The Influence of Work-Related Communication Technology Use during Non-Working Hours on Innovative Behavior: A Study on Government Employees in Hunan

Abstract

The sustainability of government innovation relies on the continuous innovative behavior of government employees. Information technology drives innovation, but its extensive use leads government employees to work outside of regular hours. This not only burdens them but also causes stress and tension, disrupting their innovative behavior. This study investigates the link between work connectivity behavior after hours (WCBA) and the innovative behavior of government employees using the resource conservation theory. It examines the mediating role of work engagement and the moderating effect of psychological resilience. Analyzing data from 275 government employees through questionnaires using Mplus 8.3, we discovered that WCBA negatively impacts innovative behavior and work engagement. Work engagement partially mediates the relationship between WCBA and innovative behavior. Additionally, psychological resilience negatively moderates the connection between WCBA and work engagement, influencing the innovative behavior of government employees. The findings offer theoretical and practical insights into reducing government burdens and fostering innovation, suggesting sustainable innovations for government employees.

1. Introduction

At present, reducing the burden on government employees to stimulate innovation and vitality has become an urgent need for government transformation and sustainable development. Information technology, a key element in government transformation, plays a crucial role in reducing workloads and promoting innovative behavior among government employees. However, excessive use of information technology leads to a significant amount of invisible digital work [1]. This situation forces government employees to stay connected to their work during non-working hours, causing stress due to technological intrusion into their personal lives [2]. For instance, government employees are required to manage numerous WeChat group messages, government apps, and repetitive data entry tasks after work [3]. This phenomenon, known as work connectivity behavior after hours (WCBA), not only results in negative behaviors such as emotional exhaustion, reduced quality of sleep, and turnover but also impacts productivity and well-being at work [4,5,6,7,8]. Does the adverse impact of WCBA on individuals affect the innovative behavior of government employees? There is limited academic research on this issue.

Innovative behavior of government employees essentially refers to the act of generating and implementing new ideas [9]. According to resource conservation theory, individuals with more resources such as time and energy are more productive and creative [10]. Therefore, when resources are scarce, government employees reduce or refrain from innovative activities to avoid resource loss [11]. Adequate free time helps employees to replenish their energy and improve their innovative performance [12]. However, off-hours work means that individuals still need to deal with work tasks after work, resulting in persistent resource depletion without sufficient time for recovery, which makes it difficult for resource-poor government employees to focus on work innovation. Despite this, current research on innovative behavior has focused on public service motivation [13], organizational climate [14], and leadership [15], neglecting the impact of off-hours work on government employees’ innovative behavior.

In research on innovative behavior, the majority of academics contend that job features influence government employees’ inventive behavior by altering their level of work engagement [16]. Continuing to work after hours, as a typical feature of the work of Chinese government employees, similarly has an impact on work engagement. The conservation of resources theory posits that the work engagement of government employees is significantly influenced by resources such as time and effort [17]. However, WCBA consumes government employees’ time and energy, and the burden of overloaded tasks makes it difficult for them to dedicate the necessary time, energy, and enthusiasm to their work. Consequently, innovative ideas or behaviors are impeded. This implies that work engagement serves as a mediator in the relationship between WCBA and innovative behavior.

Furthermore, innovative behavior is considered a potential stressor [18]. Psychological resilience is defined as an individual’s ability to adapt and recover in the face of stress and setbacks, and individuals with higher psychological resilience are more likely to adapt to stress and recover from it [19]. Therefore, psychological resilience may influence the innovative behavior of government employees. Studies have also shown that psychological resilience, as a reservoir of resources, plays a crucial role in reducing stress and restoring vitality [20]. This study argues that psychological resilience moderates the relationship between work engagement and innovative behavior. The willingness of government employees to proactively engage in innovative behavior is significantly influenced by their level of psychological resilience on an individual basis.

In conclusion, this study is based on the technology burden perspective and aims to investigate the correlation between the WCBA and innovation behavior of government employees in Hunan. This study attempts to make contributions in the following aspects: Firstly, it is grounded in the theory of resource preservation to confirm the impact of WCBA on government employees’ innovative behavior and to elucidate the mechanism of the relationship between WCBA and innovative behavior by using work input as a mediator variable. This study not only verifies that WCBA, as a resource-consuming behavior, inhibits the motivation of government employees’ innovative behaviors but also, in a Chinese context, explains the reasons why government employees are unwilling and unable to innovate. Secondly, this study analyzes the relationship between WCBA and personal resilience on the innovative behavior of government employees in Hunan and investigates the moderating role of psychological resilience between work engagement and inventive behavior. Thirdly, this study examines how work connectivity behavior after hours affect innovation behaviors. The results of this study provide useful recommendations for technology load shedding for government employees and the sustainability of government innovation.

