Public Value Creation Through Electronic Human Resource Management In Public Sector: A Mixed Methods Approach

. This study sought to evaluate the link between public value creation and e-HRM. The study draws upon public value theory to explore the role of e-HRM in value creation of public organizations. The study was conducted in Kenya in the County Government of Nyeri. A mixed methods approach was adopted, which combines quantitative surveys with qualitative interviews. The quantitative research approach involved a cross-sectional survey design. The survey respondents were 209 employees of Nyeri County Government. Data was analyzed by means of structural equation modelling (SEM). Qualitative approach involved conducting interviews. Thematization was used to analyze qualitative data. The study results show a significant relationship between quality public services, achievable desirable outcomes and development of trust and public value creation. In line with study findings, the study recommends that public sector organizations should optimize the use of e-HRM in their operations to create public value. This study is noteworthy for being one of the pioneering works to explore public value creation through e-HRM using a mixed-methodology approach.


INTRODUCTION
The operational and functional procedures within organizations have undergone significant transformation in recent decades due to ongoing efforts in digitalization, and the human resources (HR) department is no exception.Processes such as recruitment, training, and compensation, among others, have been reshaped with adoption of information and communication technology (Alomari, 2023).The digitization of HR functions and the implementation of automation have led to the emergence of electronic human resource management (e-HRM).Today, the adoption of e-HRM has become widespread across various organizations, both profit and nonprofit, with an anticipated value creation.As a result, there has been a growing academic interest in e-HRM, aiming to explore its potential for generating public value (Iqbal et al., 2019;Parry, 2011).Value creation is not a new phenomenon and has been given various views by scholars regarding its sources and targets (Lepak et al., 2007).Specifically, public sector has been known to differ from private sector in terms of formalized value creation and the ways they use ICTs (Sterrenberg & Decosta, 2023).
Scholars and professionals contend that the objectives of the public sector diverge from those of the private sector, emphasizing that governmental efforts should prioritize the generation of public value (Chantillon et al., 2020).Public value transcends economics in government to account for political and social objectives such as social inclusion, openness, equity, community well-being, stewardship and accountability (Cordella & Bonina, 2012;Moore, 1995).Cordella & Bonina, (2012) assert that a more effective way to handle the intricate socio-political effects of ICT adoption in the public sector is through the concept of public value.Despite the efforts and initiatives put in place to integrate ICT with HRM in the public sector, little is known on how the integration has boosted the provision of efficient and effective services and the value it has created to the organizations.In other words, the justification of investments of IT in HRM in the public sector cannot be easily ascertained.Though previous studies have identified the importance of e-HRM practices, the focus of those studies has not been on public value creation in public sector.Consequently, researchers are recognizing a need for more in-depth studies on ICT enabled government initiatives on public value realization (AbdulKareem et al., 2024).Therefore, purpose of this study is to assess the link between public value creation through e-HRM.http://ijstm.inarah.co.id 567

Study Objectives
The objectives of the study were to: 1. Determine the effect of e-HRM on the quality of public services within County Governments in Kenya.2. Evaluate the role of e-HRM in facilitating the achievement of desirable outcomes in County Governments in Kenya.
3. Examine the relationship between e-HRM implementation and the levels of trust in County Governments in Kenya.

II.
RELATED LITERATURE AND RESEARCH FRAMEWORK 2.1 Electronic Human Resource Management Although the concept of e-HRM has been widely used in both organizational and academic scenarios, a consensus on the definition of the concept is yet to be reached.(Sareen & Subramanian, 2012) pointed out the challenge in formulating a concise definition for e-HRM, noting that the concept relies on utilizing information technology to handle human resources data.For example, Ruël, Bondarouk, and Looise, ( 2004) described e-HRM as "the implementation of HRM strategies, policies, and practices within organizations by actively utilizing web-based technologies" (p.368).This definition emphasizes the practical application of e-HRM and underscores the concept that it is a method of achieving objectives through technology.Essentially, e-HRM involves replacing conventional HRM practices with information technology methods to carry out HR tasks.Strohmeier,(2007) broadened the perspective to underscore the concepts of individuals and interaction, asserting that e-HRM involves the planning, execution, and utilization of information technology to facilitate networking and provide support for at least two individual or collective actors engaged in collaboratively performing HR activities.
Building upon their previous work, Bondarouk and Ruël ( 2009) define e-HRM as "an umbrella term covering all possible integration mechanisms and contents between human resource management (HRM) and information technologies aiming at creating value within and across organizations for targeted employees and management".While there are slight variations among these definitions, collectively, they significantly contribute to the conceptualization of e-HRM.Key elements of e-HRM can be derived from the provided definitions.Firstly, HR technology plays a crucial role in realizing organizational objectives through HR functions.Additionally, as its objectives are oriented towards both employees and organizations, e-HRM should embody interaction among all the relevant stakeholders.Lastly, e-HRM operates in a multilayered and multi-contextual manner, resulting in impacts that extend both within and across organizations.

