Continuance Adoption of Working from Home after the COVID-19 Outbreak: Empirical Evidence from Saudi Arabia

Abstract

The COVID-19 pandemic sweeping the world has rendered a large proportion of the workforce unable to commute to work, to mitigate the spread of the virus. This has resulted in both employers and employees seeking alternative work arrangements. Due to the pandemic, most if not all workers experienced work from home Hence work from home has become a policy priority for most governments. Individuals have started to change their behavior to stick to the curfew and rapidly conform to the new way of life. This study is conducted to understand how organizations and people adjust to these developments and challenges. Numerous organizations are changing to the online method of working because of the COVID-19. Because of the continuous adoption of a specific behavior after the COVID-19 pandemic situation ended, employees were expected to continue working from home. To investigate deep into the behavioral consequences of such a pandemic situation, in-depth interviews were conducted in several companies in Saudi Arabia. This study was conducted by extending the Theory of Planned Behavior (TPB) and the theory of reasoned action (TRA) to explore and assess the various factors that determine the continuous adoption intention of work-from-home by the Saudis. The finding shows that the employees' positive attitude, subjective norms, and self-efficacy affect the employees' intention to adopt work from home. The continuous adoption of work from home has been affected by employees' relevant intention and controllability.

1. Introduction

The world witnessed the outbreak of a virus pandemic in December 2019, which is known as the coronavirus (COVID-19), which triggered a new pandemic in all countries around the world in 2020 (Herwany et al., 2021). Since October 30, 2020, COVID-19 has spread to 216 countries and territories (Khanthavit, 2021). It impacted people’s behavior, not just in Saudi Arabia but across the globe. In Saudi Arabia, 37 million people have been affected financially and economically due to the COVID-19 outbreak. The Saudi government has been taking proactive measures since the pandemic was discovered. To mitigate the spreading of the virus, restrictions have been imposed on public gatherings, including workplace attendance. Workplaces were not the only places that were shut down, as the lockdown was also applicable for schools, universities, and even government agencies. On the 14^th of March 2020, government regulations directed a strict curfew in all the cities, restricting all kinds of gatherings, and on the 16^th of March 2020, released a set of guidelines on working from home. In relation to business owners, several researchers have stated that if the work-from-home model is implemented with adequate resources to the employees, a rise in productivity and a decrease in the stress of the employees will be found (Gibson & Shrader, 2018).

Furthermore, when employees are working remotely and in different areas, it enables the firm to obtain a better understanding of the local market. Continuous adoption is the measure to which a person continuously adopts a specific behavior (Lu & Hsiao, 2010). One can assess the success of the work-from-home model by evaluating the enthusiasm exhibited by the targeted population to adopt it. If businesses execute the work-from-home model despite the lack of adoption by employees, they are unlikely to observe any significant advantages. Thus, in terms of implementing the work-from-home model post-pandemic, understanding factors that influence the continuous adoption by targeted individuals is important. The businesses must conduct an assessment to gauge if the work-from-home model is continuously supported by their employees, otherwise, it might create negative consequences of the WFH initiative. This study, therefore, aims to explore the factors influencing the continued adoption of remote work arrangements by employees after the pandemic ends. It was found that the extended models have been quite successful in predicting people’s post-adoption behavior in several contexts (Wang & Chou, 2016). As the Theory of Planned Behavior (TPB) is one of the most popular theories on human behavior, this study aims to explore and assess the various factors that determine the continuous adoption intention of the work-from-home model by the Saudi population.

2. Literature Review

COVID-19 was announced as a world health emergency by the WHO in January 2020. By early March, the focal point of the virus shifted from China to other parts of Asia and Europe, particularly Iran and Italy. By May 2020, the virus had rendered over 4 million people sick across the globe. On the 2^nd of March 2020, Saudi Arabia witnessed its first case of the novel pandemic. On the 12^th of March 2020, the Saudi Ministry of Municipal and Rural Affairs imposed a total lockdown in all areas. The first case of COVID-19 was brought into the country by a citizen, who returned to his homeland from Iran through Bahrain. The first corona virus-related death was reported on the 23^rd of March 2020. The departed was reported to be a fifty-one-year-old male, hailing from Afghanistan, but a resident of Medina. By the 24^th of March, the country had recorded 767 cases. The 12^th of May saw the country recording 1191 cases, with the total number of cases in the country escalating to 42, 925 (Weqaya, 2020).

