or renewing a license
	Applying for government job(s)
	Downloading government forms
	Researching government reports, statistics, or other official documents
	Getting tourist information
	Getting advice from a government department about a health or safety issue
	Getting information about public education
	Applying for government benefits

3.1 Personal factors interacting with behavioral factors

3.1.1 Personal factors and measures

Self-efficacy is a key component of SCT and at the center of the personal segment of the triadic reciprocal causation model (Bandura, 1989). Some of empirical studies found that self-efficacy predicts e-government adoption (e.g. Hung et al., 2013). In this study we chose specifically Internet self-efficacy (ISE) as the key variable of personal factors because it is essential to the use of e-government services. ISE refers to the self-perception and self-competency of users when they interact with the Internet (Torkzadeh et al., 2006), whereas computer self-efficacy (CSE) is defined as an individual judgment of one’s capability to use a computer (Compeau & Higgins, 1995; Marakas et al., 2007; Marakas et al., 1998). Although CSE has been used to refer to ISE in many research articles, the two constructs are not the same (Torkzadeh et al., 2006). E-government empowered by the Internet and digital technologies is significantly different from the back-office automation (computer-based automation) in government agencies that started a few decades ago. With the advent of Web 2.0-based technologies, such as social media, e-government in many countries has become ‘Government 2.0’. Web 2.0, as a personalized and communicative form of the Internet, enables active participation, connectivity, and collaboration (Wigand et al., 2010; Porumbescu, 2016). In this regard, Web 2.0 provides a better opportunity for e-government participation (i.e., e-participation) than the first generation of Web technologies. Therefore, measuring ISE becomes more relevant than CSE in terms of e-government use. In addition to ISE, we consider that other personal factors (see Table 1 for detail), including gender, age, education, and income, could have an effect on e-government adoption through interaction with behavioral factors. For example, the studies of Al-Shafi and Weerakkody (2010) and Morgeson et al. (2011) found that the e-government adoption is higher among males than females and that the majority of e-government adopters are aged between 25 and 44 and have university education backgrounds.

3.1.2 Behavioral factors (patterns) and measures

In terms of behavioral factors, we look specifically at Internet usage patterns and argue that such patterns could influence e-government adoption when interacting with personal factors (e.g. Internet self-efficacy, age and gender). To gauge the Internet usage patterns, we examine Internet users’ preferences in terms of the information they are looking for (e.g., searching for information related to work/jobs and/or study), use of social media and tools, use of online payment systems, and providing personal information online. Prior research has attempted to test the associations between Internet use and e-government adoption. For example, the empirical study of Morgeson et al. (2011) found that Internet use is significantly related to e-government use, although e-government adoption is a far more complex phenomenon than the use of the Internet. For example, trust of government and citizens’ perception of government e-services are seen as significant determinants to e-government adoption (Carter et al., 2016). The recent study of Venkatesh et al. (2014) demonstrates that personality and demographics affect people’s preferences for Internet use as well as e-government use. However, the extant studies are mostly interested in finding the linear relationships between various factors and e-government adoption. Little research has been conducted to look at the interaction effect of Internet usage pattern on e-government adoption. In other words, the effect of Internet usage patterns on e-government adoption could be affected by personal factors. We seek to address the knowledge gap.

To identify the overall interaction among personal and behavioral factors and its consequent influence on e-government adoption, we developed hypothesis 1 (H1).

H1: Personal factors and behavioral factors interact with each other, which in turn influences e-government adoption.

3.2 Environmental factors interacting with personal factors

3.2.1 Environmental factors and measures

In the context of e-government, we consider that environmental factors are those that are closely related to, and conducive to, e-government adoption. Extant e-government literature has identified the link between environment and e-government adoption (e.g., Lean et al., 2009; Alcaide-Muñoz et al., 2015). For example, the empirical study of Sharma and Mishra (2017) found that factors such as citizens’ awareness, accessibility to e-government services, ease of obtaining the services and the reliability of e-services (i.e. e-service quality) offered by governments are all significant determinants to e-government adoption. With the increasing popularity of social media use, studies have demonstrated the link between the availability of social network sites and e-government participation and adoption (e.g. Gao and Lee, 2017; Bonsón et al., 2017). Therefore, we investigated environmental factors such as public awareness about e-government services; accessibility; the quality, security, and design of government web sites; and the availability of Government 2.0 (see Table 1 for detail). Although abundant literature has studied the effect of environmental factors on e-government adoption, the prior studies have adopted mostly the TAM model, focusing on one-way causality between the two. Our study takes a different approach using the SCT to investigate the possible interactions between e-government environment and personal factors and exploring whether the interactions influence e-government adoption. Therefore, we developed hypothesis 2 (H2).

H2: Environmental factors and personal factors interact with each other, which in turn influences e-government adoption.

3.3 Behavioral factors interacting with environmental factors

According to SCT, the relationship between behavioral and environmental factors is a reciprocal one; in other words, influence is two-way. Thus, behavior can alter environmental conditions and is in turn altered by the environmental conditions it creates. It should be noted that in most cases, environment does not influence ‘until it is activated by appropriate behavior’ (Bandura, 1989). To test the concept in the context of e-government adoption, we need to find out whether behavioral patterns (i.e. Internet usage patterns) influence, and be influenced by, environmental factors, which may consequently affect e-government adoption. Some empirical research demonstrates the significant role of environment in e-government adoption. For examples, using global datasets, Zhao et al. (2013) found that e-government adoption is by and large influenced by environmental factors. Using the construct of reciprocal causation theory and the empirical findings discussed here, we developed our third hypothesis (H3).

H3: Behavioral factors (i.e. Internet usage patterns) and environmental factors interact with each other, which in turn influences e-government adoption.

