International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 A Two-Staged SEM-Neural Network Approach for Predicting the Determinants of Mobile Government Service Quality Fakhar Shahzada, Zeeshan Ahmadb, Nadir Munir Hassanc, Muhammad Rafiqd, aDepartment of Business Administration, ILMA University, Karachi, Pakistan, b,cDepartment of Business Administration, Air University Multan, Pakistan, dDepartment of Economics and Business Administration, University of Education, Lahore, Vehari Campus, Email: afshahzad51@yahoo.com, b c zeeshan.ahmad@aumc.edu.pk, nadir.magsi@aumc.edu.pk, d m.rafiq@ue.edu.pk Rapid advances in technology have transformed the world, forcing governments worldwide to move from electronic government to mobile government (m-Govt) in delivering quality services. The mGovt service quality (mGSQ) is an arrogant concept, but so far, it has been relatively rare in this growing field. Thus, this research intends to explore the most critical factors predicting mGSQ, especially overseas citizens’ cognition of the mGSQ. The online survey was conducted to collect data on overseas Pakistanis living in mainland China. A multi-analytic approach verified the valid responses of 345 overseas users of m-Govt services. The structural equation model (SEM) was used as the input of the artificial neural network (ANN) model to predict the main factors influencing mGSQ. The outcomes revealed that infrastructure quality is the most important strength of m-Govt’s quality of service. The outcomes of this investigation provide an aerial perspective for government and practice to improve mGSQ. Key words: m-Govt services; service quality; ANN; predictive modeling. 159 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 1. INTRODUCTION The electronic government (e-Govt) includes the use of information and communication technology (ICT) to bring government services and information to the individual effectively, business as well as between the government departments around the globe (Al-Hubaishi et al., 2017; Ebrahim & Irani, 2005; Rana et al., 2014; Shahzad, Xiu, Khan, & Shahbaz, 2020). Meanwhile, the active penetration of mobile devices increased the citizen habits to get government information using mobile services, which also push the governments to deliver efficient services using mobile technology in the podium of m-Govt (Al-Hadidi & Rezgui, 2010; Alsaadi et al., 2019; Iskender & Özkan, 2013; Kushchu & Kuscu, 2004). The trend of m-Govt services is rapidly growing, which denotes the provision of government services over mobile application or interactive messages (Al-Hubaishi et al., 2017; Almarashdeh & Alsmadi, 2017; Dutra & Soares, 2018; Kumar & Sinha, 2007; Shareef et al., 2014, 2016; Sheng & Trimi, 2008) However, despite the requirements, citizens cannot consider m-Govt services to be effective if they are not able to contact government services through their mobile phones to fulfill their expected tasks. The service quality in its contemporary meaning refers to comparing user-perceived expectations and a service's perceived performance. However, numerous researchers (e.g., Cronin, Brady, & Hult, 2000; Parasuraman, Zeithaml, & Berry, 1988; Shareef, Dwivedi, Stamati, & Williams, 2014; Zeithaml, Parasuraman, & Malhotra, 2002) argued that the service quality of the virtual medium is diverse in various aspects of traditional service quality. As a result, poor public service performance can undermine government performance and negatively affect government systems' fundamental credibility (Shareef et al., 2014). To evaluate the mGSQ, the researchers must first understand the dimensions affecting mGSQ. Although the previous literature, particularly in mGSQ, is not robust compared to e-Govt from the citizen’s perspective (Al-Hubaishi et al., 2017). The authors believe that this study will provide a conceptual base for studying the necessary conditions for determining m-Govt services' quality. This study offers a theoretical understanding of citizens’ perception of mGSQ and provides practical acumens for government agencies and officials to improve the sustainability of mGSQ, especially for overseas citizens. In response to this study's research question (what are the potential dimensions that affect improving the quality of the m-Govt service from overseas citizen perceptive), authors are interested in revealing the perception of overseas citizens on the service quality dimensions the podium of m-Govt. In examining the potential dimensions of mGSQ, considerable research has been conducted on literature related to e-Govt and mGSQ, which assist in exploring the concept in more detail. Literature related to e-Govt can be applied to understand the phenomena of mGSQ. However, due to the mobility of mobile technology, m-Govt has its own features which required investigation separately and need a comprehensive model to evaluate its service quality (Al-Hubaishi et al., 2017; Chanana et al., 2016; Shahzad, Xiu, Khan, & Shahbaz, 2020; 160 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Tan & Chou, 2008). In addition, there is no single way to explore the same concept in each industry. The prior researchers applied the e-Govt service quality model to the mobile environment. In order to establish a new and inclusive model to predict the dimensions of mGSQ, it is of considerable significance to explore the factors that discuss sustainability mGSQ. 2. RELEVANT LITERATURE 2.1 Conceptualization of transition from e-Govt to m-Govt The researchers define the concept of e-Govt according to its use in ICT systems and the provision of government services, which describe the governmental relationship with electronics (Ebrahim & Irani, 2005; Heeks & Bailur, 2007; Larosiliere & Carter, 2016; Shareef et al., 2015; Venkatesh et al., 2014). The scholars have claimed that m-Govt is a sub-division of e-Govt by wireless communication devices to provide government information and services (Althunibat et al., 2011; Bergman, 2001; Z. J. Chen et al., 2016). Meanwhile, it is suggested that m-Govt would not be supposed as a new type, but rather as a new method to permit an efficient government that allows accessing the services from anywhere and anytime. m-Govt is an emerging trend to reform, restructure, and transform public services to boost and modernize existing e-Govt services to make them more viable and adaptable (Shareef et al., 2014). The technological revolution in the field of the internet is not just for e-Govt. It gets the researcher’s attention towards the implementations and adoption of the m-Govt system and explores the shift from wire to wireless-based applications in public and private sectors (Kumar & Sinha, 2007). The m-Govt adds value to e-Govt because people can access government services using wireless electronic devices such as smartphones, personal digital assistants (PDAs), and Wi-Fi-enabled devices (De Lima, 2017). Regardless of its infancy, the m-Govt is growing with a set of composite tools and strategies that will change traditional governance's existing functional roles. People can carry an m-Govt access terminal while traveling, strengthening the citizen-government relationship (Al-Masaeed, 2013). This shift from e-Govt to m-Govt will raise several questions regarding the delivery and quality of m-Govt services. Therefore, this study is systematized to categorize the potential dimensions associated with the service quality of m-Govt. 2.2 Mobile Government service quality (mGSQ) The service quality can be termed as the level of difference between customer experience and perceived service performance (Zhou et al., 2010). It is also described as the link between customer and service provider and between perceived services and delivered services (Almuraqab & Jasimuddin, 2017; Markovic & Raspor, 2010). In terms of online services, e- 161 