E-HRM and Public Value Creation
Public value refers to the anticipated outcomes associated with the provision of government public services, as perceived by stakeholders such as citizens, policymakers, civil servants, and taxpayers (Moore, 1995;Sterrenberg & Decosta, 2023).Understanding this concept is crucial, especially in the context of technology utilization, as the perceived value derived from its use significantly impacts its acceptance and adoption (Gerli et al., 2021; Luna-Reyes & Zhang, 2023).In the pursuit of "public value," governments address strategic objectives that extend beyond economic gains to encompass political and social goals (Cordella & Bonina, 2012).These include enhancing efficiency in public services, ensuring equal treatment of constituents, promoting social inclusion, fostering openness, supporting community regeneration, prioritizing community well-being, and upholding principles of stewardship and accountability (Luna-Reyes & Zhang, 2023; Moore, 2014; Pooe & Munyanyi, 2022).Public value is recognized as a fundamental basis for essential elements, benchmarking, and analytical frameworks used to assess public services and evaluate the impact of technologies on the production and delivery of services (Gerli et al., 2021;Sterrenberg & Decosta, 2023;Sufna & Fernando, 2020) .
Values can be assessed and quantified on the outcome of digital services to facilitate comparisons and priorities among different provisions; for example, to cluster values and link them to operational characteristics of government websites.In line with Porter's (1985) ideas, organizations achieve value creation when they innovate their processes and adopt novel methods.In the public sector, e-HRM introduces a fresh perspective on HR management, altering the nature of HRM procedures, practices, and http://ijstm.inarah.co.id 568 policies through the effective use of information technology.The successful implementation of this technology may enable organizations to attain benefits or gain competitive advantages through this innovative approach.Several research has been conducted on value creation and e-HRM (Bondarouk & Ruel, 2013;Iqbal et al., 2019;Pyszka, 2018).The expanding body of literature focusing on e-HRM, and value creation has explored various overarching objectives associated with the introduction of e-HRM, encompassing cost and efficiency savings, strategic goals, and enhancements in client services (Iqbal et al., 2019).Its overarching goal is to generate value within and across organizations for targeted employees and management (Bondarouk & Ruël, 2009).

Research Framework
The current research builds on the theoretical groundings of Public Value Theory (PVT).Public Value Theory provides governments with a comprehensive framework for evaluating their actions, placing citizens at the center, promoting transparency, and striving for effectiveness and efficiency in the delivery of public services (Moore, 1995).According to Kearns (2004), the tenet of public value encompasses three dimensions.The first dimension revolves around the public value generated through the delivery of highquality standard public services, influenced by factors like availability, satisfaction, perceived relevance of services, and cost.The second facet of public value pertains to achieving specific quality of life outcomes, such as improved healthcare, poverty reduction, and decreased environmental pollution.The final component of public value involves the crucial concept of trust in government and its public sector organizations, playing a vital role in shaping public participation and approval of government actions.This study delves into the relationship between public value creation through e-HRM guided by a conceptual model based on public value theory and a synthesis by Kearns' public value tenet in the existing literature.The conceptual framework is depicted in Figure 1

III. METHODS
This study is based on a pragmatic worldview and employs mixed methods approach (Creswell, 2013).This combined approach facilitates a thorough analysis of the data, allowing for a deeper understanding of participants' perspectives.Through qualitative analysis, quantitative results may be confirmed, expanded upon, or potentially challenged as postulated by Government of Nyeri.The study utilized structural equation modeling (SEM) which is a contemporary statistical method for analyzing and estimating relationships between factors and constructs (Wang & Rhemtulla, 2021).These involved two steps of feasibility tests of the measurement and structural models as stipulated in Anderson and Gerbing' (1988).