According to analysts, the virus could potentially decrease the global economic growth by almost 2% every month, if the virus is not curtailed soon enough. Moreover, international trade could fall by 13% to 32. Furthermore, Saudi has lowered its interest basis points by 75, and its repo rate and its reverse repo rate have decreased to 1% and 0.5% (Aramco, 2020). Several companies in Saudi have laid off their employees, including the Seera group, which is also one of the largest travel companies in the UAE, Egypt, and some other countries. Its stock fell to an all-time low in March 2020, as it decreased by over 40%. Another example is Careem, a major corporation in Saudi, which is planning on laying off 31% of its workforce as the pandemic led to a 31% plunge in its overall business. Thus, during the COVID-19 outbreak, the companies have taken measured approaches to protect their employees from potential exposure to infection (Aramco, 2020). For example, many Saudi companies utilized the work-from-home model. Working from home, also known as remote working has seen a substantial rise in recent times due to the advent of digital communication, high-speed Internet, and collaboration mechanisms enabling staff to stay connected and conveniently work from home. According to Wade Foster (CEO of Zapier), remote working gives them access to the talent pool all over the world, while it also cuts costs effectively on parameters like office space, time lost due to traffic congestion, and other expenses associated with owning physical office spaces. Thus, cost cutting is a significant parameter that businesses can take into consideration when exploring the idea of implementing the work-from-home model.

Therefore, we must study employees’ behavior to understand and predict their continuous behavior towards adopting remote work. The theory of planned behavior (TPB) (Ajzen, 1985) is the extended version of the theory of reasoned action (TRA) (Ajzen & Fishbein, 1977). The TRA theorized that the primary determinant of behavior is the intention to exercise that behavior, provided, the behavior can be voluntarily regulated by the individual. Attitudes are the components that dictate intentions, influencing the individuals towards (or away from) the behavior, along with the perceived social pressures from important referents (subjective norms) to perform the behavior. It integrates an additional parameter determining the intentions and behavior, known as perceived behavioral control (PBC) (Ajzen & Fishbein, 1977). PBC is defined as the perception of the difficulty of enacting a behavior. PBC is the key difference between the Theory of Planned Behavior and the Theory of Reasoned Action. According to Ajzen (1985), practically no behavior is under the complete control of an individual. Internal, as well as external, factors, can prevent individuals from implementing the desired behavior. The predictive role of PBC depends on the extent to which the behavior can be voluntarily controlled, as well as the possibility of the interference of external and internal factors. PBC refers to the degree to which a person believes that he or she can perform a given behavior. PBC involves the perception of the individual’s own ability to perform the behavior. In other words, PBC is behavior- or goal-specific. That perception varies by environmental circumstances and the behavior involved). For attitudes, norms and PBC, beliefs are considered to be the primary structural units (Ajzen, 1985).

According to Ajzen (1991), the TPB, which is related to the individual’s information processing, bases their behavior on rational decisions. It offers a constrained method to predict intentions, which are considered to be the immediate precursor of any behavior. Attitudes, subjective norms, and perceived behavioral control are shown to be related to appropriate sets of salient behavioral, normative, and control beliefs about the behavior, but the exact nature of these relations is still uncertain (Ajzen, 1985). Other factors can only affect intentions indirectly by having an impact on attitudes, subjective norms, or PBC. Several meta-analytic studies have substantiated the predictive efficiency of both the TRA and the TPB across various behavioral areas (Rise et al., 2010).

3. Research Methodology

3.1. Research Model

Focusing on theoretical constructs, the Theory of Planned Behavior (TPB) is concerned with how individual motivational factors become determinants of the likelihood of performing a specific behavior (Ajzen, 1985). TPB, which focuses on the constructs of attitudes, subjective norms, and perceived control, explains a large proportion of the variance in behavioral intentions and predicts a number of different behaviors, including health behaviors, purchase behaviors, technology adoption, work behaviors, and continuous adoption (Ajzen, 1985).