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 services have been recognized as the electronic provision of interactive network-based services over the internet (Alsamydai, 2014). However, this concept is critically increasing in service organizations where the service organizations are suggested to enhance service quality by understanding user perceptions, particularly in e-Govt or m-Govt services (Brady & Cronin, 2001; Shareef et al., 2014; Yang, 2012). Former studies have revealed that service quality can trigger several positive customer reactions, such as loyalty, creating trust, satisfaction response of citizens to significant drivers of e-service adoption (Hao, 2016; Reichheld & Schefter, 2000; Schaupp & Belanger, 2005; Shahzad, Xiu, Khan, Shahbaz, et al., 2020; Welch et al., 2005). When it comes to mGSQ, research must be more specific, signifying that conceptualization and measurement should be grounded on user perception (Alsaadi et al., 2019; Chanana et al., 2016). We have not found any comprehensive way to define the quality of service of e-Govt and m-Govt. Some researchers (Ostrom et al., 2015; Shareef et al., 2014) believe that due to different service modes, the quality of virtual media service may differ from the traditional quality of service in several dimensions. Most researchers try to reveal m-Govt’s service system from the e-Govt literature, focusing on exploring its various adoption factors, which not only imitates m-Govt’s service perspective but also reflect the product, facility, and technology issues distinct to the service system, such as resource availability, awareness, cost and software (Ahmad & Khalid, 2017; Shahzad et al., 2019; Shanab & Haider, 2015; Shareef et al., 2012). Therefore, in this study, we take the prior concepts of e-commerce or e-Govt service quality as an entry point to conceptualize the public administration service quality provided through mobile devices. Consequently, this intends to categorize the influential factors affecting mGSQ, particularly on a user-centric approach. 3. MODEL DEVELOPMENT AND HYPOTHESES In this part, we explored the design of the m-Govt quality of service framework theoretically and methodologically. This study explores the critical constructs as perceived by overseas users in the cognition of mGSQ, which was not previously focused on by researchers. 3.1 Effort expectancy Effort expectancy mentions the extent of simplicity linked with a particular system (Davis et al., 1989). In line with the study of Venkatesh et al. (2003), effort expectancy summarizes the concept from previous adoption theories like the perceived ease of use and complexity. Similarities between these factors have been discussed in the previous literature concerning the user’s intention to use technology adoption research (Davis et al., 1989; Dwivedi et al., 2017; Lin et al., 2011; Shahbaz, Gao, Zhai, Shahzad, & Arshad, 2020; Shahbaz, Gao, Zhai, Shahzad, Abbas, et al., 2020; Weerakkody et al., 2013). In m-Govt services, the comprehensive and 162 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 straightforward system can reduce user efforts, increase its intention to use the system, and improve service quality. Therefore, we proposed the following hypothesis: H1. Effort expectancy will have a significant impact on mGSQ. 3.2 Perceived reliability Perceived reliability involves the user's perception that the service provider delivers the promised services accurately and reliably (Gefen, 2002). In m-Govt services, it refers to citizens' confidence in government services delivered by using mobile technology to facilitate the citizens (Ndou, 2004). Citizens are concerned with the accuracy and on-time delivery of services (Onashoga et al., 2016; Papadomichelaki & Mentzas, 2012). Reliability plays a pivotal role in ensuring citizen loyalty towards the use of m-Govt services (Gaviria-Marin et al., 2018; Hao, 2016; Shareef et al., 2014). Perceived reliability among citizens regarding the accessibility and availability of service contents would ensure that the transaction process and quality of m-Govt services will not fluctuate from the user’s end. Therefore, we proposed the following hypothesis: H2. Perceived reliability will have a positive impact on mGSQ. 3.3 Security and privacy People are more conscious regarding security and privacy infringement when interacting with virtual media such as e-Govt, e-commerce, or m-Govt, including identity theft, information, eavesdropping, and credit card abuse (Shareef, Kumar, Dwivedi, & Kumar, 2016). It also includes citizen perception regarding the security of personal information, the anonymity, risk of scam, and about the whole transaction carried by using such system (Colesca, 2009; Giovanis et al., 2012; Lallmahomed et al., 2017; Onashoga et al., 2016; Shareef et al., 2014; S. K. Sharma, 2015). The m-Govt services should require enhancing personal security and privacy by scrambling messages, gaining access to monitor, applying digital signatures, and getting usernames and passwords. Grounded in these arguments, citizen perception regarding the security and privacy issue of m-Govt is conceptualized as a pivotal factor in generating beliefs to remove the cognitive dissonance among citizens. Therefore, we proposed the following hypothesis: H3. Security and privacy will have a positive impact on mGSQ. 3.4 Information quality Information quality refers to “the system's ability to convey the intended meaning of information” (Wang & Lin, 2012). It determines the quality of information and system design and cost, completeness, accuracy, and format characteristics (AL Athmay et al., 2016; Alomari 163 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 et al., 2013). Therefore, in m-Govt services, information quality can be defined as the m-Govt system's ability to deliver the citizens with a novel, accurate, complete, transparent, and understandable information proposed as a key success factor to improve the mGSQ (Liu et al., 2008; Shahzad, Xiu, Khan, & Shahbaz, 2020; Shan et al., 2017). Due to the dearth of research in the m-Govt context, it would be fascinating to explore the government's information quality using mobile services and its impact on mGSQ. Therefore, we proposed the following hypothesis: H4. Information quality will have a positive impact on mGSQ. 3.5 Perceived valence of information Valence is mostly used in psychological research, especially when discussing individual emotions, which refers to the intrinsic attraction of an event, object, or situation (Antheunis et al., 2010). Meanwhile, the perceived valence of information refers to the degree of positive (goodness) or negative (badness) information about the target person, system, or service provider, regardless of its credibility or excellence (Al-Hubaishi et al., 2017; Dai et al., 2015; Liébana-Cabanillas & Alonso-Dos-Santos, 2017). It also affects the level of user uncertainty, which would harm the perceived service quality of the m-Govt system (Z. Chen & Dubinsky, 2003; Ismagilova et al., 2019). Therefore, we can assume that the user-perceived valence of information is a significant parameter to measure the mGSQ. Thus, we proposed: H5. Perceived valence will have a positive impact on mGSQ. 3.6 Infrastructure quality The quality of infrastructure can provide pivotal insight into system characteristics such as quality, comprehensiveness, accuracy, instant connectivity, and network availability, which assist in building and stabilizing user trust in service quality (Hole, 2016). The availability and quality of network infrastructure are significant to estimate the quality of mobile-based service systems (Chae et al., 2002; Das et al., 2017; Hsiao et al., 2010; R. Sharma & Jayasimha, 2016). The infrastructure quality of the network, would affect the user intends to engage