ScienceDirect Available online at
Download 0.63 Mb. Pdf ko'rish
|
Rwanda Electron Government
- Bu sahifa navigatsiya:
- Trust E-Government
- Trust Propensity
- Trust Internet
- 4. Results and discussion
- April
|
respectively, these values are bigger than their correlation values of their separate sections. In this way, the outcome demonstrates that the Discriminant validity is settled as appeared in the above Table 3.
Leonidas Nzaramyimana et al. / Procedia Computer Science 161 (2019) 350–358 355
Author name / Procedia Computer Science 00 (2019) 000–000 5 Variables and measurements items Code People who use e-government to obtain services have more prestige than those who do not SI4
I expect that e-government services will not take advantages of me TEGOV1
I believe that e-government services are trustworthy TEGOV2
I believe that e-government services will not act in a way that harms me TEGOV3
I trust e-government services TEGOV4
Trust Propensity
I generally do trust other people. TPROP1
I generally have faith in humanity. TPROP2
I feel that people are generally reliable TPROP3
I generally trust other people unless they give me reason not to. TPROP4
Trust Internet
The internet has enough safeguards to make me feel comfortable using it to interact with the e-government services. TI1
I feel assured that legal and technological structures adequately protect me from problems on the Internet. TI2 I feel secure sending sensitive information across the internet. TI3 In general, the internet is now a robust and safe environment in which to transact services with the e-government. TI4
3.3. Measuring of structure and measurement of model Inner model and outer model were estimated to test the fitness of the model utilizing tests convergent and Discriminant validity through PLS-SEM 3 Students Version (See Fig.1). PLS is a powerful tool to quantify the structural equation modelling. Especially with constrained members and slanted information appropriation [21]. From the reality above [21] along these lines in estimation model the measures were dependability, Discriminant validity, and discriminant legitimacy. Additionally, the basic model the measures were R 2 (explained variance), f 2 (impact size) and Q 2
Fig. 1. Proposed Model of this Research. 6 Author name / Procedia Computer Science 00 (2019) 000–000 4. Results and discussion For this situation our investigation is reflective where validity and reliability ought to be analyzed [22].in reliability the test is indicator reliability which is obtained by squaring every outer loadings respectively, the preferred value must be 0.7 or higher, for this situation outer loading must be greater than 0.708 [22]. For internal consistency reliability, Cronbach's alpha is utilized to gauge regularly in sociology looks into, yet it will in general give in preservationist estimations in PLS-SEM. Earlier writing has recommended the utilization of "Composite reliability" as substitution, the estimation of composite dependability ought to be 0.7 or higher. In Discriminant validity, the estimation of convergent validity (AVE) ought to be 0.5 or higher [23]. At long last, in discriminant validity, the test is for AVE numbers and latent variables, considering [24] the square root of AVE of each latent variable ought to be than the relationship among the idle factors, that is known as Fornel-lacker. Table 2 demonstrates reflective of the outer model of our proposed model using Smart PLS 3 students’ version and Table 3 demonstrates the Fornell-larcker