The Effects of Human Resources Quality, Infrastructure, Leadership, and Communication on E-Government Implementation: A Case of Indonesia Local Government

In the 21st century, or the era widely referred to as the digital era, the most developed information technologies are webor Internet-based (Sethuraman, 2017). Internet technologies amongst the community can meet the information needs instantly, precisely, and accurately (Setiawan, 2017). This current phenomenon makes information technologies necessary that every individual, group, company, or government institution must fulfill (Cui et al., 2015). In general, every local government strives to run the government wheel well to realize a good government through better public services (Akib et al., 2016). To this end, both central and local governments have started to implement an electronic government system commonly called e-government (Salsabila & Purnomo, 2017). To implement this egovernment system, local governments are expected to apply care in seeing and analyzing the factors that can influence its success in a government environment (Stefanovic et al., 2016). Lee (2009) states that e-government implementation in numerous countries falls short of expectations. Heeks’s research (2003) revealed that 35% of e-government projects in developing countries experienced total failures, 50% experienced partial failures, and only 15% were considered successful. Heeks (2003) argued that some factors contributed to e-government implementation failures in developing countries; for example, the public administration system, internal rejection by the government, a lack of plans and strategies, partial, unsystematic e-government introduction, and the insufficient number of human resources. The other factors are an absence of investment plans, the insufficient number of information system and technology vendors, technological immaturity, too much emphasis on technology and technology-based implementation, and hurried implementation without adequate preparation and testing. Abstract

the effects of human resources quality, infrastructure, leadership, and communication on egovernment implementation in the Yapen Islands Regency.
In this study, the researchers set a scope for the variables information and communications infrastructure and leadership to prevent too broad a discussion. The discussion of information and communications infrastructure was restricted only to Internat/Intranet/LAN network, while leadership was restricted to Echelon II.b and Echelon III local government organization leaders.

E-Government
E-government by Khalil et al. (2002) is defined as the use of information technologies, such as a wide area network (WAN), by a government agent. The Internet with mobile computing can transform relationships with the community, businesses, and government institutions. According to Indrajit et al. (2005), e-government is a novel interaction mechanism between the government and the community and other interested parties with information technology (especially the Internet) use for service quality improvement.

Human Resources Quality
As posited by Sedermayanti (2009), quality is a measure that states the extent to which various requirements, specifications, and expectations are fulfilled. Meanwhile, human resources or employees in an organization play an essential role in the organization's success. Human resources quality, according to Matindas (2002), is the willingness of every employee to complete his/her work, develop him-/herself, and encourage his/her colleagues' self-development.
Human resources are one of the factors that affect e-government implementation success (Surdin, 2017). Organizations are in grave need of competent human resources with high knowledge for accurate, practical e-government system use (Choi et al., 2016). Previous research by Ariana et al. (2020), Multama et al. (2018), Probowulan (2016), Novita (2014), and Lee (2009) showed that qualified human resources influence egovernment implementation. Hence, referring to previous empirical findings, the first hypothesis to be proposed is as follows: H1: Human resources quality affects e-government implementation in the environment of the Yapen Islands Regency government.

Infrastructure
The American Public Works Association in Kodoatie (2005) defines the infrastructure as the physical facilities developed or needed by public agents to serve government functions in the provision of water, electrical power, waste disposal, transportation, and similar services to facilitate social and economic purposes. Meanwhile, Presidential Regulation No. 38 of 2015 defines infrastructure as the technical, physical, and hardware and software facilities needed to provide the community with services and to support structural networks for the community's profitable economic and social growth.
Infrastructure directly influences e-government implementation, and the infrastructure quality determines how successful or failed an e-government program is (Gyamfi et al., 2019;Sami et al., 2011). E-government success needs an appropriate information technology infrastructure to support the systems and applications was developed before the e-government program is implemented (Sirat & Komputer, 2013). Besides, previous research by Choi et al. (2016), Sorn-In et al. (2015), Novita (2014), Al-Shlboul et al. (2014), andSirat (2013) showed that infrastructure influences e-government implementation. Hence, referring to previous empirical findings, the second hypothesis to be proposed is as follows: H2: Infrastructure has an effect on e-government implementation in the environment of the Yapen Islands Regency government.