2. Theory and Hypotheses

2.1. WCBA and Innovative Behavior

The current form of government burdens has evolved: online digital loads have replaced offline substantive burdens as the predominant form of government burden [21]. Although WeChat and government apps have eased the work process, job requirements such as “24 h a day” and “leaving traces everywhere” have become the mode of operation for the government. This means communication tools put a lot of strain on government employees and increase their workloads significantly. According to Resource Conservation Theory, in order to prevent the effects of stress, people want to acquire a variety of psychological, social, and physical resources. Resource replenishment after work is an even more effective strategy to realize resource gain [22]. However, WCBA compels government employees to respond quickly 24 h a day, which makes psychological disengagement more difficult for government employees [23]. As a consequence, life is cluttered by work, depleted resources are unable to be replenished efficiently, and energy and emotions are drained by online formalized duties [24]. Research has shown that WCBA continues to limit individual freedom, preventing individuals from achieving physical and mental detachment from work and hindering creativity [25]. Therefore, this paper proposes the following hypotheses:

H1: 

WCBA negatively affects innovative behavior.

2.2. The Mediating Role of Work Engagement

In the digital age, information technology is essential for government employees’ work, not only to enhance governance but also to facilitate evidence-based decision-making. However, it comes with challenges such as digital formalism, invisible work, and other complexities [1]. At the grassroots level, which represents the end of the bureaucratic chain, employees face mounting pressure from superiors. They have to navigate through various workgroups, contend with redundant and intricate government applications, and repeatedly complete numerical forms outside regular working hours. These obstacles often lead to negative emotions like fatigue, burnout, and anxiety, hindering their ability to fully engage in their tasks [26]. According to the Resource Conservation Theory, if an individual expends more resources managing stress than the benefits gained, the outcome is likely to be adverse [27]. Therefore, utilizing communication tools to address work responsibilities during non-working hours not only demands a significant investment of resources, including government employees’ time, energy, and focus, but also results in these resources not being promptly replenished, subsequently impacting their future work engagement [28]. This suggests that WCBA may contribute to government personnel experiencing burnout, diminishing their effectiveness, and impeding the development and execution of innovative practices [29]. Therefore, this study proposes the following hypothesis:

H2: 

Work engagement plays a mediating role in WCBA and innovative behavior.

2.3. The Moderating Role of Psychological Resilience

Government employees are required to use communication tools outside of working hours to handle tasks such as formalized work, leave-track appraisal, and show-off punching. This unintentionally extends their working hours, increases their workload, and surpasses their psychological tolerance. The negative results of WCBA, such as emotional weariness and lower job satisfaction, are signs of burnout in government employees, which is a reaction to chronic, long-term pressures on the job [30]. Studies have demonstrated that personal characteristics influence how individuals respond to stress, and psychological resilience is considered a safeguarding element that plays a crucial role in reducing stress and enhancing mental well-being. Government employees with low psychological resilience exhibit less capacity for self-regulation, hindering their ability to effectively adjust to the work environment and perhaps subjecting them to additional responsibilities [31]. Thus, when government employees possess a greater degree of resilience, they exhibit an enhanced capacity to cope with intense, stressful, and demanding jobs. As a result, the likelihood of burnout decreases as work engagement levels rise [32]. Therefore, this study proposes the following hypothesis:

H3: 

Psychological resilience plays a moderating role in WCBA and work engagement.

Psychological resilience not only regulates the link between WCBA and work engagement, but it also significantly moderates work engagement’s role as a mediator in the relationship between WCBA and innovative behavior. Resource conservation theory highlights the importance of resource investment, which implies that people would invest in resources to prevent their loss [33]. Individuals with high resilience exhibit positive coping strategies regardless of the magnitude of stress they encounter [34]. This demonstrates that psychological resilience is a valuable resource for individuals to invest in. More precisely, people with a strong level of resilience are capable of effectively managing job-related demands, thereby avoiding resource depletion and enabling them to focus on their tasks and exhibit inventive behaviors. Individuals with low resilience find it difficult to recover from the loss of resources caused by high-intensity and high-stress employment, making it challenging to have additional resources to spend on work and innovation. Therefore, this study proposes the following hypothesis:

H4: 

Psychological resilience moderates the indirect effect between WCBA and innovative behavior via work engagement.