4. Methodology

To test the hypotheses and validate our research model, we conducted a survey in the United Arab Emirates (UAE). The selection of the UAE for this study has mainly three reasons. First, the UAE has one of the most developed economies in the Middle East region and is classified as a high-income developing economy by the IMF (2010). Benchmarking conducted by the UN between 2003 and 2014, for the most part, placed the UAE in the top 50 of more than 190 states in terms of e-government development. The UAE was also ranked highest in e-government development and in e-participation in the Middle East region in the latest surveys (United Nations, 2014, 2016). The results suggest that e-government in the UAE is well developed and mature. In this regard, the findings of this study could have wider implications for the region. Second, there is limited rigorous empirical research that has been conducted about e-government adoption in the UAE. This study helps address the gap. Third, the lead author had worked and lived in the country for years before and when the data was collected. This helped the author get access to the sample population and conduct the survey as well as develop a sound understanding of the UAE context.

4.1 Survey

We conducted a quantitative survey in the UAE to examine whether and how personal, behavioral, and environmental factors interact with each other and influence citizens’ adoption of e-government services. The survey instrument, a questionnaire, was designed on the basis of our research model (see Fig. 2) and the hypotheses and measures that we developed and discussed in detail in the preceding section. The questionnaire contained three parts. Part 1 collected demographic information about participants. Part 2 attempted to examine the Internet usage patterns of participants. Part 3 sought to explore participants’ e-government use. It also sought to investigate the e-government environment by gauging participants’ views on what government can do to foster an environment that encourages e-government use. Except for Part 1, the questions were rated on a five-point Likert scale. Appendix 1 presents sample questions used for the survey. Table 1 provides more detail about the questionnaire and also shows how we operationalized the key factors – personal, behavioral and environmental as well as how we measured e-government adoption.

Before we launched the survey, we tested the questionnaire in a small group to assess and improve its clarity, logical consistency and contextual relevance. A group of colleagues (four) and students (five) at a university in the UAE participated in the test and provided constructive feedback. Consequently, some of the questions were re-worded, revised, and refined before the survey was formally launched between November and December 2014.

4.2 Sample

The population of UAE was around 9 million in 2016 according to the World Bank (2016). Constrained by the resources that we had, we were not able to conduct a nation-wide survey. Instead, we selected our sample from one university in the UAE with a student population of around 5,000 who came from various parts of the UAE. Our targeted population was Internet users who could need e-government services for personal reasons. One thousand potential participants who met the basic criteria; namely, Internet users and UAE citizens and residents (not visitors) were randomly selected using their email addresses from a list of nearly 5000 people who were a mix of students undertaking foundation studies, undergraduate and postgraduate courses, and alumni in a university in the UAE. The potential participants were invited by email and provided with a web link to our online survey. A total of 361 people completed the survey, resulting in a response rate of 36.1 percent. Research shows that the response rate for online survey is generally lower than that of paper-based survey which is administered in a face-to-face setting (Nulty, 2008). After comparing nine studies on online response rate, it is found the overall response rate is about 33 percent (Sax et al., 2003). Based on the prior studies, we considered that our response rate was acceptable. After checking the completed questionnaires, we found that 355 of them were usable. A summary of the demographic data of the respondents is presented in Table 2. Given that personal factors in our study come from demographic data, we will discuss further the sample features in our Data Analysis and Results section.

Table 2
Summary of demographic data of survey participants ( $n=$ 355)

Variable	Frequency	Frequency	Frequency	Frequency
Gender	Male: 56%	Female: 44%
Age	20–25: 52%	26–30: 27%	31–40: 6%	No answer: 15%
Education level	Bachelor: 52%	Master: 24%	Foundation study: 21%	No answer: 3%
Occupation	Professionals: 42%	Full-time students:	Non-professional workers:	Others: 16%
		31%	11%
Income	Middle income: 38%	Low income: 34%	High income: 9%	Others (non-claimed): 19%

4.3 Statistical procedures

In this study, several statistical analysis methods were taken to test the hypotheses. We used a two-step procedure. In step one, given the large number of variables and measures (see Table 1), we used principal component analysis to identify the more important variables related to environment, behavior factors, and e-government adoption. Principal component analysis (PCA) is a well-established mathematical technique and has the advantage of reducing the dimensionality of data while keeping as much variation as possible (Abdi & Williams, 2010). We used PCA to reduce environmental variables, behavioral variables and e-government adoption patterns into fewer variables. We did not reduce personal variables because we used unique scales to measure different personal variables due to the nature of the variables. For example, personal variables, such as, gender, age, education, occupation and income are different types of variables (i.e. categorical or numerical) and was measured using different scales; however, behavioral and environmental variables and e-government adoption variables were the same

Table 3
Results of principal component analysis for behavioral factors, environmental factors, and e-government adoption

Behavioral variables	Behavioral		Environmental variables	Environmental		E-government adoption	E-government
	factors			factors		variables	adoption
	Factor 1	Factor 2		Factor 1	Factor 2		Factor 1	Factor 2
Enjoy searching for information online	0.737		Raising public awareness of online government services	0.802		Looking for information about a policy/regulation		0.635
Prefer using the Internet to find information related to work/job/study	0.801		Keeping people informed of what governments are doing	0.638		Investigating what services governments are offering		0.682
Enjoy using social networking tools	0.696		Improving accessibility of government web sites and online services	0.635		Paying a fine	0.428
Feel uncomfortable making payments online		0.885	Making government web sites and online services more user-friendly	0.784		Making a payment other than a fine	0.578
Feel uncomfortable providing personal information online		0.879	Making government web sites and online services more secure		0.914	Applying for and/or renewing a license	0.854
			Using social media to offer online services to the public		0.928	Applying for government job(s)	0.633
						Downloading government forms		0.682
						Researching government reports, statistics, or other official documents		0.835
						Getting tourist information	0.703
						Getting advice from a government department about a health or safety issue	0.759
						Getting information about public education	0.690
						Applying for government benefits	0.777
Variance explained percentage	33.662	33.501	Variance explained percentage	34.954	29.095	Variance explained percentage	38.149	37.293
Eigenvalues ${}^{*}$	2.031	1.328	Eigenvalues	2.427	1.416	Eigenvalues	4.578	1.153

Note: We included the eigenvalues here to report the reliability of our factor analysis.

type of variables and were thus measured using five point Likert scale. For this purpose, the eigenvalue-one criterion and a scree plot test were used to determine which variables to retain.