3.1.1. Attitudes and Its Antecedents

Several past studies have uncovered that an individual’s needs and expectations can motivate the performance of a particular behavior. Motivation is the underlying reason why people choose to do certain behavior. Motivation can trigger employees to work harder to influence the company in achieving its goals (Pancasila et al., 2020). Motivations attached to jobs are of two types: intrinsic and extrinsic. The satisfaction and gratification derived from the work itself are known as intrinsic motivation, such as feedback, autonomy, and achievement. Extrinsic motivation is also the satisfaction and gratification derived from the job, although it is not related to the task itself, such as with pay and incentives. It has a significant positive effect on the performance of employees through employee engagement (Astuti et al., 2020).

During the COVID-19 pandemic, which forced companies to widely apply the work-from-home model, the employees were self-motivated to adopt remote working. However, after the virus outbreak ends, continuous behavior will require a motive to encourage employees to continue working from home. Therefore, continuing certain behaviors requires a motivational factor that is driven by an individual’s willingness to move towards gaining value or taking away from threats or discomfort. In the remote work context, Olson and Primps (1984) argued that remote working comes with a lack of visibility. Thus, most of the employees pursue autonomy provided by the remote setting to enhance their positive work attitude and job performance. Consequently, this study scrutinized the impact of autonomy as an intrinsic motivation on attitude, particularly in the context of remote working. Besides, it has been established that people who perceive the balance between their work and life roles are more gratified with their lives and also enjoy better health mentally and physically (Haar, 2013). Typically, it is observed that work-life balance manifests into an enhanced positive work attitude as well as productivity in performance. Several studies established that a positive work attitude is enhanced when there is a work-life balance (Beauregard & Henry, 2009). Thus, this study proposes work-life balance as a significant parameter that influences the attitude of employees regarding remote working. Consequently, autonomy and work-life balance, as intrinsic motivations, are proposed to be integrated with TPB as antecedents of attitudes to address the factors that influence the continuous adoption behavior of working from home after the COVID-19 outbreak.

3.1.2. Subjective Norms (SN) and Its Antecedents

The theory of reasoned action (TRA) (Ajzen & Fishbein, 1977) and TPB (Ajzen, 1985) purported that the influence of peers such as friends, family members, or colleagues at work, shapes one’s intentions towards a given behavior. Besides, the Social Norm Theory by Perkins and Berkowitz (1986) posits that an individual’s behavior is influenced by perceptions of how their peers evaluate their behaviors and to what extent their peers accept it. Moreover, the effects of social influence stem from the influence of loved and respected persons, such as family members and friends. In other words, individuals are influenceable and respond to the influences from their peers and superiors. While making a decision that may affect their future career path, most individuals try to take advice or seek support from their close social circles to verify the appropriateness of their decision. The received support by individuals in the community, such as family or friends, may help decrease anxiety levels. For the purpose of this study, two kinds of support will be addressed: (i) social support, which refers to the received support from the employee’s family and friends, and (ii) organizational support, which refers to the received support from the firm. Therefore, social support and organizational support are proposed to be integrated with TPB as antecedents of subjective norms for addressing the factors that influence the continuous adoption behavior of remote working after the COVID-19 outbreak.

3.1.3. Perceived Behavioral Control (PBC) and Its Components

Perceived behavioral control (PBC) relates to the individual’s evaluation of self-efficacy and perceived controllability in engaging in a behavior (Ajzen, 1991). These components are conceptually distinct in terms of their locus of control (Rhodes et al., 2006)). Self-efficacy is “the belief in one’s capabilities to organize and execute the courses of action required to manage prospective situations.” Self-efficacy is a person’s belief in his or her ability to succeed in a particular situation. Self-efficacy directs towards an internal locus of control as it emphasizes an individual’s skills and abilities, whereas controllability is based on an external locus of control over extraneous (e.g., organizational, technological) resources required to engage in that behavior (Terry & O’Leary, 1995). Terry and O’Leary (1995) uncovered that the items under self-efficacy strongly affected individuals’ intentions to practice sports repeatedly, although it was not reflected in their actual behavior. While controllability did not impact intention greatly, it predicted actual behavior to a great extent. Moreover, Rhodes and Courneya (2004) asserted that gauging Ajzen’s intended PBC subcomponents of perceived skills/ability, resources, and opportunities to engage in behavior may help in developing an enhanced model to predict behavior. Therefore, and for the purpose of this study, self-efficacy and controllability are proposed to be integrated with TPB to address the factors that influence the continuous adoption behaviour of working from home after the COVID-19 outbreak.