The theoretical model of this study is shown in Figure 1.

3. Methods

3.1. Sample and Procedure

The research object of this paper is government employees in Hunan. The reason is that compared with other employees, government employees are the key bridge connecting the government and the public. They are the actual executors of various affairs. In order to ensure that work is handled in a timely and continuous manner, they have to sacrifice a large amount of non-working time to complete tasks. However, individual resources are limited, and prolonged occupation of rest time may lead to difficulty for government employees to focus their attention and energy on work innovation. Therefore, it is important to explore the impact of work connectivity behavior after hours on the innovation behavior of government employees. This study adopts a convenience sampling method, focusing on government employees at the township level in Changsha County (district), Hunan Province. The reason for selecting government employees in Hunan Province as the research object is that in recent years, the province has actively carried out grassroots burden reduction activities. In July 2023, the “Specific Measures for Deepening the Rectification of ‘Mountains of Documents and Seas of Meetings’ and Other Prominent Problems of Formalism and Bureaucracy in Hunan Province” was issued to address issues such as document issuance, meetings, reporting, forms, supervision, assessment, and government application promotion [35]. Despite reductions in paperwork, government employees still face bureaucratic challenges and overtime work, which can hinder innovation. Therefore, it is necessary to investigate the impact of work connectivity behavior after hours on innovation among government employees. Before distributing the formal questionnaire, this paper referred to Li Zhuo et al. [36] and used a simple random sampling sample size calculation formula in statistics to determine the minimum effective sample size required for this study:

$$n={\left({\displaystyle \frac{Z_{1-\alpha /2}}{\delta }}\right)}^2\times p\times \left(1-p\right)$$

(1)

n denotes the minimum effective sample size; p denotes the proportion of the target population to the total population; α denotes the confidence level; Z_1−α/2 denotes the value of the standardized normal distribution, which can be obtained by consulting the table of standardized normal distribution; δ denotes the permissible error.

Current surveys on government employees’ work connectivity behavior after hours are limited, and working hours are one of the indicators of work connectivity behaviors. Thus, this study refers to the questionnaire survey on working hours of grassroots conducted by Chen Jagang et al. [37]. It is assumed that the grassroots civil servants who have work connectivity behaviors during non-working hours accounted for 77.92% (p = 77.92%). Let the confidence level α = 0.95. Consulting the standard normal distribution table gives Z_1−α/2 = 1.96. The allowable error is set to 5%, i.e., δ = 0.05. Based on the above data and Equation (1), a minimum of 264 samples are required for this study.

The survey process is as follows: In the pre-survey phase, interviews were conducted with ten government employees to understand the intricacies of grassroots-level work. Next, specialists translated the English scales into Chinese scales, which were then combined with Chinese government employee characteristics to develop the initial questionnaire. Subsequently, 30 pre-survey questionnaires were collected, and their reliability and validity met the standard level, fulfilling the requirements of empirical research. In the formal investigation phase, we contacted Master of Public Administration (MPA) students at universities, as most MPA students in China are government employees, making it effective to ensure sample validity by having them complete the questionnaire. We obtained a total of 100 questionnaires through this method. Secondly, we utilized social media platforms to specifically target government employees by searching for terms like “government employees” and “government employees’ jobs” on these platforms. We then messaged each individual to inquire if they were willing to participate in the survey, resulting in a total of 53 completed questionnaires. Finally, this study leveraged acquaintance networks to distribute questionnaires to government employees via WeChat and QQ. Simultaneously, the snowball method was employed to ask government employees who have completed the questionnaire if they would be willing to pass it on to their colleagues. When this process reached the third level, the total number of samples reached 275 (>264), and the process concluded.

The basic characteristics of the sample are as follows: The sample in this survey is balanced between males and females, with 119 males (43%) and 156 females (57%). In terms of age, the government employees who participated in the survey are mainly young people, among whom 55 (20%) are under the age of 24, 116 (42%) are between the ages of 25 and 34, 59 (22%) are between the ages of 35 and 44, and 45 (16%) are over the age of 45. In terms of education level, a bachelor’s degree is the main one, among which there are about 18 (7%) people with high school or below, 56 (20%) people with college degrees, 142 (51%) people with bachelor’s degrees, and 59 (22%) people with postgraduate degrees or above. In terms of years of experience, 88 (31%) had 3 years or less; 54 (20%) had 4 to 6 years; 63 (23%) had 7 to 9 years; and 70 (26%) had 10 years or more.