Evaluation of Measurement Model
The initial measurement model consists of public value creation that comprises of quality of public service (QPS), achievable desirable outcomes (ADO) and development of trust (DOT).In the initial model, 10 indicator variables (QPS1-QPS10) were postulated to load on the first order construct QPS.Another nine indicator variables (ADO1 to ADO9) were loaded on construct ADO, and 6 indicator variables (DOT) were loaded on DOT1 to DOT6).Table 1  Privacy is assured when using e-HRM systems DOT2 Information provided through e-HRM is not wrongly used.

DOT3
County Government protects information in custody of e-HRM systems DOT4 Information disseminated through e-HRM is credible DOT5 e-HRM has resulted to new ways of working and delivery of HR services DOT6 e-HRM facilitates the organization's orientations in changing organizational culture The measurement models used reflective constructs and reflective indicator variables, which means that the latent constructs are reflected, rather than caused, by the observed variables.For example, the latent construct QPS is reflected by the observed variables QPS1 to QPS10.This means that QPS is not causing employee to have perceptions about public value creation, rather, employee' perceptions of these things are reflecting their overall perception of QPS.In order to derive the best indicators of the latent variable prior to testing of the structural model, a congeneric analysis was conducted to validate measurement model through Confirmatory Factor Analysis (CFA).A congeneric analysis was conducted to test the validity of elements for measuring the construct (Hair et al., 2010).Additionally, in assessing reflective indicators, two critical aspects of the measurement model must be considered: convergent and discriminant validity (Gefen et al., 2000).Convergent validity is evaluated through indicator reliability, composite reliability, and average variance extracted (Gau, 2010).The study used goodness of fit indices to assess how well the CFA model matched the data.To assess how well the model fit the data, the study calculated three goodness-of-fit indices: the chi-squared/degrees of freedom ratio, the comparative fit index, and the root mean square error of approximation.A good model fit is achieved when the chi-squared/degrees of freedom ratio is below 3 (Malkanthie, 2015).http://ijstm.inarah.co.id 570 In this study, the ratio was 2.923, with a Bollen-Stine probability (p) value of 0.002 indicating that the model did not fit the data.The RMSEA value of 0.96 and CFI value of 0.866 for the measurement model indicate that the model did not fit the data well.The RMSEA value was higher than recommended value of 0.05 and the CFI value lower than recommended value.The parsimony fit index also confirms that the model was not a good fit.This means that the model needed to be modified to improve its fit to the data.For the first-order factors shown, three one-factor congeneric measurement models were created as the initial step in the model modification process.The one-factor models were reexamined using the cut-off values for the GOF measures, standardized factor loadings (SFLs), standardized residuals (SRs), and modification indices (Kline, 2010).The model re-specification method in this study employed cut-off values of SFLs smaller than 0.5 and SRs bigger than | ± 2.5 ≈ ± 2.58| (Hair et al., 2010).The outcomes of congeneric measurement models with new specifications are displayed in Table 2.If the SR value exceeds | ± 4.0|, it may suggest the presence of an unacceptably high degree of errors, which could result in the item being removed from the measurement model.The Goodness-of-Fit (GOF) measures of the revised measurement model indicates that the model was valid.The chi-square (χ2) divided by degrees of freedom (df) value was 1.234, with a Bollen-Stine p-value of 0.599.This means that the model fits the data well.Additionally, the Root Mean Square Error of Approximation (RMSEA) value was 0.034, with a PCLOSE value of 0.947 and Comparative Fit Index (CFI) value was 0.956 which indicates a good fit.

Construct Validity
Construct validity is a measure of how well a construct is being measured by the items in a measurement model.It is assessed through two types of validity: convergent validity and discriminant validity.Convergent validity can be assessed by (a) the significance of the factor loadings of all items, (b) the average variance extracted (AVE), and (c) the reliability of constructs (Hair et al., 2010).SFL should, as a general rule, be at least 0.5 or greater for every observed item.According to Hair et al. (2010), convergent validity can be achieved for each latent factor with an AVE of 0.5 or above (Canatay et al., 2022).The latent factor reliability can be estimated with the coefficient Cronbach alpha of 0.70 (Hair et al., 2010).Table 3 indicates the model passed construct validity.Constructs that passed convergent validity test were tested for discriminant validity.Table 5 shows that all of the constructs passed the discriminant validity test.This means that each construct was distinct from the other constructs in the model.DOT construct had the highest discriminant validity, with a square root of AVE of 0.845.The correlations between DOT and the other constructs ranged from 0.486 to 0.845.This suggests that the DOT construct was well-represented by its observed variables and that it was not too highly correlated with the other constructs in the model.Table 4