in such a system because users are apathetic to use mobile services because of disconnection and lack of access (Al-Hubaishi et al., 2017). So, the government should be more focused on improving or establishing a high-quality infrastructure for the provision of m-Govt service, which increases the mGSQ and user intention towards government services. Thus, we proposed: H6. Infrastructure quality will have a positive impact on mGSQ. 3.7 Responsiveness Responsiveness generally refers to the willingness to interact with service providers to respond to customers and expected efficiency in time and user convenience (Parasuraman et al., 1988; 164 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Shanab & Haider, 2015). Because services are essentially intangible, an interaction that occurs during service delivery and quality of responsiveness has the maximum impact on service quality (Al-Hubaishi et al., 2017; Balasubramanian et al., 2003; Shanab & Haider, 2015). Because the government is offering only services over the m-Govt system instead of a tangible good, in dealing with citizen issues, the government should pay more attention to employees' attitudes (Saghafi et al., 2016; Wulandari, 2019). A government official responsible for dealing with citizens using mGSQ should deliver quality responses to build citizen trust in them. Because, positive response from the service provider or the government, in the case of m-Govt services, will raise service quality and citizen satisfaction. Thus, we proposed: H7. Responsiveness will have a positive impact on mGSQ. 3.8 Perceived Empathy Empathy reflects the consumers’ belief that they receive personalized consideration from the service provider (Parasuraman et al., 1988). Logically, these procedures do not reconstruct the extent of empathy as a human service provider; however, they interact with vendors individually and provide personalized services to a reasonable standard. The m-Govt services are provided in a virtual medium, apparently without any physical contact with the service provider; however, the firm belief that customer service exists at the rear of the screen can offer a constructive attitude to m-Govt (Moon & Norris, 2005). Empathy in the context of m-Govt can be referred to as an emotional belief that is a sincere and timely response in government service systems to meet special citizen requirements (Balasubramanian et al., 2003; Markovic & Raspor, 2010; Shareef et al., 2016). Therefore, it is required to provide IT-mediated content functions on the m-Govt system to provide service customization features, and they only work with minimal interactivity, which improves mGSQ. Thus, we proposed the following hypothesis: H8. Empathy will have a positive impact on mGSQ. In this study, an inclusive interpretation of the concept of mGSQ has been constructed on prior literature. We conceptualized a comprehensive research framework to measure mGSQ. Figure 1 shows the proposed research model, which demonstrates the critical factors influencing mGSQ. 165 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Figure 1 Research model for mGSQ 4. RESEARCH METHODOLOGY 4.1 The context of the research Literature shows that citizens’ intentions are crucial and measured before and after the implementation of m-Govt systems, which have an impact on its service quality (Almuraqab & Jasimuddin, 2017; Burja et al., 2006; Djamal & Renaldi, 2018; Dwivedi et al., 2018; Hameed et al., 2014; Shahzad et al., 2019; Shareef et al., 2016). However, these studies collect the required information from the citizens living and using mobile services within the country. Keep into consideration the mobility of m-Govt services, and this study tries to explore the intentions of overseas citizens concerning the quality of m-Govt services, which previous scholars potentially ignored. It is important because they have an imperative contribution to the country's economic growth and are required to keep in contact with the country’s matters. Therefore, the required data were gathered from the overseas Pakistani nationals, mainly living in mainland China. Citizens living abroad (short or long-term) have no direct physical access to government services and can only use electronic or mobile services to establish contact with government affairs. The m-Govt platform facilitates citizens worldwide to make explicit content to the government using mobile applications for their queries and other matters. The availability and quality of mobile services from the government do matter for these overseas citizens. Therefore, for the first time in this study, the researchers targeted overseas citizens to investigate the potential dimensions contributing to the expansion of mGSQ according to their point of view. 166 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 4.2 Measurement development A structured questionnaire was developed for data collection and answers to the research question to ensure the research framework's content validity. All measuring items were operationalized and adapted from the preliminary investigations. The questionnaire initiates with a summary explaining the study's aim and significance, comprised of close-ended questions. To encourage respondents to express themselves freely, precise and straightforward language was used to write the scale items. The questionnaire of this study was established using a seven-point Likert scale from 1 “Strongly Disagreed” to 7 “Strongly Agreed” as strongly recommended by (Kurfalı et al., 2017; Nathan & Scobell, 2012; Rana et al., 2017; Shahzad et al., 2019). 4.3 Data collection and sampling Questionnaires were distributed by generating online survey forms and spread electronically to the target population. By availing of authors' links in mainland China, friends are requested to circulate the prescribed link of an online questionnaire among Pakistani citizens. Using a snowball sampling technique, they are requested to share the survey link to others who have experience using any m-Govt services such as m-banking, m-health, m-education, m-complaint many more. The respondents are free to respond to the questionnaire, and no reward was granted to them. Approximately in three months, we received 371 responses against the survey’ and decided to perform the empirical analysis. To ascertain the effect of independent constructs on mGSQ, we applied Structural Equation Modeling (SEM). As per the studies (Kline, 2005; Shareef et al., 2011), a sample size of more than 200 is considered satisfactory and can be used for SEM. During the primary screening of data, 26 responses were excluded, out of which 11 were not completed, and in 15 responses, the respondents select the one option for all items showing a non-serious attitude towards the survey. The eligible sample size was determined as n= 345. 5. STATISTICAL ANALYSIS AND RESULTS 5.1 Sample Descriptive The present study is based on the sample of 345 respondents shown in Table 1, out of which 58.6% are male, and 41.4% are female actively joined in the survey. Around 22.3% of respondents belong to the age group less than 20 years, 50.7% belong to 20 to 29, and 18.8% belong to 30 to 39 years old. The rest, 8.1%, belong to the more than 40 years of age group. The results showed that relatively young people respond to the questionnaire. The sample comprises comparatively well-educated respondents, as 35.9% are graduates, and 26.4% are postgraduates. Meanwhile, 27.2% of respondents are undergraduates, and the rest, 10.4%, have other educational backgrounds. The overall results from Table 1 described that mostly young 167 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 and educated individuals who can comprehend the importance of mobile services usage provided by the government; are the respondents in this study. Table 1 Demographics of respondents Category Gender Age Education Frequency Percent Male 202 58.6 Female 143 41.4 Total 345 100.0 Less than 20 years 77 22.3 20-29 175 50.7 30-39 65 18.8 40-49 21 6.1 Above 50 years 7 2.0 Total 345 100.0 Undergraduate 94 27.2 Graduate 124 35.9 Postgraduate 91 26.4 Other (Diploma/ Professional 36 10.4 345 100.0 education) Total 5.2 Reliability analysis The constructs’ reliability is used to gauge the consistency among the measurement items. Cronbach's alpha is applied to calculate the inter-item consistency, and the threshold value for the reliability scale is above 0.70 (Fornell & Larcker, 1981). The values of Cronbach’s alpha of the current study data were summarized in Table 2, ranging from 0.883 to 0.980 for all constructs showing that there is not reliability concern. 