Based on the aforementioned discussion, the proposed model consists of nine constructs that this study posits to influence an employee’s continuous adoption intention of remote working after the COVID-19 outbreak. These constructs include attitude, subjective norms, controllability, self-efficacy, perceived organizational support, social support, autonomy, work-life balance, and continuous intention. This study aims to evaluate the validity of the hypothesized relationships embedded in the theoretical model and the robustness of the model in predicting the continuous adoption intention of working from home in Saudi Arabia after a pandemic. The theoretical model is graphically presented in Figure 1.

Figure 1: The Proposed Research Models

3.2. The Development of Hypotheses

3.2.1. Autonomy

Autonomy refers to the freedom, independence, and discretion a person is provided with while performing a task. In this study, autonomy is an important job characteristic that impacts the potential positive attitudes related to a job. Therefore, a positive relationship is expected between autonomy and a positive work attitude, since employees are benefitted from the freedom of choosing how they work and how much effort they put in. Therefore, in the context of remote working, higher job autonomy will most likely result in a positive attitude to the work-from-home model. In this context, this study proposes a hypothesis as follows:

H1: Job autonomy positively influences an employee’s work-from-home attitude.

3.2.2. Work-life Balance

Work-life balance is an extremely subjective feature, which may be quite different for each person. This way, one can view work-life balance as the practice of dividing various resources such as time and labor, among different aspects of life. In the case of remote work, the employees usually have more freedom to choose the appropriate working hours for them. It is widely believed that work flexibility provided by remote working benefits the employers and the employees. From an employee’s point of view, flexibility is supposed to allow individuals to fulfill work-related responsibilities while maintaining a satisfying personal life (Casey & Grzywacz, 2008). From the point of view of employers, work flexibility benefits the organization by increasing employee productivity, commitment and loyalty to the organization. Hence, remote employees can use this flexibility provided by remote working to balance their personal lives and job obligations. Therefore, a good work-life balance provided by remote work is beneficial for both employers and employees, and also leads to a positive work attitude.

H2: Work-life balance positively influences an employee’s work-from-home attitude.

3.2.3. Attitude

An attitude emerges from the combination of an individual’s past and present experiences. For the purpose of this study, an attitude refers to an employee’s positive or negative perspective towards working from home. The employees will be motivated to continuously work from home if they believe it will result in the same positive attitude which they experienced during the first adoption.

H3: Attitudes towards working from home positively influence an employee’s continuous adoption intention of working from home.

3.2.4. Perceived Organizational Support (POS)

POS is used to indicate the degree to which a firm values any contributions made by its employees, as well as how much it cares about them (Eisenberger et al., 1986). In this study, perceived organizational support is used to indicate the employees’ perceptions regarding the extent to which their firm values their contributions and looks after their well-being while they work from home. Several studies confirmed the positive relationship of organizational support on subjective norms (Iqbal et al. 2011). Accordingly, based on the above assumption and empirical support, this study postulates:

H4: Perceived organizational support positively affects the subjective norms.

3.2.5. Social Support

Social support is an individual’s perception, and actuality, that they are valued, cared for, and a part of a supportive social network. Social support theory was developed to explain how social relationships influence cognitions, emotions, and behaviors. Social support is primarily two dimensional in nature: emotional support and informational support Therefore, when social support exists through both emotional and informational support, people tend to feel more secure and less stressed during decision-making, which then leads to positive subjective norms towards a certain behavior. Accordingly, based on the above assumption and empirical support, this study postulates:

H5: Social support from family positively affects subjective norms.

3.2.6. Subjective Norms (SN)

Subjective norms refer to the social parameters that influence an individual’s performance of a specific behavior (Ajzen & Fishbein, 1977). In this study, subjective norms are taken as the degree to which social pressure influences the employee’s perception of adopting remote working.

H6: Subjective norms positively influence an employee’s continuous adoption intention of the work-from-home model.