3.2. Measures

The main research variables in this paper are WCBA, work engagement, psychological resilience, and innovative behavior. To ensure the reliability and validity of the measurement variables, this study referenced previous research and employed well-established domestic and international scales for each variable. Apart from control variables like gender, age, education level, and work experience, the measurement scales used are Likert five-point scales.

3.2.1. WCBA

WCBA uses a scale developed by Richardson and Benbunan-Fich [4]. WCBA is measured primarily in two aspects: duration and frequency of use. In terms of time of use, it mainly examines the length of time government employees use communication tools to handle their work during various non-working hours. In terms of frequency of use, it focuses on how often government employees use communication tools to handle work in various non-work settings. The duration of use is mainly combined with Richardson and Thompson [38] and Ma Hongyu et al. [39]. The scale is divided into five parts: before work, during lunch break, after work, on weekends, and on holidays, with a Cronbach’s alpha of 0.769. For the frequency of use, we asked how frequently a technological device is used during a specific non-work activity (e.g., shopping, commuting to/from work, a meal at home/restaurant, a movie in a theater, etc.). There are a total of 6 items, the Cronbach’s alpha is 0.849, and the Cronbach’s alpha of the total scale is 0.865.

3.2.2. Psychological Resilience

We measured psychological resilience using the Connor–Davidson Resilience Scale. The CD-RISC consists of 25 items, each rated on a 5-point scale (0–4), where higher scores indicate greater resilience (Connor and Davidson, 2003) [40]. The focus of this study is government employees, so the chosen items should be relevant to this population. Following the elimination of non-conforming items, the research utilized four items to assess the psychological resilience of government employees. Sample items are able to “adapt to change, make the best effort, and handle unpleasant feelings.” The Cronbach’s alpha is 0.870.

3.2.3. Work Engagement

We measured work engagement with a scale developed by Schaufeli et al. [41]. Schaufeli measures “engagement” in three dimensions: vigor, dedication, and absorption. Vigor refers to the high levels of energy and mental resilience to work (e.g., I can continue working for very long periods at a time, I feel bursting with energy at my work). Dedication refers to a sense of significance, enthusiasm, and pride (e.g., I am enthusiastic about my job). Absorption refers to being fully concentrated and deeply engrossed in work (e.g., When I am working, I forget everything else around me). In total, there are 4 items, and the Cronbach’s alpha is 0.912.

3.2.4. Innovative Behavior

We measured innovative behavior using a scale developed by Susanne Scott and Regiaald Bruce [42]. The scale comprises four items to assess innovative behavior. Sample items include “Generates new ideas; learns new knowledge; focuses on new technologies; implements or promotes new ideas and technologies to others”. With a total of four items, the Cronbach’s alpha is 0.925.

3.2.5. Control Variables

According to previous studies, we also collected the following demographic variables as control variables: age, gender, education level, and tenure.

4. Results

4.1. Reliability and Validity Test

Before conducting structural equation modeling (SEM), this study used Mplus 8.3 to perform a confirmatory factor analysis (CFA). The results of the reliability of each variable are presented in

Table 1. Reliability and validity test.

Variable	Cronbach’s a	STD.LOADING	AVE	CR
Duration of WCBA	0.769	0.504–0.870	0.429	0.783
Frequency of WCBA	0.849	0.574–0.830	0.539	0.852
Psychological resilience	0.870	0.767–0.820	0.628	0.871
Work engagement	0.912	0.824–0.860	0.723	0.913
Innovative behavior	0.925	0.856–0.891	0.756	0.925

N = 275. WCBA = work connectivity behavior after hours, duration and frequency represent two dimensions of WCBA.

. Combined reliability (CR) was utilized to assess the internal consistency and stability of the scale, with a CR exceeding 0.6 considered an acceptable threshold [43]. The CR for the latent variables in this study ranged from 0.783 to 0.925, all surpassing the critical value of 0.6. Hence, the model developed in this study demonstrates good internal consistency and stability. Convergent validity and discriminant validity were employed to evaluate the model’s validity. The standardized factor loadings of the variables ranged from 0.504 to 0.891, surpassing the standardized value of 0.5, with p-values reaching significance levels, indicating high-quality measures. The average variance extracted (AVE) was used to assess the convergent validity of the scale, where an AVE greater than 0.36 is deemed acceptable and greater than 0.5 is considered ideal [44]. The AVE values in this study ranged from 0.429 to 0.756, suggesting strong convergence of the model. In conclusion, the variable measurements largely meet the criteria and are suitable for the subsequent stage of structural modeling.