168 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Table 2 Results of factor loadings, convergent validity, reliability, and descriptive statistics Cronbach’s SR Variables Items Loadings CR AVE Alpha EE1 .826 EE2 .831 Effort 1 0.980 0.980 0.925 expectancy EE3 .852 EE4 .834 PR1 .875 PR2 .761 Perceived 0.923 0.926 0.759 2 PR3 .872 reliability PR4 .730 SP1 .741 SP2 .883 Security and 0.933 0.937 0.789 3 SP3 .860 privacy SP4 .863 IQ1 .803 Information IQ2 .809 0.962 0.961 0.859 4 IQ3 .828 quality IQ4 .853 PV1 .756 Perceived PV2 .821 5 valance of 0.895 0.898 0.688 PV3 .863 information PV4 .755 InfQ1 .757 InfQ2 .782 Infrastructure 0.883 0.885 0.658 6 InfQ3 .768 quality .707 InfQ4 RPN1 .810 7 Responsiveness RPN2 .865 0.945 0.947 0.858 RPN3 .833 PE1 .772 PE2 .783 Perceived 0.889 0.900 0.692 8 PE3 .775 empathy PE4 .747 mGSQ1 .747 mGSQ2 .792 mGovt service 9 mGSQ3 .777 0.915 0.914 0.681 quality mGSQ4 .706 mGSQ5 .734 169 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 5.3 Measurement model Principal component analysis (PCA) was applied to observe items' exactness in their corresponding factors as per the threshold value of greater than 0.7 (Hair et al., 2010; Shahzad et al., 2019; Venkatesh et al., 2016). One of the items from responsiveness was excluded due to a low level of factor score. Table 2 describes that EFA values for all items are ranging from 0.707 to 0.875 in their respective factors. Moreover, confirmatory factor analysis (CFA) was applied to confirm model goodness-of-fit, assessment of convergent validity, and discriminant validity of the constructs in AMOS 23. 5.3.1 Convergent and discriminant validity The outcomes from Table 2 shows the values of the average variance extracted (AVE) and composite reliability (CR). The values of CR and AVE are above the threshold values of 0.7 and 0.50, respectively, which confirm the high level of convergent validity (Fornell & Larcker, 1981; Henseler et al., 2014). Table 3 explains the values of inter-construct correlations and the square root of AVE, which depict the survey instrument's discriminant validity (Fornell & Larcker, 1981). The diagonal values of the square root of AVE are higher than the values of inter-construct correlation across the table, confirming the survey's discriminant validity. The results verified that convergent and discriminant validity is not a problem in the selected survey questionnaire. 170 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Table 3 Inter-construct correlations and validity concerns Mean S. D mGSQ EE SP IQ PR PV PE InfQ mGSQ 0.361 0.920 0.825 EE 0.389 0.986 0.601 0.962 SP 0.254 0.961 0.504 0.448 0.888 IQ 0.377 0.962 0.578 0.533 0.502 0.927 PR 0.242 0.972 0.492 0.436 0.408 0.445 0.871 PV 0.250 0.908 0.500 0.440 0.339 0.455 0.413 0.830 PE 0.389 0.904 0.574 0.623 0.406 0.563 0.467 0.496 0.832 InfQ 0.377 0.891 0.583 0.578 0.463 0.614 0.416 0.498 0.536 0.811 RPN 0.281 0.956 0.530 0.463 0.492 0.519 0.388 0.430 0.467 0.496 RPN 0.926 Values are significant at <0.001. “Inclined lines rendered in Boldface show the Square Root of the AVE.” 5.3.2 Measurement model assessment CFA was employed to test the reliability and validity concerns for all factors in AMOS 23. The CFA model fit was measured using popular indices based on collected data. The values of CMIN/DF=1.803, CFI=0.965, NFI=0.925, TLI=0.960, SRMR=0.044, RMSEA=0.048, and PClose=0.716 are obtained from the results, which fulfill the minimum threshold requirements (Byrne, 2010; Hair et al., 2010). The results mentioned above lead toward confirmation of the good fitness of the measurement model. 5.4 Structural model A structural model has been used after the confirmation of reliability and validity. The results from Figure 2 describe the overall goodness-of-fit for the path model. The values of CMIN/DF=1.363, CFI=0.996, NFI=0.985, TLI=0.987, SRMR=0.055, RMSEA=0.032, and PClose=0.807 confirm the overall model fit (Byrne, 2010; Hair et al., 2010). R-square's value is 0.60, which revealed that the selected constructs caused 60% of the total variance in mGSQ in the research framework. 171 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Figure 2 SEM results for hypotheses testing To verify the support of the assumed relationship, the threshold value of the standardized path coefficient should reach the minimum value of significance at p<0.05 (Byrne, 2010, 2013). The result from Table 4 explains that the coefficient's beta values are significant at 0.01 level except “information quality,” which is significant at 0.05. All values of the path coefficient are significant, described in Table 4 and Figure 2. Infrastructure quality contributes to mGSQ, followed by perceived empathy, perceived valence of information, responsiveness, security and privacy, effort expectancy, perceived reliability, and information quality. Gender and age were measured as a control variable and did not find a meaningful relationship. Based on this empirical analysis, it is decided that the theoretical model is significantly acceptable. 172 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Table 4 Hypotheses results mGSQ mGSQ mGSQ mGSQ mGSQ mGSQ mGSQ mGSQ mGSQ mGSQ <--<--<--<--<--<--<--<--<--<--- EE PR SP IQ PV InfQ RPN PE Gender Age Estimate S.E. C.R. P .114 .112 .121 .102 .126 .155 .123 .141 -.003 .033 .039 .031 .047 .043 .046 .044 .041 .053 .049 .027 2.935 2.639 2.789 2.056 2.830 3.008 2.724 2.723 -.088 .968 .003 .008 .005 .040 .005 .003 .006 .006 .930 .333 5.5 Artificial neural network (ANN) approach An artificial neural network (ANN) is defined as a “massively parallel distributed processor made up of simple processing units, which have a natural propensity for storing experimental knowledge and making it available for use” (Tadeusiewicz, 1995). ANN model is more vital that can offer higher prediction accuracy, which conventional methods cannot perform, such as multiple regression analysis (MRA) and SEM (Chong, 2013a). The ANN has been extensively used in information system research because of its powerful computing ability and its convenience to predict the influencing factors accurately. Therefore, this study utilized the ANN model's strength to examine the most important predictor of mGSQ, which was not discussed yet (as per our best knowledge). Among various types of ANN models, this study used the most common type of Back-Propagation Multilayer Perceptron (MLP) in SPSS v25. The input layer consists of eight independent variables (EE, PR, SP, IQ, PV, InfQ, RPN, and PE), having significant SEM analysis results. In comparison, the output layer comprises a single variable named mGSQ. As per the literature, no rule-of-thumb is accessible to limit the number of hidden neurons in ANN. Therefore, this study examines the ANN model with 1 to 10 different hidden neurons, as suggested by (S. K. Sharma et al., 2017), using Hyperbolic tangent-activation function for both hidden-layer and output-layer. The results on seven nodes in a hidden layer are considered more complex and accurate to predict input variables' importance. 