The Role of Employee Performance During Covid-19: Do Motivation and Commitment Matter?

Purpose – This study tries to analyze the relationship between job satisfaction and employee performance with the mediating variables of work motivation and organizational commitment in during the Covid-19 pandemic. Methodology – Structural Equation Model was built to capture the relationship between variables using purposive sampling, with the sample criteria being lecturers who have worked in both public and private tertiary institutions with a minimum work period of 2 years. When the questionnaire was distributed, the respondent was working through working-from-home scheme. In order to fulfil requirement from SEM model, 150 respondents were taken, and 128 questionnaire that can be used in estimation process. Findings – The results showed that work motivation and organizational commitment had no effect on employee performance, nor were they mediating variables on job satisfaction. This weak relationship proves that the pandemic is a difficult time for workers. Furthermore, satisfaction from a well-completed job is objectively beneficial, but the feeling appears confined to a task without high motivation and commitment. Originality – In case of pandemic time, Work motivation and Organizational commitment fail as a Mediating effect that connects job satisfaction to employee performance. Here’s proves that negative impact of the working-from-home policy is unavoidable .


Introduction
The Covid-19 virus pandemic has forced people worldwide to adapt to new habits. This new habit was required to prevent the virus from becoming more widespread. Likewise, the Indonesian government has already set several policies, including large-scale social restrictions, such as working-from-home (WFH) and school-from-home (SFH) for workers and students. This policy is a strategic step taken by the government to limit peoples' movement in order to prevent the spread of the Coronavirus, but not everyone is able to adapt to this new habit. The combination of knowledge and attitude becomes the most crucial adaptability factor (Reuben et al., 2020) Working-from-home for workers entails using online communication networks (International Labour Organization, 2020). It is difficult in some areas of work. Various pros and cons have emerged due to this policy (Rokhani, 2020). However, several obstacles occur during working-from-home. These include the fact that not all work can be done at home, inadequate facilities when working, and limited communication and coordination when using online media (Mustajab et al., 2020). Additionally, not all organizations are ready for the working-from-home system.
The working-from-home concept started from balancing work and personal life in early 1981 (Crosbie & Moore, 2004). An individual in a certain job position will adopt a leisure seeker lifestyle in order to maintain their job performance. Work-life balance alludes to individuals investing adequate energy in their positions while also devoting more time to other interests, such as family, leisure activities, and network inclusion (Carolinda & Hyronimus, 2020;Smith et al., 2008Smith et al., , 2016. The significant difference in the working-from-home pattern is in determining the working hour, which is usually set by the company. However, workers must make their working hours adjustments.
During the pandemic, where all aspects of life are uncertain, the economy is weak, and the spectre of a recession looms, the working-from-home policy has implications, especially in the motivation of workers to complete their work. Here, working from home during Covid-19 is a condition that must be done not by choice but by necessity. It raises two possible conditions, such as increasing motivation which will affect worker productivity on the comfort of working at home. On the other hand, it eliminates motivation due to several obstacles faced when work is done at home.
Within educational institutions, face-to-face lectures were replaced by online lectures, and campus-based coordination works are now required to be completed online. Given the dual role of lecturers as both lecturers and executors of campus institutions, there is a burden to ensure the achievement of learning targets even in working-from-home conditions while coordinating with other work units. Lecturers have more time at home, thereby increasing work-life balance, but it is quite difficult to maintain working performance (Palumbo, 2020;Palumbo et al., 2020). As part of the organization, lecturers are also burdened with performance achievements. However, given the current conditions, achieving the established targets is certainly not easy. Employee performance is measured based on the standards or criteria set by the management company, which primarily intends to improve overall employee performance. Even though working-from-home also gains a spillover effect, its positive effect also remains marginal (Veld et al., 2016).
Various methods are used to improve performance, one of which is the creation of a comfortable work environment for employees. According to Fluegge (2008), the creation of job satisfaction is one of the factors that influence performance. Job satisfaction is a person's emotional response to their jobs, wages or salaries, promotion opportunities, supervision, and coworkers' relationships. Job satisfaction is a major predictor of employee performance Landis et al., (2015). Satisfied employees tend to respond to their positive experiences by talking positively about the organization, helping colleagues, and exceeding expectations in completing their work. Studies from Hartono & Nurwati (2021;Iqbal et al., (2021); Sitohang & Budiono, (2021);Syardiansah et al., (2020) ;Wahjoedi, (2021) show that job satisfaction has a positive and significant effect on employee performance. Meanwhile, Gunawan et al., (2022); Harianto et al., (2020); Husein & Hanifah, (2019) show that job satisfaction does not significantly affect employee performance.
In order to support high employee performance, job satisfaction and organizational commitment need to get the main attention (Luthans, 2011). Devyanti & Satrya, (2020); Dharmayanti et al., (2020); Saputra et al., (2019) showed that there is a significant positive relationship between job satisfaction and organizational commitment. The higher the level of satisfaction in the work environment, the more employee commitment. On the other hand, motivation is the main driver in improving employee performance. According to several previous studies, motivation's impact on employee performance differs for each case (Christian et al., 2020). In high-pressure organizational conditions, motivation has no significant impact on employee performance but has a significant impact in the opposite group. Therefore, workingfrom-home implementation in an organization needs to be evaluated regularly, especially in a firsttime implementation.
It is important to remember that the variables above were conditions when there was no pandemic, indicating that the relationship between variables is still a condition that occurs in a state without coercion. Therefore, this study aims to examine the relationship between job satisfaction and employee performance mediating by motivation and commitment.
The measurement of the influence of the independent variables, namely job satisfaction (X1), work motivation (Y1), and Organizational Commitment (Y2), on the dependent variable, namely Employee Performance (Y3), was carried out with the framework shown in Figure 1.