basis investigation for checking discriminant validity. Once Smart PLS was running, the principal results were not substantial for some constructs and items, where there have been another run and after that the information turned out to be clearly reliable and valid outcomes. From the procedure, a portion of the item (TEGOV1=0.369 and SI2=0.634) from the latent variables TRUST E_GOVERNMENT and SOCIAL INFLUENCE were dropped out, this is because of the paradigm that the indicator reliability quality was under 0.4, while the remainder of the show were reliable in light of the fact that the indicator reliability is more than the prescribed estimation of 0.4, some are close and other are over the favoured dimension of 0.7 [25]. Table 2. A reflective measure of the outer model. Constructs Variables Loadings IR AVE CR Results BEHAVIOURAL INTENTION BI_1
0.884 0.781
0.766 0.929 Reliable
BI_2
0.904 0.817 Reliable
BI_3 0.904 0.817 Reliable
BI_4 0.804 0.646 Reliable SOCIAL INFLUENCE SI_1 0.884
0.781 0.680 0.864 Reliable
SI_3 0.793 0.629 Reliable
SI_4 0.793 0.629 Reliable TRUST PROPENSITY TPROP_1 0.912
0.832 0.805 0.943 Reliable
TPROP_2 0.933 0.870 Reliable
TPROP_3 0.907 0.823 Reliable
TPROP_4 0.834 0.696 Reliable TRUST E_GOVERNMENT TEGOV_2
0.901 0.812
0.820 0.932 Reliable
TEGOV_3 0.895 0.801 Reliable
TEGOV_4 0.920 0.846 Reliable TRUST INTERNET TI_1
0.877 0.769
0.699 0.902 Reliable
TI_2 0.869 0.755 Reliable
TI_3 0.790 0.624 Reliable
TI_4 0.803 0.645
Reliable
Considering the Fornell-larcker rule in Table 3, the discriminant validity measure recommended as the square root of AVE in each construct can be utilized to set up discriminant if the value is bigger the other correlation diagonally. For instance, from the above table the construct AVE of BEHAVIOURAL INTENTION, SOCIAL INFLUENCE, TRUST E_GOVERNMENT, TRUST INTERNET and the TRUST PROPENSITY found to have square root of 0.875, 0.824, 0.906, 0.836, and 0.897 respectively, these values are bigger than their correlation values of their separate sections. In this way, the outcome demonstrates that the Discriminant validity is settled as appeared in the above Table 3.
356 Leonidas Nzaramyimana et al. / Procedia Computer Science 161 (2019) 350–358
Author name / Procedia Computer Science 00 (2019) 000–000 7
Table 3. Discriminant Validity test (the Fornell-larcker).
BI SI TEGOV TI TPROP BI
0.875
SI
0.537 0.824
TEGOV 0.793 0.530 0.906
TI 0.628 0.514 0605 0.836
TPROP 0.558 0.399 0.605 0.513 0.897
For the basic model, the measure is R 2 (explained variance), f 2 (effect size) and Q 2 (Predictive relevance) [26][21]. The scope of R 2 is from 0 to 1 which abnormal states demonstrating a more elevated amount of prediction accuracy, the estimation of 0.75, 0.50 and 0.25 can be depicted as substantial, moderate and weak [26]. The current correlation between the dependent variable and independent variable shown by various R and its statistical significance at p<0.05. table below demonstrates a summary of coefficient of determination of our model. The value R 2 in our proposed model is 0.674 implying that the model is very fitted, and it covers over half. Table below demonstrates the coefficient of determinant with R 2 value together with R 2 adjusted. The effect size f 2 enables us to watch the impact of each exogenous construct on the endogenous construct, estimation of 0.02, 0.15 and 0.35 represent to small, medium and large impact of the exogenous inactive variable [26]. Table 4. The coefficient of Determination of the proposed model.