5.5.1 Validation of neural network In order to evade the issue of over-fitting, 10-fold cross-validation was executed, in which 70% of data points were used for network training, and the rest 30% were used for testing to 173 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 investigate the prediction exactness of the training network (Chong, 2013b; S. K. Sharma et al., 2017). The Root Mean Square of Error (RMSE) is commonly used as an indicator to evaluate the ANN model's predictive accuracy of both training and testing data of all ten neural networks. The average value for both training and testing datasets are also calculated and given in Table 5. From Table 5, average (training = 0.124, testing = 0.128) values of RMSE in the ANN model are also small, indicating that the prediction is entirely accurate and reliable. Table 5 Validation results of ANN model (RMSE values) Network configures Training Testing ANN1 .124 .128 ANN2 .124 .126 ANN3 .125 .122 ANN4 .123 .123 ANN5 .126 .129 ANN6 .123 .126 ANN7 .127 .137 ANN8 .122 .133 ANN9 .121 .130 ANN10 .119 .126 Average .124 .128 5.5.2 Sensitivity analysis The sensitivity analysis is described as the “importance of each independent variable” to investigate the extent to which the NN model's predicted value changes with the value of the independent variable (Chong, 2013b). Based on the findings shown in Table 6, “Infrastructure quality” is the most influencing predictor, which is consistent with SEM analysis outcomes. However, in contrast with the SEM results, security and privacy is the second most influencing factor from ANN analysis. Responsiveness is the third important factor, followed by effort expectancy, perceived empathy, perceived valence of information, perceived reliability, and information quality. Meanwhile, the normalized importance is arranged to describe influencing factors from higher to lower in Table 6. 174 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Table 6 Independent variable importance Variables Importance Infrastructure quality Security and privacy Responsiveness Effort expectancy Perceived empathy Perceived valence of information Perceived reliability Information quality .275 .149 .123 .109 .103 .096 .073 .071 Normalized Importance 100.0% 54.3% 44.7% 39.6% 37.4% 34.9% 26.4% 25.7% 6. DISCUSSION AND IMPLICATIONS This study intends to discover the sustainability of mGSQ by predicting the factors that affect overseas citizens’ perceptions. Based on our research analysis, we will address several imperative discussions. The developed model was tested in two stages; first, the causal association among the latent factors was analyzed using SEM; the second ANN model was applied to predict the most influencing factor contributing to m-Govt services quality from user perspectives. The SEM results confirmed that the proposed model could reach an acceptable level in the predictive capability of mGSQ. Meanwhile, all the structural model results are significant, and the indicators of measurement and structural models are within the threshold limits. The value of R-square describes the 60% variation due to selected factors. The values of the beta coefficient for all constructs are also significant; therefore, hypotheses are accepted. Moreover, SEM results indicated that “infrastructure quality” is the most influencing factor predicting the quality of the m-Govt services based on the overseas citizen's perception. Providing government services to overseas nationals is not an easy task. It requires an up-todate and vast development of organizational and technical infrastructure (R. Sharma & Jayasimha, 2016). Low quality of infrastructure will not be able to stipulate the necessary services to the overseas citizen and the rural areas. Therefore, the infrastructure quality of mGovt services influences the overall service quality, as also proved in the current study context. Meanwhile, the perceived empathy is the second imperative predictor of mGSQ, followed by the perceived valence of information, responsiveness, security and privacy, effort expectancy, perceived reliability, and information quality from the SEM outcomes. The positive outcomes are consistent with previous literature; however, the importance of these factors contributing to mGSQ is collectively investigated for the first time in this study. Similar to SEM analysis results, the results from ANN analysis found the “information quality” as the most influencing predictor of m-Govt services quality. However, unlike the SEM results, “security and privacy” is the second most influencing predictor, followed by responsiveness, effort expectancy, perceived empathy, the perceived valence of information, perceived 175 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 reliability, and information quality. This may be because the NN model detects the non-linear relationship between decision factors rather than the linear as detected in the structural model. 6.1 Theoretical implications The outcome theoretically contributes by integrating the prior narrative on the concepts of mGSQ. Second, this research model focuses on expanding the knowledge that affects the quality dimension of m-Govt services to formulate guidelines for the development and enactment of the m-Govt system. Third, from the theoretical viewpoint, this research will provide further information in improving online service quality aspects and particularly in mobile application research. Finally, an innovative two-stage (SEM-ANN) modeling is applied to provide an overall understanding from an analytical perspective, which overcomes the limitation in several prior studies on m-Govt services. Moreover, the ANN model's noncompensatory feature overcomes the deficiency of the compensatory nature of linear SEM. SEM evaluated the causal relationship among the constructs and verified by the ANN model, which allows sorting independent constructs based on their importance. 6.2 Practical implications Practically, this investigation supports the researchers and practitioners to develop an understanding of the most influential factors of mGSQ. The findings are based on data collected by overseas individuals, which represent their preference for mGSQ. This presents a bird’s eye view for the government to design a better mobile service system, especially when providing m-Govt services to overseas citizens. The findings broaden the insight of the mGSQ and factors that might involve m-Govt service development from the outset. Moreover, the results revealed that infrastructure quality is a highly stimulating factor in predicting mGSQ. Government officials are recommended to make strategies to improve the organizational and technical infrastructure quality to improve the service quality of m-Govt systems. The service quality of any system cannot be separated from personal data's security and privacy, which is also verified in the context of mGSQ. Therefore, the developers should ensure the “security and privacy” of citizens' data, which raises the willingness to adopt such services and perception regarding service quality. 