4.2. Common Method Bias

To assess the impact of common method bias on the models, this study employs Harman’s one-factor test. The fit indices for the one-factor model are as follows (refer to

Table 2. Fit indices of each model.

Model	Factor	χ2	df	χ2/df	RMSEA	CFI	TLI	SRMR
Five-factor model	A1,A2,B3,C4,D5	272.952	199	1.372	0.037	0.982	0.979	0.036
Four-factor model	A1,A2,B3,C4 + D5	305.191	203	1.503	0.043	0.975	0.972	0.037
Three-factor model	A1,A2,B3 + C4 + D5	379.049	206	1.840	0.055	0.958	0.953	0.041
Two-factor model	A1 + A2,B3 + C4 + D5	492.789	208	2.369	0.071	0.931	0.923	0.048
One-factor model	A1 + A2 + B3 + C4 + D5	790.532	209	3.782	0.101	0.859	0.844	0.072

N = 275. A1 and A2 represent two dimensions of WCBA: duration and frequency; B3 = Psychological resilience; C4 = Work engagement; D5 = Innovative behavior.

): χ² = 790.532, df = 209, RMSEA = 0.101, CFI = 0.859, TLI = 0.844, and SRMR = 0.072. For the five-factor model, the fit indices are as follows: χ² = 272.95, df = 199, RMSEA = 0.037, CFI = 0.982, TLI = 0.979, and SRMR = 0.036. The fit indices of the five-factor model are significantly superior to those of the single-factor model, suggesting that common method bias has a minimal impact on this study. Additionally, the five-factor model outperforms other models, demonstrating good discriminant validity.

4.3. Descriptive Statistics and Correlations

This study shows the results of descriptive statistics and correlation analysis in

Table 3. Means, standard deviations, and correlations of the research variables.

Variable	1	2	3	4	5	6	7	8
1. Gender	1
2. Age	−0.258 **	1
3. Education	0.168 **	−0.433 **	1
4. Tenure	−0.269 **	0.738 **	−0.318 **	1
5. WCBA	0.138 *	0.123 *	0.010	0.116	1
6. Psychological resilience	0.071	−0.208 **	0.067	−0.141 *	−0.617 **	1
7. Work engagement	0.082	−0.101	−0.004	−0.066	−0.666 **	0.826 **	1
8. Innovative behavior	0.127	−0.120 *	0.065	−0.111	−0.679 **	0.766 **	0.889 **	1
M	1.570	2.340	2.880	2.420	3.275	3.312	3.137	3.021
SD	0.496	0.978	0.817	1.182	0.727	1.004	1.175	1.233

N = 275. * p < 0.05. ** p < 0.01. Gender: 1 = ’male’, 2 = ’female’; Age: 1 = 24 years old and below, 2 = 25–34 years old, 3 = 35–44 years old, 4 = 45 years old and above; Education: 1 = High school and below, 2 = Junior college, 3 = Bachelor’s degree, 4 = Graduate and above; Tenure: 1 = 3 years and below, 2 = 4 to 6 years, 3 = 7 to 9 years, 4 = 10 years and above.

. WCBA is negatively correlated with work engagement (r = −0.666, p < 0.01) and negatively correlated with innovative behavior (r = −0.679, p < 0.01); work engagement is significantly positively correlated with innovative behavior (r = 0.889, p < 0.01); and psychological resilience is significantly positively correlated with work engagement (r = 0.826, p < 0.01). The results of the correlation analysis initially verify the hypotheses proposed and lay the foundation for better hypothesis testing in the next step.

4.4. Regression of the Study Variables

This study uses regression analysis to test each hypothesis, and the results are presented in

Table 4. Regression analysis.