These retained variables were used for further analysis in step two when we generated one variable by multiplying Factor 1 and Factor 2 from the e-government adoption variables (see Table 3 for details). This variable was used as the regressed variable (dependent variable) in our regression model. We used ordinary least squares with autoregressive errors for this purpose. There are two reasons for such an approach. First, our model is linear with data gathered from a random population sample. Second, we intended to remedy errors arising from independent variables that were strongly collinear (Godfrey, 2011). The interaction effects for the personal, behavioral and environmental factors were calculated by multiplying all the variables together.

To identify the triadic and reciprocal interactions among personal, behavioral, and environmental variables and their impact on e-government adoption, we used multivariate analysis to identify the effect of the overall interactions of multiple variables. Statistically, multivariate interaction has three advantages over bivariate interaction. First, multivariate analysis is highly sensitive to a number of measures that interact with each other and influence e-government adoption, while bivariate analysis is not (Ductor et al., 2011; Diamond & Sekhon, 2013). Second, multivariate interaction has a stronger statistical power than bivariate interaction. Third, multivariate interaction models provide greater insight among moderators, regressors, and regressed variables. This degree of insight is absent from bivariate interaction (Mencía & Sentana, 2012). Overall, multivariate analysis is more robust than bivariate analysis.

Furthermore, after investing the broad interaction effects between the behavioral, personal and environmental factors, we also investigated their partial or full interaction effects, using path analysis.

5. Data analysis and results

As discussed in the preceding section of this article, SCT research considers gender, age, education, and income as personal factors. We adopted this view and hypothesized that these demographics could interact with other factors (i.e. behavioral and environmental factors) and have an effect on e-government adoption. Table 2 indicates that most of the participants received tertiary and above education. This is consistent with the recent survey results of the UAE’s Human Development Index (HDI). The UAE’s HDI value for 2012 is 0.818, a figure that falls into the ‘very high’ human development category (United Nations Development Program, 2013). The majority of the participants are between 20 and 30 years old, which is consistent with the young age structure of the UAE (CIA, 2013). The gender ratio between the male and female participants is 1.15:1, which is representative of the gender ratio in the UAE (CIA, 2013). In terms of the participants’ occupations, 42 percent are professionals, 31 percent are full-time students, 11 percent are blue-collar workers, and 16 percent are others. As far as income is concerned, we classify participants’ monthly income into three levels: high-income (i.e., above US$10,000), middle-income (i.e., between US$4,000 and US$10,000), and low-income (i.e., below US$4,000). This classification is based on the results of a UAE salary survey (Salary Explorer, 2014). Of the participants, most (72 percent) are from middle and low-income groups. This is not surprising given the young age of the majority of participants. Overall, the participants in the survey are representative of the UAE Internet population who may need to use e-government services. The demographics are further analyzed in the following sections with regards to their relationships with behavioral and environmental factors.

Table 3 presents the results of our first step analysis – principal component analysis performed on behavioral factors, environmental factors, and e-government adoption by using a varimax rotation procedure. Two factors that explain the 67.163% variance in behavioral factors are citizens’ online information-seeking behavior and their level of comfort with online payment systems. Two factors that explain the 64.049% variance in environmental factors are the government’s role in e-government development and the level of e-government service security. Two factors that explain the 56.232% variance in e-government adoption patterns are the use of e-government for seeking government information and the use of e-government for making payments. All the factors that are retained have eigenvalues greater than one in accordance with the eigenvalues-greater-than-one rule proposed by Kaiser (1960).

Table 4 presents the results from our second step of analysis – regression results of our hypotheses H1–H3, suggesting that the impact of the interactions between personal and behavioral factors, and personal and environmental factors on e-government adoption (H1 & H2) is significant, while the impact of the interaction between behavioral and environmental factors on e-government adoption (H3) is not significant. Further, the regression results indicate that the interaction of personal and behavioral factors ( $t=$ 4.810; $\beta=$ 0.382; $p=$ 0.000) has greater influence on e-government adoption as compared to the interaction of personal and environmental factors ( $t=-$ 2.526; $\beta=-$ 0.225; $p=$ 0.012). These results demonstrate that the impacts of these factors on e-government adoption vary when they interact with each other.

Table 4
Regression results of ordinary least squares with autoregressive errors for interactions among personal, behavioral, and environmental factors for hypotheses: H1–H3

	Unstandardized		Standardized	$t$	Sig.
	coefficients		coefficients
	B	Std. error
(Constant)	0.062	0.054	1.143	1.143	0.254
Personal factors x behavioral factors (PF-BF)	0.382	0.079	4.810 ${}^{**}$	4.810	0.000
Personal factors x environmental factors (PF-EF)	0.225	0.089	2.526 ${}^{*}$	2.526	0.012
Behavioral factors x environmental factors (BF-EF)	0.019	0.024	0.799	0.799	0.425

a. Dependent Variable: Zscore: E-government Adoption Factor 1. No significant results were found when regression was conducted for factor 2, so it is not reported. $R=$ 0.295. $F=$ 8.860. ${}^{*}p<$ 0.05, ${}^{**}p<$ 0.01.

In addition to investigating the overall interaction effects of personal, behavioral and environmental factors, we used path analysis to look further into the interaction effects of each of the variables of the three sets of variables (personal, behavioral and environmental) and constructed a path analysis diagram. Figure 3 shows the path analysis diagram indicating the significant interaction effects between personal and behavioral factors, and between personal and environmental factors. It also shows that the interaction between behavioral and environmental factors is partially significant. The details of the full and partial interaction effects are captured in the results of our multiple regression analysis of the three sets of variables presented in Appendix 2. In terms of the partial interaction effects between behavioral and environmental factors, Appendix 2A shows that looking for work information, and uncomfortable in providing personal information online, interact with environmental factors and influences E-government Adoption Factor 1. Similarly, uncomfortable in making payments online, and enjoying searching online, interact with environmental factors and influence E-government Adoption Factor 2 (see Appendix 2B).