7. CONCLUSION, LIMITATIONS, AND RESEARCH DIRECTIONS With the rapid development in mobile phones, governments around the globe are using technological advancements to discover innovative ways for service delivery while the demand for improved government service quality is also increasing over time (Al-Hubaishi et al., 2017; Chanana et al., 2016; Shareef et al., 2014). Because the m-Govt initiative is required to put tremendous efforts and the government’s budget allocations. Therefore, user perception regarding the mGSQ is affected by several critical factors required to be explored before 176 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 establishing and implementing such systems. Therefore, based on the two-stage analysis (SEM-ANN), this research extended the existing literature on mGSQ, which has different features from the e-Govt and any other online services with crucial consideration to technical and non-technical aspects. The results from both analyses founded that “infrastructure quality” is the most influencing predictor of mGSQ. The model outcomes could serve as a steppingstone to fill the existing gap in a growing field of mGSQ, particularly in emerging nations. Regardless of the prodigious contribution, the study acknowledges some limitations. First of all, the research's cross-sectional nature allows future researchers to conduct longitudinal research to evaluate user preferences more accurately over time. Second, keep in view the unique characteristics of culture; the proposed model can be applied to various cultural contexts to measure its impact on mGSQ. Third, the researchers collected the data only from those overseas citizens who have the experience to use m-Govt services only from China. Future researchers are recommended to obtain data from citizens across the globe to prove this research's generalizability. Finally, this study used snowball sampling for the collection of data. Larger sample sizes allow for the use of efficient sampling techniques, such as cluster sampling should be used to authenticate the results. 177 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 REFERENCES Ahmad, S. Z., & Khalid, K. (2017). The adoption of M-government services from the user’s perspectives: Empirical evidence from the United Arab Emirates. International Journal of Information Management, 37(5), 367–379. https://doi.org/10.1016/j.ijinfomgt.2017.03.008 Al-Hadidi, A., & Rezgui, Y. (2010). Adoption and diffusion of m-government: Challenges and future directions for research. IFIP Advances in Information and Communication Technology, 336 AICT, 88–94. https://doi.org/10.1007/978-3-642-15961-9_9 Al-Hubaishi, H. S., Ahmad, S. Z., & Hussain, M. (2017). Exploring mobile government from the service quality perspective. Journal of Enterprise Information Management, 30(1), 4–16. https://doi.org/10.1108/JEIM-01-2016-0004 Al-Masaeed, S. (2013). Towards a User-Centric Mobile Government in Jordan (Issue April). Brunel University. AL Athmay, Al. A. A., Fantazy, K., & Kumar, V. (2016). E-government adoption and user’s satisfaction: an empirical investigation. EuroMed Journal of Business, 11(1), 57–83. https://doi.org/10.1108/EMJB-05-2014-0016 Almarashdeh, I., & Alsmadi, M. K. (2017). How to make them use it? Citizens acceptance of M-government. Applied Computing and Informatics, 13(2), 194–199. https://doi.org/10.1016/j.aci.2017.04.001 Almuraqab, N. A. S., & Jasimuddin, S. M. (2017). Factors that Influence End-Users’ Adoption of Smart Government Services in the UAE: A Conceptual Framework. The Electronic Journal Information Systems Evaluation, 20(1), 11. Alomari, M. A., Ellrehail, H. H., & Al Shibly, H. (2013). Mobile-Government : Challenges and Opportunities Jordan as Case study Alomari Mohammad A . Balqa Applied University Balqa Applied University. International Journal of Business and Social Science, 4(12), 244–250. Alsaadi, M. R., Ahmad, S. Z., & Hussain, M. (2019). Improving the quality of mobile government services in the Gulf Cooperation Council: A quality-function-deployment approach. Journal of Systems and Information Technology, 21(1), 146–164. https://doi.org/10.1108/JSIT-11-2017-0116 Alsamydai, M. J. (2014). Adaptation of the Technology Acceptance Model (TAM) to the Use of Mobile Banking Services. International Review of Management and Business Research, 3(4), 2016–2028. www.irmbrjournal.com Althunibat, A., Zain, N. A. M., & Ashaari, N. S. (2011). Modelling the factors that influence mobile government services acceptance. African Journal of Business Management, 5(34), 13030–13043. https://doi.org/10.5897/AJBM11.2083 Antheunis, M. L., Valkenburg, P. M., & Peter, J. (2010). Getting acquainted through social network sites: Testing a model of online uncertainty reduction and social attraction. Computers in Human Behavior, 26(1), 100–109. https://doi.org/10.1016/j.chb.2009.07.005 178 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Balasubramanian, S., Konana, P., & Menon, N. M. (2003). Customer Satisfaction in Virtual Environments: A Study of Online Investing. Management Science, 49(7), 871–889. https://doi.org/10.1287/mnsc.49.7.871.16385 Bergman, M. K. (2001). The deep web: Surfacing hidden value. Journal of Electronic Publishing, 7(1), 1–11. https://doi.org/10.3998/3336451.0007.104 Brady, M. K., & Cronin, J. J. (2001). Some New Thoughts on Conceptualizing Perceived Service Quality: A Hierarchical Approach. Journal of Marketing, 65(3), 34–49. https://doi.org/10.1509/jmkg.65.3.34.18334 Burja, L., Mazare, D., & Tanase, M. (2006). Towards a Governmental Platform for Citizen Safety and Crisis Management Using Advanced Mobile Technologies. In 2nd European Mobile Government Conference (Euro mGov 2006) (Issue November). http://www.m4life.org/proceedings/2006/PDF/5_Burja.pdf%5Cnpapers2://publication/ uuid/12D0F1C0-B4DE-4815-BB28-9A953CAE21B8 Byrne, B. M. (2010). Structural Equation Modeling with AMOS. In Structural Equation Modeling (Vol. 22). https://doi.org/10.4324/9781410600219 Byrne, B. M. (2013). Structural equation modeling with AMOS: Basic concepts, applications, and programming, second edition. In Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming, Second Edition. https://doi.org/10.4324/9780203805534 Chae, M., Kim, J., Kim, H., & Ryu, H. (2002). Information Quality for Mobile Internet Services: A Theoretical Model with Empirical Validation. Electronic Markets, 12(1), 38–46. https://doi.org/10.1080/101967802753433254 Chanana, L., Agrawal, R., & Punia, D. K. (2016). Service Quality Parameters for Mobile Government Services in India. Global Business Review, 17(1), 136–146. https://doi.org/10.1177/0972150915610702 Chen, Z., & Dubinsky, A. J. (2003). A Conceptual Model of Perceived Customer Value in ECommerce: A Preliminary Investigation. Psychology and Marketing, 20(4), 323–347. https://doi.org/10.1002/mar.10076 Chen, Z. J., Vogel, D., & Wang, Z. H. (2016). How to satisfy citizens? Using mobile government to reengineer fair government processes. Decision Support Systems, 82, 47–57. https://doi.org/10.1016/j.dss.2015.11.005 Chong, A. Y. L. (2013a). A two-staged SEM-neural network approach for understanding and predicting the determinants of m-commerce adoption. Expert Systems with Applications, 40(4), 1240–1247. https://doi.org/10.1016/j.eswa.2012.08.067 Chong, A. Y. L. (2013b). Predicting m-commerce adoption determinants: A neural network approach. Expert Systems with Applications, 40(2), 523–530. https://doi.org/10.1016/j.eswa.2012.07.068 Colesca, S. E. (2009). Increasing E-Trust: A Solution To Minimize Risk In E-Government Adoption. Journal of Applied Quantitative Methods, 4(1), 1–16. https://doi.org/http://inzeko.ktu.lt/index.php/EE/article/viewFile/11637/6319 Cronin, J. J., Brady, M. K., & Hult, G. T. M. (2000). Assessing the effects of quality, value, 179 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 and customer satisfaction on consumer behavioral intentions in service environments. Journal of Retailing, 76(2), 193–218. https://doi.org/10.1016/S0022-4359(00)00028-2 Dai, H., Luo, X., Liao, Q., & Cao, M. (2015). Explaining consumer satisfaction of services: The role of innovativeness and emotion in an electronic mediated environment. Decision Support Systems, 70, 97–106. https://doi.org/10.1016/j.dss.2014.12.003 Das, A., Singh, H., & Joseph, D. (2017). A longitudinal study of e-government maturity. Information and Management, 54(4), 415–426. https://doi.org/10.1016/j.im.2016.09.006 Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User Acceptance of Computer Technology: A Comparison of Two Theoretical Models. Management Science, 35(8), 982–1003. https://doi.org/10.1287/mnsc.35.8.982 De Lima, C. C. B. (2017). M-government: Limits and possibilities for the use of mobile apps as a citizenship tool in Brazil. ACM International Conference Proceeding Series, Part F1280, 245–252. https://doi.org/10.1145/3047273.3047383 Djamal, E. C., & Renaldi, F. (2018). Developing citizen centric integrated systems for local community: An approach towards M-Government application. 