Factorial Validity
The factorial validity test conducted to evaluate whether the factors that passed the convergent validity and discriminant validity tests represented the same higher-level construct (PV).Factorial validity test also detects and remove any cross-loading items (Cheung et al., 2023).The results shows that the higherorder (second-order) factor model had sufficient validity.The goodness-of-fit (GOF) of the final measurement model was within the acceptable range.The p-value for the model was 0.046, which was less than 0.05.This means that the model fitted the data.Furthermore, the GFI (0.965), TLI (0.976), and CFI (0.968) were all greater than 0.95, which proves that the model approached a perfect fit.Finally, the RMSEA (0.028) was less than 0.05, with a PCLOSE value of 0.984 and a lower end of the 90% confidence interval (LO 90) equal to zero (0).This provides strong evidence that the final model was a good fit to the data.

Structural model and hypothesis testing
After the researchers checked that the scales were reliable and valid, and that the model fits the data well, they evaluated the parameters of the structural model.The model had 3 latent constructs and 18 observed variables.The researchers used these variables to estimate the path coefficients, the explanatory power, and the relationships between the constructs in the structural model.In hypothesis testing, the Critical Ratio (CR) is a significant test statistic in structural equation modeling.It is calculated by dividing the unstandardized regression weight (URW) by its standard error.CR values greater than +1.96 and probability (p) values less than 0.05 indicates that the regression coefficient is statistically significant at the .05level ) (Byrne, 2013).Figure 2 shows the results of the structural model and Table 5 exhibits the path coefficients and the results of the hypothesis testing.As indicated in Table 5, quality of public services has a positive significant impact on public value creation (β = .068;CR = 5.545; p =.0.000),hence the hypothesis was supported.Similarly, achievable desirable outcomes and development of trust was found to have a significant impact on public value creation (β = 0.070; CR = 7.577; p = .000)(β = .0.078; CR = 6.861; p =.000).The results indicate that delivering e-HRM services results to creation of public value.This aligns with other research that were conducted previously (Abdulkareem & Mohd Ramli, 2022;Karkin & Janssen, 2014;Pooe & Munyanyi, 2022).

Qualitative Research Findings
In the qualitative strand, most employees interviewed affirmed creation of public value through e-HRM under the three themes presented as follows.

e-HRM results to improved quality of service delivery.
Extant literature indicates that investing in e-HRM would simplify administrative processes and improve HR service delivery (De Alwis et al., 2022).The participants were asked whether they felt that using e-HRM for service delivery made a difference to the process.Majority of the participants indicated that they felt that it does because information form e-HRM system is update, reliable, saves time, cuts down on administrative costs and is accurate.For instance, some employees responded as follows: Yes, information from e-HRM systems is reliable and update.Yes, use of e-HRM results to reduced use of papers therefore cutting down on administrative costs".Improved accuracy of the information contained in HR systems was achieved by removing duplicate keying of data.Use of e-HRM disseminates timely information"."The implementation of e-HRM in the County Government, particularly has aided in the resolution of numerous inexpensive errors such as information loss and unnecessary delays in some expected planned activities".

e-HRM results to achievable of desirable outcomes.
Research has shown that investing in e-HRM contributes to desirable outcomes ( (Nyathi & Kekwaletswe, 2024).In this study, participants were asked to share their insights on how e-HRM adoption could contribute to achievement of desirable outcomes.The participants highlighted that using e-HRM enhances transparency and fairness in HR functions.The participants also alluded that e-HRM facilitates the dissemination of timely information for decision making.Also, participants stated that e-HRM results to environmental sustainability.Some of their responses were as follows: "Use of e-HRM systems makes HR processes open /transparent and fair.""In the past, we used to submit paper forms, which would circulate from one person to another.This process often led to forms being misplaced, resulting in accusations of non-submission.The introduction of the new system allows us to track the progress of an action at any given moment, eliminating the issues associated with lost forms and ensuring better transparency in the process"."Use of e-HRM reduces the use of papers hence environmental sustainability"."Implementation of e-HRM has resulted in improving transparency and enhancing the HR image".