	Work Engagement						Innovative Behavior
	Model 1	p	Model 2	p	Model 3	p	Model 4	p	Model 5	p	Model 6	p	Model 7	p
Gender	0.054	0.392	0.028	0.556	−0.022	0.511	−0.023	0.488	0.080	0.206	0.053	0.255	0.031	0.264
Age	−0.135	0.154	−0.061	0.390	0.051	0.313	0.042	0.402	−0.069	0.464	0.006	0.929	0.054	0.192
Education	−0.062	0.354	−0.009	0.861	−0.018	0.612	−0.025	0.471	0.010	0.876	0.065	0.191	0.072	0.016
Tenure	0.028	0.758	0.060	0.374	0.010	0.829	0.009	0.853	−0.036	0.692	−0.003	0.97	−0.05	0.208
WCBA			−0.664	0.000	−0.251	0.000	−0.246	0.000			−0.676	0.000	−0.154	0.000
PR					0.687	0.000	0.719	0.000						0.000
WCBA × PR							−0.096	0.004
WE													0.787
R2	0.016		0.447		0.728		0.736		0.022		0.469		0.812
△R2	0.001		0.436		0.722		0.729		0.007		0.459		0.807
F	1.089		43.403		119.309		106.193		1.49		47.454		192.499
p	0.362		0.000		0.000		0.000		0.205		0.000		0.000

N = 275. The regression coefficient β is the standardized coefficient. PR = psychological resilience, WE = Work engagement.

. Model 1 included control variables in the analysis, revealing that gender, age, education, and tenure were not strongly associated with work engagement. In Model 2, WCBA was introduced, showing a significant prediction of work engagement (β = −0.664, p < 0.01). Model 3 incorporated psychological resilience, which also significantly predicted work engagement (β = 0.687, p < 0.01). Interestingly, the regression coefficient of WCBA and work engagement changed after the inclusion of psychological resilience (β = −0.251, p < 0.01). Model 4 explored the interaction effect between WCBA and psychological resilience on work engagement, indicating that psychological resilience negatively influenced the relationship between WCBA and work engagement. The interaction term was found to be negatively significant (β = −0.096, p < 0.05). Combining the results from Models 2, 3, and 4, it was observed that psychological resilience acts as a negative moderator in the relationship between WCBA and work engagement. Consequently, with an increase in psychological resilience, the adverse impact of WCBA on work engagement diminishes, thereby supporting hypothesis H3. Model 6 demonstrated that WCBA significantly and negatively predicted innovative behavior (β = −0.676, p < 0.01), thereby supporting hypothesis H1. Model 7 revealed that work engagement significantly and positively predicted innovative behavior (β = 0.787, p < 0.01). Additionally, the regression coefficient of WCBA changed (β = −0.154, p < 0.01), suggesting that work engagement acts as a mediator between WCBA and innovative behavior. The study provides support for hypothesis H2.

4.5. Mediating Effect Testing

To further verify the mediating effect of work input, we conducted a bootstrap test using Mplus 8.3 with 1000 repeated samples.

Table 5. The mediating role of work engagement.

ITEM	Point Estimate	Product of Coefficients			BC 95% CI
		S.E.	Est./S.E.	p-Value	Lower	Upper
Total	−0.734	0.07	−15.301	p < 0.01	−0.725	−0.594
Total indirect	−0.393	0.065	−10.343	p < 0.01	−0.492	−0.349
Total direct	−0.341	0.083	−4.952	p < 0.01	−0.352	−0.156

N = 275.

presents the measures of the indirect effect. For WCBA on innovative behavior, the direct and indirect effects are −0.341 (p < 0.01) and −0.393 (p < 0.01), respectively, accounting for 38% and 62%. The bootstrap analysis shows a 95% confidence interval of [−0.725, −0.594], excluding 0, indicating that work engagement played a significant partially mediating role between WCBA and innovative behavior, further supporting hypothesis H2.

4.6. Moderating Effect Testing

To further confirm the moderating effect of psychological resilience, we conducted a moderating effect test. The interaction index comprises the WCBA index and the psychological resilience index. Subsequently, we tested the moderating effect of the product and present the results in

Table 6. The moderating role of psychological resilience.

Variable	Work Engagement		Work Engagement
	β	t	β	t
WCBA	−0.253	−6.232	−0.249	−6.201
Psychological resilience	0.670	16.483	0.703	16.894
Product interaction term			−0.099	−2.966
R2	0.722		0.731
F	352.902		244.946

. Table 6 indicates that the interaction term of WCBA and psychological resilience significantly predicts work engagement (r = −0.099, p < 0.01). The F-test reveals a significant regression equation (R² = 0.731, F = 244.946, p = 0.03). A simple slope test was performed, a moderating effect graph was created, and the results are displayed in Figure 2. The interaction term exhibits a negative slope, suggesting that psychological resilience exerts a negative moderating effect between WCBA and work engagement. In comparison to low psychological resilience, as psychological resilience increases, the negative relationship between WCBA and work engagement persists, albeit weakened. This further supports hypothesis H3.