2017 International Conference on ICT for Smart Society, ICISS 2017, 2018-Janua, 1–4. https://doi.org/10.1109/ICTSS.2017.8288887 Dutra, D., & Soares, D. (2018). Decision-making factors for the provision of mobile government solutions. ACM International Conference Proceeding Series, 706–708. https://doi.org/10.1145/3209415.3209500 Dwivedi, Y. K., Rana, N. P., Janssen, M., Lal, B., Williams, M. D., & Clement, M. (2017). An empirical validation of a unified model of electronic government adoption (UMEGA). Government Information Quarterly, 34(2), 211–230. https://doi.org/10.1016/j.giq.2017.03.001 Dwivedi, Y. K., Sahu, G. P., Baabdullah, A. M. A., & Rana, N. P. (2018). Citizens’ awareness, acceptance and use of mobile government services in India: An exploratory research. ACM International Conference Proceeding Series, 236–239. https://doi.org/10.1145/3209415.3209491 Ebrahim, Z., & Irani, Z. (2005). E-government adoption: Architecture and barriers. Business Process Management Journal, 11(5), 589–611. https://doi.org/10.1108/14637150510619902 Fornell, C., & Larcker, D. F. (1981). Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. Journal of Marketing Research, 18(1), 39. https://doi.org/10.2307/3151312 Gaviria-Marin, M., Merigo, J. M., & Popa, S. (2018). Twenty years of the Journal of Knowledge Management: a bibliometric analysis. Journal of Knowledge Management, 22(8), 1655–1687. https://doi.org/10.1108/JKM-10-2017-0497 Gefen, D. (2002). Customer Loyalty in E-Commerce. Journal of the Association for Information Systems, 3(1), 27–53. https://doi.org/10.17705/1jais.00022 Giovanis, A. N., Binioris, S., & Polychronopoulos, G. (2012). An extension of TAM model 180 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 with IDT and security/privacy risk in the adoption of internet banking services in Greece. EuroMed Journal of Business, 7(1), 24–53. https://doi.org/10.1108/14502191211225365 Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate Data Analysis. Prentice Hall, 816. https://doi.org/10.1016/j.ijpharm.2011.02.019 Hameed, A. G., Shukur, B. S., Al-Khafaji, N. J., & Al-Farhan, H. A. (2014). Utilizing mobile applications as a channel of interaction between the citizens and the government: Design M-police in Iraq. 2014 3rd International Conference on E-Technologies and Networks for Development, ICeND 2014, 1–7. https://doi.org/10.1109/ICeND.2014.6991182 Hao, Y. (2016). Exploring undergraduates’ perspectives and flipped learning readiness in their flipped classrooms. Computers in Human Behavior, 59(January), 82–92. https://doi.org/10.1016/j.chb.2016.01.032 Heeks, R., & Bailur, S. (2007). Analyzing e-government research: Perspectives, philosophies, theories, methods, and practice. Government Information Quarterly, 24(2), 243–265. https://doi.org/10.1016/j.giq.2006.06.005 Henseler, J., Ringle, C. M., & Sarstedt, M. (2014). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8 Hole, K. J. (2016). Building Trust in E-Government Services. Computer, 49(1), 66–74. https://doi.org/10.1109/MC.2016.4 Hsiao, K.-L., Lin, J. C.-C., Wang, X.-Y., Lu, H.-P., & Yu, H. (2010). Antecedents and consequences of trust in online product recommendations: An empirical study in social shopping. Online Information Review, 34(6), 935–953. https://doi.org/10.1108/14684521011099414 Iskender, G., & Özkan, S. (2013). E-government transformation success: An assessment methodology and the preliminary results. Transforming Government: People, Process and Policy, 7(3), 364–392. https://doi.org/10.1108/TG-09-2012-0008 Ismagilova, E., Slade, E. L., Rana, N. P., & Dwivedi, Y. K. (2019). The Effect of Electronic Word of Mouth Communications on Intention to Buy: A Meta-Analysis. Information Systems Frontiers, May, 1–4. https://doi.org/10.1007/s10796-019-09924-y Kline, R. B. (2005). Principles and practice of structural equation modeling. In Methodology in the Social Sciences (Vol. 156, Issue 3958). The Guilford Press. https://doi.org/10.1038/156278a0 Kumar, M., & Sinha, O. P. (2007). M-Government - Mobile Technology for e-Government. International Conference on E-Government, 294–301. https://doi.org/10.1504/EG.2008.016124 Kurfalı, M., Arifoğlu, A., Tokdemir, G., & Paçin, Y. (2017). Adoption of e-government services in Turkey. Computers in Human Behavior, 66, 168–178. https://doi.org/10.1016/j.chb.2016.09.041 Kushchu, I., & Kuscu, M. H. (2004). From E-government to M-government: Facing the 181 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Inevitable. Proceedings of the 3rd European Conference on EGovernment, 1–13. http://citeseerx.ist.psu.edu Lallmahomed, M. Z. I., Lallmahomed, N., & Lallmahomed, G. M. (2017). Factors influencing the adoption of e-Government services in Mauritius. Telematics and Informatics, 34(4), 57–72. https://doi.org/10.1016/j.tele.2017.01.003 Larosiliere, G. D., & Carter, L. D. (2016). Using a fit-viability approach to explore the determinants of e-government maturity. Journal of Computer Information Systems, 56(4), 271–279. https://doi.org/10.1080/08874417.2016.1163995 Liébana-Cabanillas, F., & Alonso-Dos-Santos, M. (2017). Factors that determine the adoption of Facebook commerce: The moderating effect of age. Journal of Engineering and Technology Management, 44(2016), 1–18. https://doi.org/10.1016/j.jengtecman.2017.03.001 Lin, F., Fofanah, S. S., & Liang, D. (2011). Assessing citizen adoption of e-Government initiatives in Gambia: A validation of the technology acceptance model in information systems success. Government Information Quarterly, 28(2), 271–279. https://doi.org/10.1016/j.giq.2010.09.004 Liu, Y., Huang, X., An, A., & Yu, X. (2008). Modeling and predicting the helpfulness of online reviews. Proceedings - IEEE International Conference on Data Mining, ICDM, 443–452. https://doi.org/10.1109/ICDM.2008.94 Markovic, S., & Raspor, S. (2010). Measuring Perceived Service Quality Using SERVQUAL: A Case Study of the Croatian Hotel Industry. Management, 5(3), 195– 209. Moon, M. J., & Norris, D. F. (2005). Does managerial orientation matter? the adoption of reinventing government and e-government at the municipal level. Information Systems Journal, 15(1), 43–60. https://doi.org/10.1111/j.1365-2575.2005.00185.x Nathan, A. J., & Scobell, A. (2012). How China sees America. Foreign Affairs, 91(5), 425– 478. https://doi.org/10.1017/CBO9781107415324.004 Ndou, V. D. (2004). E – Government for Developing Countries: Opportunities and Challenges. EJISDC The Electronic Journal on Information Systems in Developing Countries, 18(1), 1–24. https://doi.org/. Onashoga, A., Ogunjobi, A., Ibharalu, T., & Lawal, O. (2016). A secure framework for SMSbased service delivery in m-government using a multicast encryption scheme. African Journal of Science, Technology, Innovation and Development, 8(3), 247–255. https://doi.org/10.1080/20421338.2016.1156837 Ostrom, A. L., Parasuraman, A., Bowen, D. E., Patrício, L., & Voss, C. A. (2015). Service Research Priorities in a Rapidly Changing Context. Journal of Service Research, 18(2), 127–159. https://doi.org/10.1177/1094670515576315 Papadomichelaki, X., & Mentzas, G. (2012). E-GovQual: A multiple-item scale for assessing e-government service quality. Government Information Quarterly, 29(1), 98–109. https://doi.org/10.1016/j.giq.2011.08.011 Parasuraman, a, Zeithaml, V. a, & Berry, L. L. (1988). SERQUAL: A Multiple-Item scale 182 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 for Measuring