Figure 3.

Path analysis diagram of the interaction between personal, behavioral and environmental factors and their impacts on e-government adoption ${}^{*}$ . ${}^{*}$ Note: The solid lines between personal and behavioral factors and between personal and environmental factors in this Figure indicate statistically significant relationships while the dash line between behavioral and environmental factors indicates partially significant relationship.

6. Discussion and implications

This study takes a new approach to investigating citizens’ e-government adoption, drawing on the triadic reciprocal model embedded in SCT (see Fig. 2). The overall findings suggest that e-government adoption is influenced by the reciprocal interactions of personal, behavioral, and environmental factors. In other words, the triadic interactive relations among the three are likely to determine citizens’ e-government adoption. The findings support the call for a more holistic and integrative approach to e-government adoption research (Zhao et al., 2014a). The recent empirical study of Zhao et al. using Fountain’s technology enactment framework (TEF) (Fountain, 2001) suggests that the process of e-government diffusion involves multiple relationships and interactions among various contextual factors such as socio-economic, cultural, technological and legal factors, and user behavior (Zhao et al., 2014b). The findings also provide, from a different theoretical perspective, empirical evidence that the relationships between the environmental factors such as government organizations and technology to enable e-government, and citizens’ e-government adoption can be reciprocal, For example, citizens e-participation in decision-making helps improve the effectiveness of e-services, which in turn helps increase e-government adoption (Zhao et al., 2018).

It is interesting to find that the level of influence of the interaction between personal and environmental factors on e-government adoption is greater than that of interactions between personal and behavioral, and behavioral and environmental factors. For example, ISE (as a personal factor) is found to have a greater effect on e-government adoption when the e-government environment is favorable than the effect generated when ISE interacts with behavioral factors (e.g., Internet usage patterns), and when behavioral factors interact with environmental factors. This finding provides empirical evidence to support Bandura’s theory that the reciprocal interactions among the three factors do not mean ‘the different sources of influence are of equal strength’ (Bandura, 1989).

6.1 Theoretical contribution and implications

To the best of our knowledge, this study is the first of its kind that tests empirically the triadic reciprocal model of SCT in e-government adoption research, although previous research such as Rana and Dwivedi (2015) applied some of the SCT constructs to their e-government adoption research. Thus, it enriches e-government adoption literature. Our research model based on the triadic reciprocal theory provides a theoretical alternative to various technology acceptance models (TAMs). As TAMs do not address social and environmental factors, they may limit their predictive ability for e-government adoption. To explain and predict e-government adoption and also to understand the complex interactions and relationships between key factors, this study takes a new perspective, focusing on an investigation into the effects of the reciprocal interactions of personal, behavioral and environmental factors. The interactions of personal, behavioral, and environmental factors are found to influence e-government adoption in our empirical testing.

Our approach and research model opens up a novel lens through which to view and study the intricate interplays and reciprocal relations among the three sources of influence on e-government adoption. As more and more countries are making significant progress in e-government, more countries have reached to the mature stage of e-government development at which e-participation (i.e. engaging citizens in government decision making) and e-democracy (e.g. online voting and on-line consultation) are the key indicators, as found in 2016 UN e-government survey (United Nations, 2016). Our research approach is particularly useful to explore reciprocal interactions between users (by studying their personal and behavioral factors), environment and e-government adoption. Moreover, enabled by increasingly pervasive social media, more governments are moving towards participatory decision-making (United Nations, 2016). In this regard, studying the interdependence between the key factors is as important as studying one-way causality between behavioral factors and e-government adoption. Our study paves the way for further research in this endeavor.

6.2 Strategic implications

In addition to the theoretical contribution made by this study, the results have implications for the development of e-government strategy in the UAE. Policy-makers in other countries, particularly in the Middle East region, may also find the study useful in improving their understanding of citizen’s e-government adoption, given the homogeneity of the countries in the region in terms of culture, social norms and political systems. Overall, the results suggest that personal, behavioral, and environmental factors are interrelated and influence one another when it comes to e-government adoption. Therefore, when formulating e-government strategy, policy makers need to address in a holistic and integrative way the issue of e-government environment (e.g., accessibility of e-government services and contents) and its alignment with citizens’ Internet use preferences (e.g., making online payments) and personal factors (e.g., citizens’ Internet efficacy). Simply improving e-government environment alone is not sufficient to improve e-government adoption. For example, the provision of electronic platforms for the public to participate in online consultation and policy making of governments is one way to achieve e-democracy. However, the e-democracy may not happen automatically unless the platform (i.e. an environmental factor) fits with the personal and behavioral factors of users, as suggested in our findings. Moreover, the reciprocal relationships (interactions) found in this study suggest that the adoption of e-government is determined by the interaction of two factors, for example, e-government environment on the one hand, and citizens’ usage patterns on the other. In other words, citizens have a role to play in shaping this environment. It is a two-way process and interaction. Thus, for e-government providers, this implies that citizens’ involvement in e-government (i.e., e-participation) is essential to improve e-government adoption. For example, our result (see Appendix 2A) shows that user friendliness of e-government web sites (environmental factor) interacts with personal factors and the interaction influences e-government adoption. This finding suggests that citizens and e-government providers should work together in the design process, which could lead to e-government adoption. Likewise, the accessibility of e-government services (environmental factor) and personal factors interact with each other and influence e-government adoption. This implies that both governments and citizens play important roles in improving the accessibility of e-government services to improve e-government adoption. For example, governments’ consultation with citizens about accessibility is as important as citizens’ active participation in the consultation process.