3.2.7. Self-efficacy

Self-efficacy is looked at as the belief in the ability to engage in the behavior that is required to attain a goal (Bandura, 1977). In this study, perceived self-efficacy indicates the degree to which a person has the ability to organize and implement the courses of action required to work from home. In the post-adoption context, there are greater chances of self-efficacy impacting their continued intention as opposed to continued behavior. Therefore, if an employee, after initially adopting the work-from-home model, believes that they are less capable of working remotely, they will be against working from home. Based on this, the study proposes this hypothesis:

H7: Self-efficacy positively influences an employee’s continuous adoption intention of working from home.

3.2.8. Continuance Adoption Intention

Continuous adoption intention refers to the extent to which an individual intends to accept doing a certain course of action consistently. In this study, continuous adoption intention refers to an individual’s intention to continue working from home. In the post-acceptance context, continuing remote working is also a rational choice, driven by instrumentality as well as other considerations, and thus, the same extension is relevant in the context of remote work continuance. A person will more likely continue to adopt remote work if they have positive continuous intentions towards the same behavior.

H8: Continuance adoption intention is positively affected by the continuous adoption behavior of working from home.

3.2.9. Controllability

Controllability, or facilitating conditions, is defined as the extent to which a person believes that organizational and technical resources are available. In the post-adoption context, employees may not be aware of the external locus prior to their continuance choice, or it may evolve before the behavior of working from home develops, and therefore influence their actual continuous behavior rather than continuance intention. Therefore, if the employees have control over these resources, they may choose to adopt remote work in a continuous manner. Many studies have uncovered a significant relationship between controllability and continuous behavior. Based on this context, this study proposes a hypothesis as follows:

H9: Controllability is positively affected by the continuous adoption behaviour of working from home.

3.3. Data Collection and Analysis

The research is descriptive in nature where data collection is based on primary data collection done through online surveys (i.e., google forms). The population of this study comprised employees from several companies in Saudi Arabia. Sampling profiles included employees working in various fields who are working from home in different capacities. Samples have been derived using snowball sampling depending upon the nature of the study and specific constructs (i.e., working from home). As 300 responses have been received, the sample size for this study is 300 (n = 300). The demographic information regarding the respondents has been analyzed on MS Excel, whereas data analysis has been carried out using SmartPLS 3.

4. Results

4.1. Demographic Analysis

Data has been collected from a diverse range of Saudi employees who differ based on gender, age, marital status, and qualifications. The detailed analysis revealed that respondent profiles comprised 62.33% males and 37.67% females. Respondents belonged to different age brackets (from 18 years to above 55 years). 9.67% belonged to the age bracket ranging from 18 to 24 years. 26% belonged to 35 to 44 years, and 9% belonged to the bracket ranging from 45 to 55 years. The fewest number of respondents belonged to above 55 years of age, and the largest number was from 25 to 35 years (2.33% and 53% respectively). Upon inquiring about the marital status, the largest portion of the respondents was single at 64.33%, with married ones at 28.67%. 4% of the respondents were revealed to be divorced, and 3% of them selected “Other”. The higher number of respondents possessed a master’s degree with 52.67%, followed by diploma holders with 24.67%. Only 17% revealed to have a bachelor’s degree, whereas 5.67% reported having a doctorate level qualification. The demographic profile of respondents is presented in Table 1.

Table 1: Demographic Profile

4.2. Evaluation of Measurement Model

The measurement model has been evaluated as per the recommendation of Hair et al. (2019) through construct reliability, convergent validity, and discriminant validity. The ideal values for item loadings are reported to be 0.708 or above. However, the acceptable range is 0.5 and above (Hair et al., 2019). All the items in the study exhibit ideal item loadings, whereas three of them are below 0.708. Out of these three, two are still in the acceptable range, i.e., 0.50 (ATT1, WLB1), whereas WLB5 has been removed from the analysis due to loadings of <0.50. The internal consistency of constructs has been reported by ρ_A. According to Hair et al. (2019), ρ_A is the most appropriate measure of internal consistency of constructs, keeping into account the “too conservative” and “too liberal” nature of Cronbach’s Alpha and Composite reliability. The variance among constructs has been measured by the average variance extracted (AVE), which bears the cut-off value of 0.5. The assessment of the measurement model depicts that all the values are above the threshold, demonstrating more than 50% variance among the constructs under study. The extremely high value (1) of continuous behavior to adopt WFH (CBA) is due to the single item scale used to measure the construct (Table 2).