Leadership
Northouse (2003) says that leadership is an individual's endeavor to control a group for his/her interests. Meanwhile, Dubrin (2015) states that leadership is about instilling belief to win members' support to attain an organization's goals. It also has characteristics of originality, anti-establishment, the acknowledgment that responsibility is in the leader's hands, and commitment to the right things (Rizanuddin, 2020).
Strong leadership is needed to accelerate the e-government implementation process and to deal with the problems that frequently arise due to the involvement of coordination with a variety of sectors (Rante & Warokka, 2016;CP & Susanto, 2019). Strong leadership is also vital for ensuring the success of e-government success as an effective leader will influence his organization's employees (Al-Azri et al., 2010). Previous research by CP and Susanto (2019), Al-Shboul (2014), Sirat (2013), Al-Kaabi (2010), andAl-Azri et al. (2010) showed that leadership influences e-government implementation. Thus, referring to previous empirical findings, the third hypothesis proposed is as follows: H3: Leadership has an effect on e-government implementation in the environment of the Yapen Islands Regency government.

Communication
As stated by Mulyana (2012), communication can be discerned in three manners: communication as one-way action, communication as interaction, and communication as a transaction. Tubbs and Moss, as quoted by Mulyana (2005), further explain that communication is a process of meaning construction between two people or more. Lasswell (2017) defines communication as a one-way message delivery process that may generate an effect.
Communication has a significant effect on the implementation success of a policy because good communication will smoothen the policy implementation according to the goals that have been set when the policy was made (Suriyani, 2018;Hartati, 2020). Ziadi et al. (2016) and Al-Kaabi (2010) showed that communication influences e-government implementation. Thus, referring to previous empirical findings, the fourth hypothesis to be proposed is as follows: H4: Communication has an effect on e-government implementation in the environment of the Yapen Islands Regency government.
The four hypotheses proposed are then summarized in the following research model ( Figure 1).

III. Research Methods
This research used quantitative methods, which included data collection and analysis with statistical testing methods. The population of this research was 650 state civil officers in several local government agencies in the Yapen Islands Regency in Papua. Slovin's formula was used to determine the sample size. From the returned 246-questionnaires, only 245 questionnaires were used for the next process.
The data analysis methods used in this research consisted of descriptive statistical analysis, research instrument test, classical assumption test, multiple regression analysis, and hypotheses testing. In this research, data were processed with the SPSS version 22. The variables in this research consisted of independent and dependent variables. The independent variables were human resources quality, infrastructure, leadership, and communication, while the dependent variable was e-government implementation.