R R Square R square Adjusted BEHAVIOURAL INTENTION 0.820 0.674
0.659
In predictive relevance Q 2 the assessment of R 2 value as basis of predictive accuracy we ought to likewise inspect Stone-geisser's Q 2 value, the value is more noteworthy than 0, shows that the model has relevance [22]. The table below demonstrates the summarized results of the inner model for the examined model. The t-statistical was assumed for the significance of model of 97% assurance level, with BEHAVIOURAL INTENTION dependent variables of our model as predicators. Ordinary least relapse (OLS) can be utilized to decipher the individual path coefficient by institutionalizing the beta coefficient. One-unit change of exogenous constructs changes the endogenous construct by the measure of the path coefficient while everything else stays steady path coefficient will be significant at the T- Statistics is bigger than 1.96 at p<0.05 [27]. Table 5. Summary of T-Statistics of path of coefficient (inner model). Construct relation P(β) t-value
p-value Results
SOCIAL INFLUENCE->BEHAVIOURAL INTENTION 0.105 1.347
0.178 Not Support TRUST E_GOVERNMENT->BEHAVIOURAL INTENTION 0.584 8.295 0.000 Support TRUST INTERNET->BEHAVIOURAL INTENTION 0.186 2.090 0.037 Support TRUST PROPENSITY->BEHAVIOURAL INTENTION 0.068 0.859
0.391 Not Support
Nonetheless, all variables have positive path coefficient, this implies that they all have positive effect on BEHAVIOURAL INTENTION as it is shown in Table 5 above. The relationship between the inner model and outer model will be noteworthy when T-Statistics are bigger than 1.96 and p-value are under 0.05 at a significant of level 8 Author name / Procedia Computer Science 00 (2019) 000–000 5%. The relationship (SOCIAL INFLUENCE->BEHAVIOURAL INTENTION, TRUST INTERNET- >BEHAVIORAL INTENTION, TRUST PROPENSITY->BEHAVIOURAL INTENTION) do not support the proposed theory in the applied model, in this way, the hypothesis H1 and H4 are rejected however we accept the hypothesis H2 and H3 which are significant. 5. Conclusion In this investigation, Analysis of factors affecting Behavioural intention to use e-government services in Rwanda. we broke down five elements and we quantified the relationship among them toward the utilization of E-government driving public institutions in Rwanda. The outcomes from this investigation covered from a proposed model which is a construct of five components, two factors have a steady supportive significant on citizens Behavioural intention to use e-government services in Rwanda. This investigation will improve the comprehension of variables that impact the utilization of e-government supporting public institutions in developing countries like Rwanda and can be applied to the other developing nations to test a similar hypothesis, especially with regards to confiding in administration-based innovation. along these lines, future research ought to analyze why citizens would not utilize E-government driven organizations while they are available for use, and thus research ought to use a bigger sample measure and other various elements, techniques and methodologies to discover what truly caused the downturn of administration services which were highly demanded before the implementation of the e-government platform, further tests can be led on different respondent's gatherings. This paper recommends the government of Rwanda to make internet technologies and new government innovation like e-government services trustworthy as it has been investigated that they are main factors that influence to not use them.
Reference [1] Noguera, M. A. P. (2014) “Opportunities and Challenges of e-Governance: A Reality or Science Fiction for the Chinese Government?” Anal. y Pensam. Iberoam. sobre China, m41–17. [2] Djermani, F., A. Shahzad, A. A. Sheikh, J. Mohammed, and E. Alekam. (2016) “Factors Influencing The Intention to Use E-Government Services in Algeria: An Empirical Study.” Stud. Ubb Negot.
[3] Nasri, W., and H. A. Abbas. (2015) “Determinants Influencing Citizens’ Intention to Use E-Gov in the State Of Kuwait: Application of UTAUT.” Int. J. Econ. Commer. Manag.
[4] Fakhoury, R., and B. Aubert. (2015) “Citizenship, Trust, and Behavioural Intentions to Use Public E-Services: The Case of Lebanon.” Int. J. Inf. Manage.
[5] Chatzoglou, P., D. Chatzoudes, and S. Symeonidis. (2015) “Factors Affecting The Intention to Use E-Government Services.” Proc. 2015 Fed. Conf. Comput. Sci. Inf. Syst.