Consumer Perceptions of Service Quality. Journal of Retailing, 64(1), 12–40. https://doi.org/10.1016/S0148-2963(99)00084-3 Rana, N. P., Dwivedi, Y. K., Lal, B., Williams, M. D., & Clement, M. (2017). Citizens’ adoption of an electronic government system: towards a unified view. Information Systems Frontiers, 19(3), 549–568. https://doi.org/10.1007/s10796-015-9613-y Rana, N. P., Dwivedi, Y. K., Williams, M. D., & Weerakkody, V. (2014). Investigating success of an e-government initiative: Validation of an integrated IS success model. Information Systems Frontiers, 17(1), 127–142. https://doi.org/10.1007/s10796-0149504-7 Reichheld, F. F., & Schefter, P. (2000). E-Loyalty: Your secret weapon on the web. Harvard Business Review, 78(4), 105–113. https://doi.org/10.1007/PL00012187 Saghafi, F., Zarei, B., & Fadaei, M. (2016). The conceptual model for the effect of technology on the political and social aspects of Iran’s e-government. Quality and Quantity, 50(4), 1765–1780. https://doi.org/10.1007/s11135-015-0234-9 Schaupp, L. C., & Belanger, F. (2005). A Conjoint Analysis of Online Consumer Satisfaction. Journal of Electronic Commerce Research, 6(2), 95–111. Shahbaz, M., Gao, C., Zhai, L., Shahzad, F., Abbas, A., & Zahid, R. (2020). Investigating the impact of big data analytics on perceived sales performance: The mediating role of customer relationship management capabilities. Complexity, 2020, 1–17. https://doi.org/10.1155/2020/5186870 Shahbaz, M., Gao, C., Zhai, L., Shahzad, F., & Arshad, M. R. (2020). Moderating Effects of Gender and Resistance to Change on the Adoption of Big Data Analytics in Healthcare. Complexity, 2020, 1–13. https://doi.org/10.1155/2020/2173765 Shahzad, F., Xiu, G., Khan, M. A. S., & Shahbaz, M. (2020). Predicting the adoption of a mobile government security response system from the user’s perspective: An application of the artificial neural network approach. Technology in Society, 62, 101278. https://doi.org/10.1016/j.techsoc.2020.101278 Shahzad, F., Xiu, G. Y., Khan, I., Shahbaz, M., Riaz, M. U., & Abbas, A. (2020). The moderating role of intrinsic motivation in cloud computing adoption in online education in a developing country: a structural equation model. Asia Pacific Education Review, 21(1), 121–141. https://doi.org/10.1007/s12564-019-09611-2 Shahzad, F., Xiu, G. Y., Khan, I., & Wang, J. (2019). m-Government Security Response System: Predicting Citizens’ Adoption Behavior. International Journal of HumanComputer Interaction, 35(10), 899–915. https://doi.org/10.1080/10447318.2018.1516844 Shan, S., Liu, M., & Xu, X. (2017). Analysis of the key influencing factors of haze information dissemination behavior and motivation in WeChat. Information Discovery and Delivery, 45(1), 21–29. https://doi.org/10.1108/IDD-09-2016-0029 Shanab, E. A., & Haider, S. (2015). Major factors influencing the adoption of m-government in Jordan. Electronic Government, an International Journal, 11(4), 223. https://doi.org/10.1504/EG.2015.071394 183 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 Shareef, M. A., Archer, N., & Dwivedi, Y. K. (2012). Examining adoption behavior of mobile government. Journal of Computer Information Systems, 53(2), 39–48. https://doi.org/10.1080/08874417.2012.11645613 Shareef, M. A., Archer, N., & Dwivedi, Y. K. (2015). An empirical investigation of electronic government service quality: from the demand-side stakeholder perspective. Total Quality Management and Business Excellence, 26(3–4), 339–354. https://doi.org/10.1080/14783363.2013.832477 Shareef, M. A., Dwivedi, Y. K., Stamati, T., & Williams, M. D. (2014). SQ mGov: A Comprehensive Service-Quality Paradigm for Mobile Government. Information Systems Management, 31(2), 126–142. https://doi.org/10.1080/10580530.2014.890432 Shareef, M. A., Kumar, V., Dwivedi, Y. K., & Kumar, U. (2016). Service delivery through mobile-government (mGov): Driving factors and cultural impacts. Information Systems Frontiers, 18(2), 315–332. https://doi.org/10.1007/s10796-014-9533-2 Shareef, M. A., Kumar, V., Kumar, U., & Dwivedi, Y. K. (2011). E-Government Adoption Model (GAM): Differing service maturity levels. Government Information Quarterly, 28(1), 17–35. https://doi.org/10.1016/j.giq.2010.05.006 Sharma, R., & Jayasimha, K. R. (2016). Assessment of e-government service quality: An emerging market perspective. The Marketing Review, 16(1), 62–77. https://doi.org/10.1362/146934716x14636478977313 Sharma, S. K. (2015). Adoption of e-government services: The role of service quality dimensions and demographic variables. Transforming Government: People, Process and Policy, 9(2), 207–222. https://doi.org/10.1108/TG-10-2014-0046 Sharma, S. K., Gaur, A., Saddikuti, V., & Rastogi, A. (2017). Structural equation model (SEM)-neural network (NN) model for predicting quality determinants of e-learning management systems. Behaviour and Information Technology, 36(10), 1053–1066. https://doi.org/10.1080/0144929X.2017.1340973 Sheng, H., & Trimi, S. (2008). M-government: Technologies, applications and challenges. Electronic Government, 5(1), 1–18. https://doi.org/10.1504/EG.2008.016124 Tadeusiewicz, R. (1995). Neural networks: A comprehensive foundation. In Control Engineering Practice (Vol. 3, Issue 5). Englewood Cliffs, NJ: Prentic Hall. https://doi.org/10.1016/0967-0661(95)90080-2 Tan, F. B., & Chou, J. P. C. (2008). The relationship between mobile service quality, perceived technology compatibility, and users’ perceived playfulness in the context of mobile information and entertainment services. International Journal of HumanComputer Interaction, 24(7), 649–671. https://doi.org/10.1080/10447310802335581 Venkatesh, V., Sykes, T. A., & Venkatraman, S. (2014). Understanding e-Government portal use in rural India: Role of demographic and personality characteristics. Information Systems Journal, 24(3), 249–269. https://doi.org/10.1111/isj.12008 Venkatesh, V., Thong, J. Y. L., Chan, F. K. Y., & Hu, P. J. H. (2016). Managing citizens’ uncertainty in e-government services: The mediating and moderating roles of transparency and trust. Information Systems Research, 27(1), 87–111. 184 International Journal of Innovation, Creativity and Change. www.ijicc.net Volume 15, Issue 9, 2021 https://doi.org/10.1287/isre.2015.0612 Wang, R. T., & Lin, C. P. (2012). Understanding innovation performance and its antecedents: A socio-cognitive model. Journal of Engineering and Technology Management, 29(2), 210–225. https://doi.org/10.1016/j.jengtecman.2012.01.001 Weerakkody, V., El-Haddadeh, R., Al-Sobhi, F., Shareef, M. A., & Dwivedi, Y. K. (2013). Examining the influence of intermediaries in facilitating e-government adoption: An empirical investigation. International Journal of Information Management, 33(5), 716– 725. https://doi.org/10.1016/j.ijinfomgt.2013.05.001 Welch, E. W., Hinnant, C. C., & Moon, M. J. (2005). Linking citizen satisfaction with egovernment and trust in government. In Journal of Public Administration Research and Theory (Vol. 15, Issue 3, pp. 371–391). https://doi.org/10.1093/jopart/mui021 Wulandari, S. S. (2019). Employee commitment and service performance. Human Systems Management, 37(4), 381–386. https://doi.org/10.3233/HSM-17122 Yang, S. C. (2012). Mobile applications and 4G wireless networks: A framework for analysis. In Campus-Wide Information Systems (Vol. 29, Issue 5, pp. 344–357). https://doi.org/10.1108/10650741211275107 Zeithaml, V. A., Parasuraman, A., & Malhotra, A. (2002). Service quality delivery through web sites: A critical review of extant knowledge. In Journal of the Academy of Marketing Science (Vol. 30, Issue 4, pp. 362–375). https://doi.org/10.1177/009207002236911 Zhou, T., Lu, Y., & Wang, B. (2010). Integrating TTF and UTAUT to explain mobile banking user adoption. Computers in Human Behavior, 26(4), 760–767. https://doi.org/10.1016/j.chb.2010.01.013 185