4.7. Moderated Mediation Effect Testing

To verify the moderated mediation model, we conducted bootstrapping tests and path analysis using PROCESS with 1000 replicate samples. The results are presented in

Table 7. Results of the moderated mediation effect.

Test-Condition	WCBA (X)→Work Engagement (M)→Innovative Behavior (Y)
	Indirect Effect	SE	BC 95% CI
Low-Psychological resilience (−1 SD)	−0.185	0.083	−0.363	−0.036
High-Psychological resilience (+1 SD)	−0.478	0.094	−0.675	−0.304
Difference	−0.293	0.104	−0.523	−0.101

and Figure 3. It can be observed that when psychological resilience is low, the value of the indirect effect of WCBA influencing the innovative behavior of government employees through work engagement is −0.185, with a confidence interval of [−0.363, −0.036], excluding 0. When psychological resilience is high, the value of the indirect effect of WCBA on government employees’ innovative behavior through work engagement is −0.478, with a confidence interval of [−0.675, −0.304], excluding 0. There is a significant difference in the indirect effect at the high and low levels, with a difference value of −0.293 and a confidence interval of [−0.523, −0.101], not including 0. Therefore, the higher the psychological resilience of government employees, the weaker the value of the indirect effect of WCBA in influencing the innovative behavior of employees through work engagement. Hypothesis H4 is supported.

5. Discussion and Implications

This study utilizes resource preservation theory to explore the impact of work connectivity behavior after hours on innovative behaviors among government employees in Hunan. It also assesses the mediating role of work engagement and the moderating influence of psychological resilience. This study’s results suggest that work connectivity behavior after hours negatively affects government employees’ work engagement and innovative behavior. The main reason behind this is that, according to resource conservation theory, government employees have limited resources, such as time and energy. Downtime serves as the primary method to replenish and rejuvenate these resources. Consequently, government employees who are heavily burdened by technology struggle to allocate additional energy and time for innovation, so government employees’ innovative behaviors are discouraged, which hinders the sustainability of government innovation. Additionally, although resilience acting moderates the relationship work connectivity behaviors after hours and work engagement as well as innovative behaviors, for government employees with low resilience, the use of information technology causes leads to rather than efficiency, productivity. These findings align Jarvenpaa et al. [25].

5.1. Theoretical Implications

First, this study is based on the theory of resource conservation to investigate the antecedent characteristics of innovative behavior among government employees, thus enhancing the research scope of grassroots innovation. Current research on work connectivity behavior after hours focuses on corporate employees and rarely includes government employees. In fact, in the Chinese context, it has become common for government employees to work overtime, and continuing to work after hours is considered necessary by most supervisors under the concept of being people-centered and dedicated to serving the people. However, according to the theory of resource conservation, people’s time, energy, and other resources are limited. Once the resources of government employees are undermined, it will be difficult for government employees to continue to work effectively, which not only affects the performance of government employees but also hinders the sustainability of the government. So based on this research gap, this paper attempts to explore the impact of work connectivity behavior after hours on the group of government employees. In addition, most of the studies on innovative behavior of government employees have focused on normal working hours, and little attention has been paid to the impact of off-hours work on innovative behavior. According to the resource conservation theory, work connectivity behavior after hours is a continuous resource consumption behavior. Innovation is a long process that requires the support of a large amount of resources, so it may be difficult for resource-poor government employees to generate innovative behavior, but few studies have focused on this point. Based on this research gap, this study empirically confirms the correlation between work connectivity behavior after hours and innovation behavior in the current Chinese context. Furthermore, it contributes to the understanding of how technology capability can deteriorate into negative capability, as discussed in the study by Hu Weiwei et al. [45].

Second, this study reveals the mediating role of work engagement in the relationship between work connectivity behavior after hours and innovation behavior based on resource preservation theory. This paper argues that government information technology, as a mode of work, should be matched with the resources possessed by government employees, such as time, energy, and skills. Work connectivity behavior after hours, however, not only disregards the availability of government employees’ time, energy, and other resources but also exceeds reasonable working hours and the psychological capacity of individuals, exposing individuals to the danger of resource loss and the failure to replenish and renew resources. According to Resource Conservation Theory, the consequence of coping is likely negative if one invests more resources than the resultant benefits. Thus, the need to use electronic communication technology to stay in touch with work during non-working hours may lead to engagement becoming ineffective and even affecting performance the next day. The results of this study address Tian et al.’s [28] concerns about non-working time crowding being negatively related to job performance.