7. Limitations and future research

As with any piece of research, this study has limitations. First, the study focuses on investigating the effects of bi-directional interactions of personal, behavioral, and environmental factors on e-government adoption. Our overall findings validate empirically and broadly these reciprocal relationships captured in our e-government adoption model. However, given the scope of the study, we do not explore in detail why some of the personal factors (e.g. Internet self-efficacy and age) are more likely to interact and influence e-government adoption than others, and some of the environmental factors (e.g. public awareness of e-government service, and user friendliness of e-government web sites) are more likely to play a significant role in influencing e-government adoption than other environmental factors when interacting with a specific personal factor (see Appendix 2). Further research is needed to expand this study to tackle these research questions. Second, the causality among personal, behavioral, and environmental factors can only be inferred from SCT but not empirically tested because testing causality requires time-series data. The data that we collected are cross-sectional but not longitudinal because of our resource constraints. We acknowledge this limitation and call for longitudinal studies to be conducted in future to test the causality. However, this study has achieved our primary objective of identifying the bidirectional relationships and interactions among the factors that we studied and their effects on e-government adoption. Third, the empirical study was conducted in the UAE only. Research shows that a nation’s culture influences the adoption of technologies as well as e-government adoption (e.g. Im et al., 2011; Zhao & Khan, 2013). Our one-country sample does not allow us to study cultural differences and their effect on e-government adoption. Fourth, given that our survey sample was associated with one university in the UAE, the majority of the participants (76 percent) in our survey received tertiary education or above. Although this demographic data was consistent with the high UAE’s Human Development Index (United Nations Development Program, 2013), caution should be taken when making any generalizations based on this study. Therefore, to further validate our e-government adoption model and to improve its external validity, we suggest that cross-country studies with larger and more representative samples are conducted. These studies could operationalize and examine comprehensively the key factors that could interact and influence e-government adoption.

8. Conclusion

E-government projects form a critical part of government operations and investment. However, e-government adoption by citizens is generally low in many countries, according to United Nations’ survey (United Nations, 2016). This study seeks to improve our understanding of e-government adoption. We draw on SCT to identify the reciprocal interactions among personal, behavioral, and environmental factors and their effects on e-government adoption. Although SCT is one of the most important contemporary theories to study and understand human behavior, few research efforts have been made to test its triadic reciprocal model in the context of e-government adoption. The findings of the study may lead to a new and advanced way to understand e-government adoption behavior.

Footnotes

Appendix 1. Sample questions used for the survey for this study *

Internet self-efficacy

I am good at using the Internet.

Internet usage pattern

I enjoy using social networking tools (e.g. Facebook, Twitter, blogs, etc.).

E-government environment

Governments make their websites and online services more accessible to the public.

E-government adoption

I use e-government to look for information about a policy/regulation

${}^{*}$ Note: Please see Table 1 for more detail on the survey items.

Appendix 2

Appendix 2A: Regression results for dependent variable e-government adoption factor 1
	Unstandardized		Standardized	$t$	Sig.
	coefficients		coefficients
	B	Std. error
Constant	0.077	0.048		1.615	0.107
Internet self-efficacy x Internet usage patterns	0.206	0.098	0.340 ${}^{*}$	2.113	0.036
Gender x Internet usage patterns	0.065	0.078	0.065	0.834	0.405
Education x Internet usage patterns	0.802	0.399	0.415 ${}^{*}$	2.011	0.045
Age x Internet usage patterns	0.037	0.143	0.058	0.258	0.797
Income x Internet usage patterns	0.084	0.151	0.074	0.555	0.580
Public awareness of e-government x personal factors	0.184	0.115	0.257	1.603	0.110
Keeping public informed of e-government offerings x	1.217	0.387	0.474	3.144	0.002
personal factors
Accessibility of e-government x personal factors	1.305	0.476	0.380 ${}^{*}$	2.742	0.007
User friendliness of web sites x personal factors	0.574	0.203	0.561 ${}^{*}$	2.835	0.005
Security of web sites x personal factors	0.020	0.399	0.018	0.050	0.960
Availability of Government 2.0 x personal factors	0.311	0.191	0.139	1.626	0.105
Enjoy searching online x environmental factors	0.025	0.014	0.144	1.760	0.080
Look for work information x environmental factors	0.560	0.090	0.410 ${}^{**}$	6.223	0.000
Enjoy using SNs x environmental factors	0.013	0.099	0.009	0.127	0.899
Uncomfortable paying online x environmental factors	0.019	0.080	0.018	0.231	0.818
Uncomfortable providing personal information online x	0.275	0.039	0.427 ${}^{**}$	7.026	0.000
environmental factors

a. Dependent Variable: Zscore: REGR factor score E-government Adoption Factor 1. $R^{2}=$ 0.351. $\Delta R^{2}=$ 0.312. $F=$ 8.957. ${}^{*}p<$ 0.05, ${}^{**}p<$ 0.01.

Appendix 2B: Regression results for dependent variable e-government adoption factor 2
	Unstandardized		Standardized	$t$	Sig.
	coefficients		coefficients
	B	Std. error
Constant	0.014	0.056		0.256	0.798
Internet self-efficacy x Internet usage patterns	0.080	0.115	0.128	0.699	0.485
Gender x Internet usage patterns	0.138	0.092	0.133	1.503	0.134
Education x Internet usage patterns	0.526	0.469	0.264	1.122	0.263
Age x Internet usage patterns	0.388	0.168	0.594 ${}^{*}$	2.315	0.021
Income x Internet usage patterns	0.188	0.178	0.160	1.058	0.291
Public awareness of e-government x personal factors	0.161	0.135	0.218	1.192	0.234
Keeping public informed of e-government offerings x	0.845	0.455	0.319	1.857	0.064
personal factors
Accessibility of e-government x personal factors	0.216	0.559	0.061	0.386	0.700
User friendliness of web sites x personal factors	0.309	0.238	0.292	1.297	0.196
Security of web sites x personal factors	0.078	0.469	0.068	0.166	0.868
Availability of Government 2.0 x personal factors	0.194	0.225	0.084	0.863	0.389
Enjoy searching online x environmental factors	0.042	0.017	0.231 ${}^{*}$	2.483	0.014
Look for work information x environmental factors	0.132	0.106	0.094	1.252	0.212
Enjoy using SNs x environmental factors	0.030	0.116	0.021	0.261	0.794
Uncomfortable paying online x environmental factors	0.381	0.095	0.362 ${}^{**}$	4.032	0.000
Uncomfortable providing personal information online x	0.075	0.046	0.113	1.628	0.105
environmental factors

a. Dependent Variable: Zscore: REGR factor score E-government Adoption Factor 2. $R^{2}=$ 0.158. $\Delta R^{2}=$ 0.108. $F=$ 3.118. ${}^{*}p<$ 0.05, ${}^{**}p<$ 0.01.