IV. Results and Discussion
In this research, the respondents were public officers who worked in the Yapen Islands Regency. Some selected local government agencies served as representatives. These local apparatus organizations were the ones that used e-government the most. The respondents' characteristics were focused on gender, age, latest education level, number of years worked, and position (see Table 1). Based on the respondents' characteristics data, the majority of the respondents were male (62.86%), aged older than 40 years (79.6%), in possession of a Bachelor's degree (79%), and Echelon III position (34%).
Descriptive analysis was used to describe the conditions and characteristics of the respondents' answers for each construct/factor or variable studied. The analysis results were used to identify the tendency of the respondents' answers regarding each variable studied. The variable human resources quality consisted of eight indicators, infrastructure five indicators, leadership 14 indicators, communication nine indicators, and e-government nine indicators.
The respondents' answers were categorized with an interval calculated by dividing the delta of maximum and minimum scores by five. The calculation yielded an interval of 0.80. According to Sugiyono (2013), with an interval of 0.80, the categorization system was as follows: 1.00-1.80 (very low); 1.81-2.60 (low); 2.61-3.40 (fair); 3.41-4.20 (high); and 4.21-5.00 (very high).  Based on the respondents' opinions distribution presented in Table 2, the respondents had an overall judgment that human resources quality, infrastructure, leadership, and communication had strong effects on e-government implementation. This finding is seen in the mean scores generated-3.52, 3.75, 3.95, 4.08, and 3.56, respectively which belonged to the high category (within the range 3.41-4.20).  (Ridwan & Sunarto, 2013). Validity testing was conducted with the Pearson Product Moment test. The result shows that all the statements had r-statistic values more significant than the t table of 0.125, meaning that all the statements were valid. Other than the validity test, a reliability test was also conducted to examine whether the instrument had a high-reliability level. The result shows that all the variables had Cronbach's alpha values greater than 0.6, meaning that all the variables were reliable. The classical assumption test is one of the requirements that must be fulfilled before conducting multiple regression analysis. This test consists of the normality test, multicollinearity test, and heteroscedasticity test. A normality test was conducted to examine whether the variables in the regression model had normally-distributed data or not. The result shows that Asymp. Sig obtained was 0.386, which was greater than 0.05. Thus, it can be said that the data were normally distributed.
A multicollinearity test was conducted to figure out whether in the regression model a correlation between independent variables was present. The result shows that all the independent variables had tolerance values > 0.10. This finding means that no problem was found in the multicollinearity test. Besides, from the VIF calculation, all the independent variables had VIF values < 10. It can be concluded that no multicollinearity was present between the independent variables in the regression model. Then, a heteroscedasticity test was conducted to figure out whether there was a similarity in the variability of residuals across different observations. A regression model meets the requirement if it has the same variability of residuals from one observation to another.
Based on the heteroscedasticity testing with the Glejser test, the result of which is presented in Table 4, it can be concluded that each independent variable had a sig.-value greater than 0.05. In other words, no heteroscedasticity was present.
Based on the result of the SPSS-(Statistical Product and Service Solutions) aided multiple regression analysis as shown in Table 4.5, the following regression equation was obtained: Y = 2.944 + 0.306X1 + 0.102X2 + 0.020X3 + 0.385X4  To examine the effects of independent variables on the dependent variable, a t-test was conducted. The t-statistics of the variable human resources quality was 7.207, which was greater than the t table 1.651, and the sig.-level was 0.000. It could be concluded that human resources quality (X1) significantly affected e-government implementation in the Yapen Islands Regency. Thus, the first hypothesis saying that human resources quality affects e-government implementation in the Yapen Islands Regency, was accepted.
This result is in line with the research by Ariana et al. (2020), Multama et al. (2018), Probowulan (2016, Novita (2014), and Lee (2009), which stated that human resource quality affects e-government implementation. This finding means that when human resources quality is improved, e-government implementation will become easier. In the context of the quality of the human resources of public officers (ASN) in the Yapen Islands Regency government, the role of human resources was central in the effort to achieve the regency's vision and missions. Human resources quality was highly influential to egovernment implementation. With human resources of quality, the existing e-government applications in local government organizations will be operable.
The variable infrastructure had t-statistics of 1.477, which was smaller than the ttable of 1.651, and the sig.-level was 0.141. It could be concluded that infrastructure (X2) had no effect on e-government implementation in the Yapen Islands Regency government. Thus, the second hypothesis saying that infrastructure affects e-government implementation in the environment of the Yapen Islands Regency, was rejected.
This result is not in line with the research by Choi et al. (2016), Sorn-In et al. (2015), Novita (2014), Al-Shlboul et al. (2014), andSirat (2013), but it agrees with that by Riyadh et al. (2019). Infrastructure had no significant effect on e-government implementation because of the challenges the government was facing in information technology infrastructure development (Riyadh et al., 2019).