[6] Robinson, S. E., J. W. Stoutenborough, and A. Vedlitz. (2017) “Understanding Trust in Government.” Underst. Trust Gov. 3: 7–15. [7] Twizeyimana, J. D., and A. Andersson. (2019) “The Public Value of E-Government – A Literature Review.” Gov. Inf. Q. 36 (2): 167–178. [8] Goldfinch, S. (2012) “Public Trust in Government, Trust in E-Government, and Use of E-Government.” Encycl. Cyber Behav., February. pp. 987–995. [9] Abdelghaffar, H. A., S. H. Kamel, and P. Duquenoy. (2010) “Studying eGovernment Trust in Developing Nations: A Case of University and Colleges Admissions and Services in Egypt.” Commun. IIMA
[10] Wangpipatwong, S., W. Chutimaskul, and B. Papasratorn. (2008) “Understanding Citizen’s Continuance Intention to Use E-Government Website: A Composite View of Technology Acceptance Model and Computer Self-Efficacy.” Electron. J. e-Government
[11] Sparks, P., I. Ajzen, and T. Hall-box. (2002) “Perceived Behavioral Control, Self-Efficacy, Locus of Control, and the Theory of Planned Behavior.” pp. 665–683. [12] Camilleri, M. A. (2015) “Exploring the Behavioral Intention To Use E-Government Services: Validating the Unified Theory of Acceptance and Use of Technology.” [13] Xie, Q., W. Song, X. Peng, and M. Shabbir. (2017) “Predictors for E-Government Adoption: Integrating TAM, TPB, Trust and Perceived Risk.” Electron. Libr.
[14] Tan, M. (2000) “Factors Influencing the Adoption of Internet Banking.” Jais 1: 1–42. [15] Pavlou, P. A., and M. Fygenson. (2014) “Planned Behavior Qjarteny and Predicting Understanding Electronic An Extension of Commerce Adoption : the Theory of Planned formed.”
[16] Eynon R., and W. H. Dutton. (2007) “Barriers to Networked Governments: Evidence from Europe.” Prometh. (United Kingdom) 25 (3): 225–242. [17] Hole, K. J. (2016) “Building Trust in E-Government Services.” Computer (Long. Beach. Calif).
Leonidas Nzaramyimana et al. / Procedia Computer Science 161 (2019) 350–358 357
Author name / Procedia Computer Science 00 (2019) 000–000 7
Table 3. Discriminant Validity test (the Fornell-larcker).
BI SI TEGOV TI TPROP BI
0.875
SI
0.537 0.824
TEGOV 0.793 0.530 0.906
TI 0.628 0.514 0605 0.836
TPROP 0.558 0.399 0.605 0.513 0.897
For the basic model, the measure is R 2 (explained variance), f 2 (effect size) and Q 2 (Predictive relevance) [26][21]. The scope of R 2 is from 0 to 1 which abnormal states demonstrating a more elevated amount of prediction accuracy, the estimation of 0.75, 0.50 and 0.25 can be depicted as substantial, moderate and weak [26]. The current correlation between the dependent variable and independent variable shown by various R and its statistical significance at p<0.05. table below demonstrates a summary of coefficient of determination of our model. The value R 2 in our proposed model is 0.674 implying that the model is very fitted, and it covers over half. Table below demonstrates the coefficient of determinant with R 2 value together with R 2 adjusted. The effect size f 2 enables us to watch the impact of each exogenous construct on the endogenous construct, estimation of 0.02, 0.15 and 0.35 represent to small, medium and large impact of the exogenous inactive variable [26]. Table 4. The coefficient of Determination of the proposed model.