References

Abdi,

, & Williams,

L.J.

(2010). Principal component analysis. Wiley Interdisciplinary Reviews: Computational Statistics, 2(4), 433-459.

Ajzen,

(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.

Al-Hujran,

Al-Debei,

M.M.

Chatfield,

, & Migdadi,

(2015). The imperative of influencing citizen attitude toward e-government adoption and use. Computers in Human Behavior, 53, 189-203.

Alomari,

Woods,

, & Sandhu,

(2012). Predictors for e-government adoption in Jordan: Deployment of an empirical evaluation based on a citizen-centric approach. Information Technology and People, 25(2), 207-234.

Alsaif,

(2014). Factors affecting citizens’ adoption of e-government moderated by socio-cultural values in Saudi Arabia (Doctoral dissertation, University of Birmingham).

Al-Shafi,

, & Weerakkody,

(2010). Factors affecting e-government adoption in the state of Qatar. EMCIS2010.

Bandura,

(1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.

Bandura,

(1986). Social foundations of thought and action (5-107). Prentice Hall: Englewood Cliffs, NJ.

Bandura,

(1989). Social cognitive theory. In Vasta,

(Ed.), Annals of Child Development. Vol. 6. Six Theories of Child Development (pp. 1-60). JAI Press: Greenwich, CT. p2.p4

10.

Bandura,

(1997). Self-efficacy: The exercise of control. W.H. Freeman: New York.

11.

Bandura,

(2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52(1), 1-26.

12.

Bandura,

(2012). On the functional properties of perceived self-efficacy revisited. Journal of Management, 38(1), 9-44.

13.

Bélanger,

, & Carter,

(2012). Digitizing government interactions with constituents: An historical review of e-government research in information systems. Journal of Association for Information System, 13(5), 363-394. p333.

14.

Bonsón,

Royo,

, & Ratkai,

(2017). Facebook practices in Western European municipalities: An empirical analysis of activity and citizens’ engagement. Administration & Society, 49(3), 320-347.

15.

Carter,

, & Bélanger,

(2005). The utilization of e-government services: Citizen trust, innovation and acceptance factors. Information System Journal, 15(1), 5-25.

16.

Carter,

Weerakkody,

Phillips,

, & Dwivedi,

Y.K.

(2016). Citizen adoption of e-government Services: Exploring citizen perceptions of online services in the United States and United Kingdom. Information Systems Management, 33(2), 124-140.

17.

Central Intelligence Agency (CIA) (2013). The World Factbook – United Arab Emirates.

18.

Chang,

I.C.

Li,

Y.-C.

Hung,

W.-F.

, & Hwang,

H.-G.

(2005). An empirical study on the impact of quality antecedents on taxpayers’ acceptance of Internet tax-filing systems. Government Information Quarterly, 22(3), 389-410.

19.

Compeau,

D.R.

, & Higgins,

C.A.

(1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19(2), 189-211.

20.

Coursey,

, & Norris,

D.F.

(2008). Models of e-government: Are they correct? An empirical assessment. Public Administration Review, 68(3), 523-536.

21.

Davis,

F.D.

(1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-40.

22.

Davis,

F.D.

Bagozzi,

R.P.

, & Warshaw,

P.R.

(1992). Extrinsic and intrinsic motivation to use computers in the workplace. Journal of Applied Social Psychology, 22(14), 1111-1132.

23.

Deci,

E.L.

, & Ryan,

R.M.

(2010). Self-Determination. John Wiley and Sons, Inc.: New York.

24.

Diamond,

, & Sekhon,

J.S.

(2013). Genetic matching for estimating causal effects: A general multivariate matching method for achieving balance in observational studies. Review of Economics and Statistics, 95(3), 932-945. DOI: 10.1016/j.im.2014.06.004. Available at: https://www-sciencedirect-com.web.bisu.edu.cn/science/article/pii/S0378720614000810.

25.

Doong,

Wang,

, & Foxallc,

G.R.

(2010). Psychological traits and loyalty intentions towards e-government services. International Journal of Information Management, 30(5), 457-64.

26.

Ductor,

Fafchamps,

Goyal,

, & van der Leij,

M.J.

(2011). Social networks and research output. Review of Economics and Statistics, 96(5), 936-948.

27.

Dunleavy,

, & Margetts,

(2010). The second wave of digital era governance. In American Political Science Association Conference, September 4, 2010, Washington DC, USA. http://eprints.lse.ac.uk/27684/.

28.

Dunleavy,

Margetts,

Bastow,

, & Tinkler,

(2006). Digital era governance: IT corporations, the state, and e-government. Oxford University Press.

29.

Fishbein,

, & Ajzen,

(1975). Belief, attitude, intention and behavior: An introduction to theory and research. Addison-Wesley: Reading.

30.

Fountain,

F.J.

(2001). Building the virtual state: Information technology and institutional change. Brookings Inst. Press: Washington, DC.

31.

Gao,

, & Lee,

(2017). E-government services and social media adoption: Experience of small local governments in Nebraska state. Government Information Quarterly, 34(4), 627-634.

32.