Third, this study examines psychological resilience as a boundary condition for the relationship between work connectivity behavior after hours and innovative behaviors. It sheds light on the impact of government employees’ use of electronic communication tools on innovative behaviors under varying levels of resilience. WeChat, microblogs, government apps, and other tools provide autonomy and flexibility for government employees but also demand instant responses, photo-taking, clocking in, and form submissions, blurring the lines between work and personal life. Staying connected to work through electronic communication during off-hours can lead to role conflicts for individuals with low resilience. Government employees may find it challenging to cope with high-intensity, long work hours, leading to negative emotions like burnout and exhaustion, which can impact work engagement and innovative behaviors. These findings build on Zhao Yanxin’s [31] research on the causes of overlapping responsibilities at the grassroots level.

5.2. Practical Implications

Firstly, attention is paid to the double-edged sword effect of information technology in the process of change and innovation in government departments. While the use of information technology has increased the autonomy and flexibility of government employees’ work, this study has found that the use of information technology to handle work during non-working hours has negative direct and indirect effects on government employees’ innovative behavior. According to the theory of resource conservation, an individual’s resources are limited, and the lack of resources interrupts the continuity of an individual’s work. Therefore, government departments should be aware that the excessive use of technology can easily lead to the loss of resources of government employees, making it difficult for them to focus their time, energy, attention, and other resources on the change and innovation of government organizations. Specifically, government departments should make rational use of information technology, strengthen the standardized management of WeChat workgroups and government apps based on the overall needs of the public, reduce unnecessary work, especially repetitive and formal tasks, and allow sufficient rest time for government employees to ensure the restoration and sustainable renewal of resources and to improve the performance of innovation.

Secondly, there should be focus on the critical role of work inputs in the process of change and innovation in government departments. Job characteristics influence changes in work engagement. Continuing to work after hours is a distinctive feature of Chinese government employees. For government departments, requiring government employees to continue to be engaged in work hours work is conducive to increasing the amount of work accomplished. However, for government employees, it is difficult to have additional resources to continue working hours. The result may be burnout rather than increased productivity and innovative performance. Therefore, government departments should focus on the quality of work input rather than overemphasizing work input hours. Specifically, government departments should balance the length of work with the length of rest and should rationalize work tasks to free government employees from pseudo-busy or tedious work. This will not only help government employees to devote themselves to actual work in a more enthusiastic and focused state but also help to promote substantive work progress and facilitate the generation and practice of innovation. At the same time, government departments should show humanistic care and pay attention to the emotional changes of government, and positive emotions will help with work and innovative behavior.

Thirdly, it is crucial to emphasize the significant contribution of psychological resilience in facilitating change and promoting innovation within the government sector. Psychological resilience plays a role in modulating the association between work connection behaviors after hours and innovative behaviors among government employees. Specifically, the way government employees stay connected to work outside of regular hours has a detrimental impact on their ability to be inventive. However, this negative effect is reduced when individuals have strong levels of psychological resilience. Government departments need to focus on developing psychological resilience in government employees. This can be achieved by enhancing their psychological well-being and self-regulation skills, as well as integrating psychological counseling into regular management practices. These measures will ultimately enhance the overall ability of government employees to cope with pressure. Simultaneously, government employees at the bottom of the bureaucratic hierarchy have to manage workloads systematically delegated and intensified by higher authorities. However, government employees face resource constraints, and even individuals with great resilience require sufficient time and space to alleviate work-related stress.

5.3. Limitations

This study also has the following shortcomings: First, the variables in this study mainly focus on the individual level of government employees. Future studies could include the families of government employees to more comprehensively measure the impact of work connectivity behavior after hours on government employees. Second, the data in this study mainly come from government employees in Hunan Province. Future research can expand the scope of the study to increase the scientific rigor and applicability of the study. Third, this study only explores the mediating effect of work connectivity behavior after hours affecting government employees’ innovative behaviors through work engagement and the boundary role of psychological resilience. Future studies on the relationship between off-hours work connectivity behaviors and government employees’ innovative behaviors can consider factors such as psychological disengagement, information burnout, and other role mechanisms or the impact of boundary mechanisms to more comprehensively and deeply clarify the relationship between the two.