R R Square R square Adjusted BEHAVIOURAL INTENTION 0.820 0.674
0.659
In predictive relevance Q 2 the assessment of R 2 value as basis of predictive accuracy we ought to likewise inspect Stone-geisser's Q 2 value, the value is more noteworthy than 0, shows that the model has relevance [22]. The table below demonstrates the summarized results of the inner model for the examined model. The t-statistical was assumed for the significance of model of 97% assurance level, with BEHAVIOURAL INTENTION dependent variables of our model as predicators. Ordinary least relapse (OLS) can be utilized to decipher the individual path coefficient by institutionalizing the beta coefficient. One-unit change of exogenous constructs changes the endogenous construct by the measure of the path coefficient while everything else stays steady path coefficient will be significant at the T- Statistics is bigger than 1.96 at p<0.05 [27]. Table 5. Summary of T-Statistics of path of coefficient (inner model). Construct relation P(β) t-value
p-value Results
SOCIAL INFLUENCE->BEHAVIOURAL INTENTION 0.105 1.347
0.178 Not Support TRUST E_GOVERNMENT->BEHAVIOURAL INTENTION 0.584 8.295 0.000 Support TRUST INTERNET->BEHAVIOURAL INTENTION 0.186 2.090 0.037 Support TRUST PROPENSITY->BEHAVIOURAL INTENTION 0.068 0.859
0.391 Not Support
Nonetheless, all variables have positive path coefficient, this implies that they all have positive effect on BEHAVIOURAL INTENTION as it is shown in Table 5 above. The relationship between the inner model and outer model will be noteworthy when T-Statistics are bigger than 1.96 and p-value are under 0.05 at a significant of level 8 Author name / Procedia Computer Science 00 (2019) 000–000 5%. The relationship (SOCIAL INFLUENCE->BEHAVIOURAL INTENTION, TRUST INTERNET- >BEHAVIORAL INTENTION, TRUST PROPENSITY->BEHAVIOURAL INTENTION) do not support the proposed theory in the applied model, in this way, the hypothesis H1 and H4 are rejected however we accept the hypothesis H2 and H3 which are significant. 5. Conclusion In this investigation, Analysis of factors affecting Behavioural intention to use e-government services in Rwanda. we broke down five elements and we quantified the relationship among them toward the utilization of E-government driving public institutions in Rwanda. The outcomes from this investigation covered from a proposed model which is a construct of five components, two factors have a steady supportive significant on citizens Behavioural intention to use e-government services in Rwanda. This investigation will improve the comprehension of variables that impact the utilization of e-government supporting public institutions in developing countries like Rwanda and can be applied to the other developing nations to test a similar hypothesis, especially with regards to confiding in administration-based innovation. along these lines, future research ought to analyze why citizens would not utilize E-government driven organizations while they are available for use, and thus research ought to use a bigger sample measure and other various elements, techniques and methodologies to discover what truly caused the downturn of administration services which were highly demanded before the implementation of the e-government platform, further tests can be led on different respondent's gatherings. This paper recommends the government of Rwanda to make internet technologies and new government innovation like e-government services trustworthy as it has been investigated that they are main factors that influence to not use them.
Reference [1] Noguera, M. A. P. (2014) “Opportunities and Challenges of e-Governance: A Reality or Science Fiction for the Chinese Government?” Anal. y Pensam. Iberoam. sobre China, m41–17. [2] Djermani, F., A. Shahzad, A. A. Sheikh, J. Mohammed, and E. Alekam. (2016) “Factors Influencing The Intention to Use E-Government Services in Algeria: An Empirical Study.” Stud. Ubb Negot.
[3] Nasri, W., and H. A. Abbas. (2015) “Determinants Influencing Citizens’ Intention to Use E-Gov in the State Of Kuwait: Application of UTAUT.” Int. J. Econ. Commer. Manag.
[4] Fakhoury, R., and B. Aubert. (2015) “Citizenship, Trust, and Behavioural Intentions to Use Public E-Services: The Case of Lebanon.” Int. J. Inf. Manage.
[5] Chatzoglou, P., D. Chatzoudes, and S. Symeonidis. (2015) “Factors Affecting The Intention to Use E-Government Services.” Proc. 2015 Fed. Conf. Comput. Sci. Inf. Syst.