Gilbert,

Balestrini,

, & Littleboy,

(2004). Barriers and benefits in the adoption of e-government. International Journal of Public Sector Management, 17(4), 286-301.

33.

Godfrey,

L.G.

(2011). Robust non-nested testing for ordinary least squares regression when some of the regressors are lagged dependent variables. Oxford Bulletin of Economics and Statistic, 73(5), 651-668.

34.

Goodhue,

D.L.

, & Thompson,

R.L.

(1995). Task-technology fit and individual performance. MIS Q, 213-236.

35.

Gouscos,

Kalikakis,

Legal,

, & Papadopoulou,

(2007). A general model of performance and quality for one-stop e-government service offerings. Government Information Quarterly, 24(4), 860-885.

36.

Gunawong,

, & Gao,

(2017). Understanding e-government failure in the developing country context: a process-oriented study. Information Technology for Development, 23(1), 153-178.

37.

Hung,

S.-Y.

Chang,

C.-M.

, & Kuo,

S.-R.

(2013). User acceptance of mobile e-government services: An empirical study. Government Information Quarterly, 30(1), 33-44.

38.

Hung,

S.-Y.

Chang,

C.-M.

, & Yu,

T.-J.

(2006). Determinants of user acceptance of the e-government services: The case of online tax filing and payment system. Government Information Quarterly, 23(1), 97-122.

39.

Im,

Hong,

, & Kang,

M.S.

(2011). An international comparison of technology adoption: Testing the UTAUT model. Information Management, 48(1), 1-8.

40.

International Monetary Fund (IMF) 2010, World Economic Outlook Database. IMF: Washington, D. C.

41.

Kaiser,

H.F.

(1960). The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20(1), 141-151.

42.

Alcaide-Muñoz,

, & Rodríguez Bolívar,

M.P.

(2015). Understanding e-government research: A perspective from the information and library science field of knowledge. Internet Research, 25(4), 633-673.

43.

Lean,

O.K.

Zailani,

Ramayah,

, & Fernando,

(2009). Factors influencing intention to use e-government services among citizens in Malaysia. International Journal of Information Management, 29(6), 458-475.

44.

Legris,

Ingham,

, & Collerette,

(2003). Why do people use information technology? A critical review of the technology acceptance model. Information and Management, 40(3), 191-204.

45.

Leonard-Barton,

, & Deschamps,

(1988). Managerial influence in the implementation of new technology. Management Science, 34(10), 1252-1265.

46.

Liang,

S.W.

, & Lu,

H.P.

(2013). Adoption of e-government services: an empirical study of the online tax filing system in Taiwan. Online Information Review, 37(3), 424-442.

47.

Liu,

S.M.

, & Yuan,

(2015). The evolution of information and communication technology in public administration. Public Administration and Development, 35, 140-151.

48.

Loo,

W.H.

Paul,

H.P.

Yeow,

P.H.P.

, & Chong,

S.C.

(2009). User acceptance of Malaysian government multipurpose smartcard applications. Government Information Quarterly, 26(2), 358-367.

49.

Luna-Reyes,

L.F.

Gil-Garcia,

J.R.

, & Romero,

(2012). Towards a multidimensional model for evaluating electronic government: Proposing a more comprehensive and integrative perspective. Government Information Quarterly, 29(3), 324-334.

50.

Marakas,

G.M.

Johnson,

R.D.

, & Clay,

P.F.

(2007). The evolving nature of the computer self-efficacy construct: An empirical investigation of measurement construction, validity, reliability and stability over time. Journal of Association for Information System, 8(1), 16-46.

51.

Marakas,

G.M.

Yi,

M.Y.

, & Johnson,

R.D.

(1998). The multilevel and multifaceted character of computer self-efficacy: Toward clarification of the construct and an integrative framework for research. Information System Research, 9(2), 126-163.

52.

Mencía,

, & Sentana,

(2012). Distributional tests in multivariate dynamic models with normal and student-t innovations. Review of Economics and Statistics, 94(1), 133-152.

53.

Morgeson,

F.V.

VanAmburg,

, & Mithas,

(2011). Misplaced trust? Exploring the structure of the e-government-citizen trust relationship. Journal of Public Administration Research and Theory, 21(2), 257-283.

54.

Nulty,

D.D.

(2008). The adequacy of response rates to online and paper surveys: what can be done? Assessment & Evaluation in Higher Education, 33(3), 301-314.

55.

Perrin,

(2015). Social media usage: 2005–2015. Pew research center, 52-68.

56.

Phang,

C.W.

Li,

Sutanto,

, & Kankanhalli,

(2005). Senior citizens’ adoption of e-government: In quest of the antecedents of perceived usefulness. In Paper presented at the 38th Hawaii International Conference on System Sciences, Big Island, HI, January 3-6.

57.

Porumbescu,

G.A.

(2016). Comparing the effects of e-government and social media use on trust in government: evidence from seoul, South Korea. Public Management Review, 18(9), 1308-1334.

58.

Rana,

N.P.

, & Dwivedi,

Y.K.

(2015). Citizen’s adoption of an e-government system: Validating extended social cognitive theory (SCT). Government Information Quarterly, 32(2), 172-181.

59.

Reece,

(2006). E-government literature review. Journal of e-Government, 3(1), 69-110.

60.

Rogers,

E.M.

(1995). Diffusion of innovations. The Free Press: New York.

61.

Salary Explorer. (2014). Average and Median Monthly Salary Comparison in Dubai, http://www.salaryexplorer.com/salary-survey.php?loc=2426andloctype=3.

62.

Sang,

Lee,

J.D.

, & Lee,

(2009), E-government adoption in ASEAN: the case of Cambodia. Internet Research, 19(5), 517-534.

63.

Sax,

L.J.

Gilmartin,

S.K.

, & Bryant,

A.N.

(2003). Assessing response rates and nonresponse bias in web and paper surveys. Research in Higher Education, 44(4), 409-432.

64.

Shareef,

M.A.