Referring to the descriptive analysis results, the statement "Connection to LAN & the Internet is available" had a mean value of 3.42. The statement in the indicator X2.4 was perceived as the lowest among the five statements (indicators). This low score came from the fact that only a few local government organizations were connected to LAN and Internet networks. Based on the fact in the field, the e-government of the Yapen Islands Regency had featured the Corruption Eradication Commission program through the Corruption Eradication Commission Action Plan of 2017. However, to date, information and communications technology infrastructure have yet to become a priority in the development program of the Yapen Islands Regency. The local government was still dependent on central government aids, which were only aimed to stimulate local governments. This fact is as evident in the smaller amount of budget allocated for information and communications technologies than to other sectors. The 2020 activity programs in the Local Information and Communications Office, which was assigned with main tasks in information and communications technology infrastructure development, were allocated with an infrastructure fund of no more than Rp1.5 billion, equal to USD 100,000 (DPA Dinas Kominfo 2020).
The t-statistics of the next variable, leadership, was 0.520, which was smaller than the t table of 1.651, and the sig.-value was 0.603. It could be concluded that leadership (X3) had no partial effect on e-government implementation in the Yapen Islands Regency. Thus, the third hypothesis saying that leadership affects e-government implementation in the Yapen Islands Regency was rejected.
This result is not in line with the research by CP and Susanto (2019) Husain and Fauziati (2018) and Wicaksono (2013). Based on the respondents' opinions distribution, the statement "The leaders put broader interests first" had a mean value of 3.43. In other words, the statement in the indicator X3.13 was perceived as the lowest among the 14 statements (indicators), with a mean value (3.43) belonging to the neutral category. This finding is attributed to the leaders' differing leadership styles, in which case there were some who were focused more on broader interests and there were some who focused more on the internal interests of their respective local government agency or on doing something according to their main tasks and functions.
The t count of the variable communication was 5.201, which was greater than the t table of 1.651, and the sig.-level was 0.000. It could be concluded that communication (X4) had a significant effect on e-government implementation. Thus, the fourth hypothesis saying that communication affects e-government implementation in the Yapen Islands Regency, was accepted.
This result is in line with the research by Ziadi et al. (2016) and Al-Kaabi (2010), which stated that communication affects e-government implementation. The fact in the field indicates that communication in the environment of the Yapen Islands Regency government had a central role, especially in the satu (one) data program or the data center of the regency. Communication is critical for improving coordination between local apparatus organizations, particularly in data exchange or processing. For instance, the Regional Development Planning Agency (Bappeda) of the Yapen Islands Regency established good communication for smooth collection of local government agency's work planning data to draft the regional planning documents (RKPD) available in the E-Planning application. The implementation of e-government should be supported by good communication between local apparatus organizations. In other words, the better the communication established, the better the e-government implementation. Thus, it can be said that communication significantly influenced e-government implementation in the Yapen Islands Regency.
Lastly, the R -squared (coefficient of determination) test was conducted to measure the ability of the variables human resources quality (X1), infrastructure (X2), leadership (X3), and communication (X4) to explain the variation in the variable e-government implementation (Y). The adjusted R-squared value obtained from the test was 0.466. The interpretation for this value is that 46.6% of the dependent variable e-government implementation was influenced by the independent variables human resources quality, infrastructure, leadership, and communication. The remaining 53.34% was influenced by other variables unexplored in this research.

V. Conclusion
This research examined the effects of human resources quality, infrastructure, leadership, and communication on e-government implementation in the Yapen Islands Regency. Based on the results of the tests conducted, the following conclusions were drawn. First, human resources quality affected e-government implementation. This finding means that if the human resources' quality is improved, e-government implementation in the Yapen Islands Regency will become easier. Second, infrastructure and leadership did not influence the success of e-government implementation in the Yapen Islands Regency. Finally, communication had an effect on e-government implementation in the Yapen Islands Regency. The better the communication, the better the e-government implementation.
The results of this research have several critical implications. First, the local government agencies in the Yapen Islands Regency should continuously improve the quality of their public officers through both improved education and training to enhance their ability to run e-government successfully. Then, better and improved-communication networks and platforms should be established among local government agencies in the Yapen Islands Regency to ensure a better and coordinated e-government implementation.
It is also expected that leaders of local government agencies play a greater role in egovernment implementation, giving a more significant force to their leadership role in influencing e-government implementation. Concerning infrastructure, the Yapen Islands Regency, through local government agencies' main tasks and functions, are expected to continuously improve and give priority to network infrastructure development for the egovernment implementation to operate as designed.