e-HRM results to Development of Trust
Trust denotes the anticipation of favorable outcomes regarding public services (Abdulkareem & Mohd Ramli, 2022).While Grimsley and Meehan (2007) suggest that trust could be considered an outcome, to underscore its significance and align with Moore's model, this study does not incorporate trust into the http://ijstm.inarah.co.id 573 outcome construct.The majority of respondents argued that the trustworthiness of e-HRM services is contingent on safeguarding the privacy and confidentiality of employees' information, a sentiment highly valued by the public.Existing literature emphasizes the crucial role of ensuring the privacy and security of employees (Bondarouk et al., 2017).Most of the interviewees agreed that e-HRM ensures the privacy and confidentiality of employee's information.This was reflected by Interviewees following respondents' assertions: "Information stored in e-HRM system is credible"."I believe information is well secured in the e-HRM systems".

V. CONCLUSION
The aim of the study was to evaluate how implementation of e-HRM contributes to the creation of public value in Kenya County Governments.The structural model analysis revealed significant positive relationships between the latent constructs (quality of public services, achievable desirable outcomes, development of trust) and public value creation.The path coefficients, supported by Critical Ratio (CR) values and p-values, demonstrate that quality of public services, achievable desirable outcomes, and development of trust all contribute significantly to public value creation through e-HRM implementation.These findings corroborate previous research, reinforcing the notion that e-HRM systems play a pivotal role in enhancing public value in organizational settings.Further, the qualitative findings provide nuanced insights into how e-HRM contributes to public value creation from the perspective of employees.The themes emerging from the qualitative data underscore the tangible benefits of e-HRM, including improved quality of service delivery, achievement of desirable outcomes, and development of trust.Employees perceive e-HRM as a tool that streamlines processes, enhances transparency, and facilitates better decision-making, ultimately leading to greater public value.The findings indicate that investing in e-HRM within the public sector is indeed associated with the generation of public value.Notably, these research outcomes align with and corroborate the findings of previous studies underscoring the significance of this relationship (AbdulKareem et al., 2024) .
This study provides several contributions to the field: Firstly, the study enriches the existing literature by analyzing the e-HRM proposition based on public value theory to explain what type of values are offered by e-HRM.By integrating quantitative and qualitative approaches, the study offers a comprehensive understanding of the complex relationship between e-HRM services and public value in County Government of Nyeri.The findings have several implications for practice and policy in County Government of Nyeri.Implementing e-HRM systems can lead to tangible benefits such as improved service delivery, cost savings, and enhanced transparency.Policymakers should prioritize investment in e-HRM infrastructure and training to maximize its potential for public value creation.However, despite the robust methodology employed in this study, there are limitations that should be acknowledged.The research was conducted in a specific context, limiting generalizability to other settings.Future research could explore the impact of e-HRM services in different organizational contexts and sectors.Additionally, longitudinal studies could investigate the long-term effects of e-HRM implementation on public value creation and organizational performance.Qualitative research could delve deeper into specific aspects of e-HRM, such as user experience and organizational culture, to provide richer insights into its implications for public value.

Fig 1 .
Fig 1. Conceptual Framework Fetters et al. (2013).The quantitative strand employed a cross-sectional survey design.The survey involved collecting data through a selfadministered questionnaire.The survey respondents were 209 employees of Nyeri County Government.Data was analyzed by means of structural equation modelling (SEM).The qualitative strand of the study sought views from employees of the County Government of Nyeri.Purposeful sampling was used to select participants interviewed while thematic technique was used to analyze qualitative data.IV.RESULTS AND DISCUSSIONS 4.1 Quantitative research findings This part of the findings presents a statistical analysis of the 209 respondents from the quantitative component of the study.The responses were randomly collected by a survey questionnaire in County http://ijstm.inarah.co.id 569

Table 1 .
exhibits a description of the indicator variables employed in the survey.Items Descriptions used in Measurement Model

Table 2 .
Table 2 exhibit results of GOF tests.The GOF results of initial and re-specified congeneric measurement models

Table 4 .
exhibit discriminant validity of the model Discriminant Validity Tests

Table 5 .
Hypotheses Test Results the Structural Model