Kumar,

, & Dwivedi,

Y.K.

(2011). E-government adoption model (GAM): Differing service maturity levels. Government Information Quarterly, 28(1), 17-35.

65.

Sharma,

, & Mishra,

(2017). Investigating the role of intermediaries in adoption of public access outlets for delivery of e-government services in developing countries: An empirical study. Government Information Quarterly, 34(4), 658-679.

66.

Straub,

E.T.

(2009). Understanding technology adoption: Theory and future directions for informal learning. Review of Educational Research, 79(2), 625-649.

67.

Taylor,

, & Todd,

P.A.

(1995). Understanding information technology usage: A test of competing models. Information System Research, 6(2), 144-176.

68.

Thompson,

R.L.

Higgins,

C.A.

, & Howell,

J.M.

(1991). Personal computing: Toward a conceptual model of utilization. MIS Quarterly, 15(1), 125-143.

69.

Torkzadeh,

Chang,

J.C.J.

, & Demirhan,

(2006). A contingency model of computer and Internet self-efficacy. Information and Management, 43(4), 541-550.

70.

United Nations. (2012). United Nations e-government survey 2012. E-government for the people. United Nations: New York.

71.

United Nations. (2014). United Nations e-government survey 2014): E-government for the future we want. United Nations: New York.

72.

United Nations. (2016). United Nations e-government survey 2016: E-government in support of sustainable development. United Nations: New York.

73.

United Nations Development Program, (2013). 2013 Human Development Report, United Nations: New York.

74.

Venkatesh,

(1999). Creation of favorable user perceptions: Exploring the role of intrinsic motivation. MIS Quarterly, 239-260.

75.

Venkatesh,

Morris,

M.G.

Davis,

G.B.

, & Davis,

F.D.

(2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425-478.

76.

Venkatesh,

Sykes,

T.A.

, & Venkatraman,

(2014). Understanding e-government portal use in rural India: role of demographic and personality characteristics. Information Systems Journal, 24(3), 249-269.

77.

Warkentin,

Gefen,

Pavlou,

, & Rose,

(2002). Encouraging citizen adoption of e-government by building trust. Electronic Market, 12(3), 157-62.

78.

Weerakkody,

El-Haddadeh,

Al-Sobhi,

Shareef,

M.A.

, & Dwivedi,

Y.K.

(2013). Examining the influence of intermediaries in facilitating e-government adoption: An empirical investigation. Information and Management, 33(5), 716-725.

79.

Wigand,

R.T.

Wood,

J.D.

, & Mande,

D.M.

(2010). Taming the social network jungle: From Web 2.0 to social media. AMCIS 2010 Proceedings. Paper 416.

80.

World Bank. (2008). Definition of e-government. Available at: http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTINFORMATIONANDCOMMUNICATIONANDTECHNOLOGIES/EXTEGOVERNMENT/0,,contentMDK:20507153∼menuPK:702592∼pagePK:148956∼piPK:216618∼theSitePK:702586,00.html [July 9, 2009].

81.

The World Bank (2016). United Arab Emirates. https://data.worldbank.org/country/united-arab-emirates?view=chart.

82.

Wu,

, & Chen,

(2017). Continuance intention to use MOOCs: Integrating the technology acceptance model (TAM) and task technology fit (TTF) model. Computers in Human Behavior, 67, 221-232.

83.

Wyer Jr,

R.S.

(Ed.). (2014). The automaticity of everyday life: Advances in social cognition (Vol. 10). Psychology Press: Oxford.

84.

Yildiz,

(2007). E-government research: Reviewing the literature, limitations, and ways forward. Government Information Quarterly, 24(3), 646-665.

85.

Zhao,

(2013). An empirical study of cultural dimensions and e-government development – Implications of the findings and strategies. Behaviour and Information Technology, 32(3), 294-306.

86.

Zhao,

, & Khan,

(2013). An empirical study of e-government service adoption: Culture and behavioral intention. International Journal of Public Administration, 36(10), 710-722.

87.

Zhao,

Collier,

, & Deng,

(2014a). A multidimensional and integrative approach to study global digital divide and e-government development. Information Technology and People, 27(1), 38-62.

88.

Zhao,

Shen,

K.N.

, & Collier,

(2014b). Effects of culture on e-government diffusion: A global study of 55 countries. Information & Management, 51(8), 1005-1016.

89.

Zhao,

Naidu,

Singh,

Sewak,

Chand,

, & Karan,

(2017). An empirical study of e-government diffusion in Fiji: a holistic and integrative approach. Public Management Review, https://doi.org/10.1080/14719037.2017.1400585.

90.

Zhou,

(2011). Understanding online community user participation: A social influence perspective. Internet Research, 21(1), 67-81.

An empirical study of e-government adoption in the United Arab Emirates: A social cognitive perspective

Abstract

Keywords

1. Introduction

2. Social cognitive theory and technology adoption model

3.1.1 Personal factors and measures

3.1.2 Behavioral factors (patterns) and measures

3.2 Environmental factors interacting with personal factors

3.2.1 Environmental factors and measures

3.3 Behavioral factors interacting with environmental factors

4. Methodology

4.1 Survey

4.2 Sample

Table 2 Summary of demographic data of survey participants ( n = 355)

Table 3 Results of principal component analysis for behavioral factors, environmental factors, and e-government adoption

Table 4 Regression results of ordinary least squares with autoregressive errors for interactions among personal, behavioral, and environmental factors for hypotheses: H1–H3

6.1 Theoretical contribution and implications

6.2 Strategic implications

7. Limitations and future research

8. Conclusion

Footnotes

Appendix 1. Sample questions used for the survey for this study *

Appendix 2

References

Table 2
Summary of demographic data of survey participants ( $n=$ 355)

Table 3
Results of principal component analysis for behavioral factors, environmental factors, and e-government adoption

Table 4
Regression results of ordinary least squares with autoregressive errors for interactions among personal, behavioral, and environmental factors for hypotheses: H1–H3