[6] Robinson, S. E., J. W. Stoutenborough, and A. Vedlitz. (2017) “Understanding Trust in Government.” Underst. Trust Gov. 3: 7–15. [7] Twizeyimana, J. D., and A. Andersson. (2019) “The Public Value of E-Government – A Literature Review.” Gov. Inf. Q. 36 (2): 167–178. [8] Goldfinch, S. (2012) “Public Trust in Government, Trust in E-Government, and Use of E-Government.” Encycl. Cyber Behav., February. pp. 987–995. [9] Abdelghaffar, H. A., S. H. Kamel, and P. Duquenoy. (2010) “Studying eGovernment Trust in Developing Nations: A Case of University and Colleges Admissions and Services in Egypt.” Commun. IIMA
[10] Wangpipatwong, S., W. Chutimaskul, and B. Papasratorn. (2008) “Understanding Citizen’s Continuance Intention to Use E-Government Website: A Composite View of Technology Acceptance Model and Computer Self-Efficacy.” Electron. J. e-Government
[11] Sparks, P., I. Ajzen, and T. Hall-box. (2002) “Perceived Behavioral Control, Self-Efficacy, Locus of Control, and the Theory of Planned Behavior.” pp. 665–683. [12] Camilleri, M. A. (2015) “Exploring the Behavioral Intention To Use E-Government Services: Validating the Unified Theory of Acceptance and Use of Technology.” [13] Xie, Q., W. Song, X. Peng, and M. Shabbir. (2017) “Predictors for E-Government Adoption: Integrating TAM, TPB, Trust and Perceived Risk.” Electron. Libr.
[14] Tan, M. (2000) “Factors Influencing the Adoption of Internet Banking.” Jais 1: 1–42. [15] Pavlou, P. A., and M. Fygenson. (2014) “Planned Behavior Qjarteny and Predicting Understanding Electronic An Extension of Commerce Adoption : the Theory of Planned formed.”
[16] Eynon R., and W. H. Dutton. (2007) “Barriers to Networked Governments: Evidence from Europe.” Prometh. (United Kingdom) 25 (3): 225–242. [17] Hole, K. J. (2016) “Building Trust in E-Government Services.” Computer (Long. Beach. Calif).
358 Leonidas Nzaramyimana et al. / Procedia Computer Science 161 (2019) 350–358
Author name / Procedia Computer Science 00 (2019) 000–000 9 [18] Zambrano, R. (2008) “E-Governance and Development : Service Delivery to Empower the Poor.” Int. J. Electron. Gov. Res. 4 (2): 1–11. [19] Santamaria-Philco, A., P. Q. Palma, W. Z. Mero, D. Macias-Mendoza, and E. Panchana. (2018) “Trust in Electronic Governments: A Vision of Their Influence Factors.” Iber. Conf. Inf. Syst. Technol. Cist. (
[20] Baldwin, A., S. Shiu, and M. C. Mont. (2002) “Trust Services: A Framework for Service-Based Solutions”, in Proceedings-IEEE Comput. Soc. Int. Comput. Softw. Appl. Conf., no.
[21] Wong, K. K. (2013) “Partial Least Squares Structural Equation Modeling ( PLS-SEM ) Techniques Using SmartPLS.” [22] Gorla, N., T. M. Somers, and B. Wong. (2010) “Journal of Strategic Information Systems Organizational Impact of System Quality, Information Quality, and Service Quality.” J. Strateg. Inf. Syst. 19 (3): 207–228. [23] Bagozzi, R., and Y. Yi. (1988) “On the Evaluation of Structure Equation Models.” January. [24] Ghadi, Ibrahim, Nor Alwi, Kamariah Abu Bakar, and Othman Talib. (2012) “Construct Validity Examination of Critical Thinking Dispositions for Undergraduate Students in University Putra Malaysia.”
[25] Ringle, C. M., and R. R. Sinkovics, (2004) “The Use of Partial Least Squares Path Modeling in International Marketing.” 20 (2009): 277– 319.
[26] Cohen, J. (n.d.) Statistical Power Analysis. [27] Haenlein, M., and A. M. Kaplan. (2004) “A Beginner ’ s Guide to Partial Least Squares Analysis.” 3 (4): 283–297.
Download 0.63 Mb. Do'stlaringiz bilan baham: 
|