Table 2: Construct Reliability and Validity

The individuality of multiple constructs under study has been ensured by measuring the discriminant validity through the Fornell-Larcker criterion (Table 3).

Table 3: Inter-Construct Correlation

Besides this, Hair et al. (2019) suggest the heterotraitmonotrait ratio (being a less controlled approach) to be used for the said purpose in PLS. The cut-off value for the HTMT ratio is 0.85 for conceptually different constructs (Table 4).

Table 4: HTMT Ratio

Before assessing the structural relationships, the variance inflation factor (VIF) has been evaluated to check for any collinearity issues (Figure 2 and Table 5). The ideal VIF value for a reflective measurement model has been suggested to be ≤10 (Hair et al., 2019).

Figure 2: Measurement Model

Table 5: Collinearity Statistics

4.3. Evaluation of Structural Model

The nature and magnitude of hypothesized relationships have been measured through the evaluation of the structural model, specifically using PLS-SEM. A summary of the results has been presented in Table 6 and Figure 3. The statistical significance of the results has been evaluated in terms of t-values (>1.64) and p-values (<0.05).

Table 6: Hypothesis Testing

Figure 3: Structural Model

The path coefficients demonstrated the extent and magnitude of relationships projected by the proposed hypothesis (Table 6).

The overall impact on endogenous constructs is also demonstrated in Table 6 by the values of R².

5. Discussion

Considering the plethora of research illuminating the consequences of COVID-19 in all walks of life, this current study highlighted the continuous adoption of working from home regarding the sample taken from Saudi companies. The employees’ attitudes towards working from home (WFH) are measured in terms of job autonomy and work-life balance. Depending on the benefits of remote working, such as flexible working hours and freedom to work in selected hours (Casey & Grzywacz, 2008), working from home also aids in maintaining the work-life balance of employees. Such flexibility allows the employees to develop a positive attitude towards working from home in the long run. Perceived organizational support and social support from family have been measured as antecedents of subjective norms towards working from home. In addition to a positive attitude and subjective norms, self-efficacy has also been evaluated as an antecedent for employees’ intention to adopt work from home. If employees, after initially adopting work-from-home, believe that they are less capable of working remotely, they will be against the said working modus. The continuous adoption of work from home has been measured in terms of employees’ relevant intention and controllability. This implies that employees possessing positive intentions towards continuous remote working, will more likely to opt for the same behavior.

The present study found the impact of controllability on employees’ continuous adoption of working from home to be insignificant. Controllability, also termed as “facilitating conditions”. Therefore, Lu et al. (2005) stated that “non-technical external controls” such as assistance and direction should be provided to users to completely embrace the technological shift, such as working from home. The findings of this study illuminate the factors leading to the continuous adoption of WFH. The upshots motivate the human resource managers to facilitate their employees, especially in the tough times of pandemic. For the said purpose, employees have to be encouraged for their jobs done from home, through relevant mentoring, training and promoting knowledge acquisition. Promoting a healthy work-life balance is the key to provide social support to the employees. Managers are suggested to put a great deal in providing supportive and productive work where employees can flexibly adapt to working from home.

From the academic perspective, it demonstrates noteworthy empirical evidence inferences regarding the continuous adoption of working from home in the context of COVID-19. It attempts to measure the employees’ post adoption of working from home in light of the eminent theories in social sciences (i.e., the theory of planned behavior and theory of reasoned actions). From a practical viewpoint, the paper presents implications to businesses worldwide and HR managers in particular. The study recommends that HR managers prioritize providing a supportive and constructive work environment employees’ so that they are flexible to adopt working from home, for greater benefits overall.

6. Conclusion

The employees’ positive attitude towards working from home is measured in terms of job autonomy and work-life balance. Job autonomy leads to a positive work attitude during working from home. Perceived organizational support and social support have been measured as antecedents of subjective norms towards working from home. In addition to a -positive attitude and subjective norms, self-efficacy has also been evaluated as an antecedent for employees’ intention to adopt work from home. The continuous adoption of work from home has been measured in terms of employees’ relevant intention and controllability. The present study found an insignificant impact of controllability on employees’ continuous adoption of working from home.