Research Methods
The study objects include 197.380 lecturers at universities in Indonesia. The Structural Equation Modeling method requires an appropriate sample size ranging from 100-200 respondents (Hair, 2010), therefore, 150 were selected for this study. The purposive sampling technique was used, with the sample criteria being lecturers with a minimum work period of 2 years in both public and private tertiary institutions. Only 128 out of 150 questionnaires were filled out. Furthermore, most of the respondents were women (61.719%) with master (73.42) and doctoral degrees (26.58%). The remaining 32.81% were male respondents, with 57.14% master and 42.86% doctoral degrees. For most respondents, accounting for 79.69%, working from home was their first experience, and 20.31% stated that they had worked from home before the pandemic.

Results and Discussions
The study was conducted using Structural Equation Modelling (SEM), and all estimates were conducted by the AMOS 23.0 program (Arbuckle, 2003). The first step is to calculate the validity of the structure, including confirmatory factor analysis, and the second is to analyze the convergence of the theoretical model and determine its validity.

Confirmatory Factor Analysis
The initial theoretical model was tested to check for the fitness of the data. It was also used to determine whether necessary modifications are required to adjust the theoretical model to fit the data better. The result of the confirmatory factor analysis is shown in Table 1.

Validity Reliability and Confirmatory Factor Analysis (CFA)
The Corrected Item-Total Correlation and Cronbach Alpha results show that all indicators are valid and reliable. The first estimation CFA for the exogen factor shows Chi-Square (χ2): 22.610, df=5, and Significant Probability (≥0.05): 0.000. To regain the goodness of fit of the model, indicators that do not meet the criteria were excluded, such as the smallest Loading Factor (A3), and then recalculate. After the exclusion, the model feasibility value shows a fit result, which is indicated by the Chi-Square value (χ2): 3.579, df=2 Significant Probability (≥0.05): 0.167. the values for CFA endogen are Chi-Square (χ2): 167.824, df=101, and Significant Probability (≥0.05): 0.000. For CFA endogenous, the smallest loading factor was excluded, namely C1.

Full Model Structural
The first estimation for the full model structure shows a Significant Probability greater than (< 0.05). Therefore, the Modification Indices Suggestion was adopted. It provides modification suggestions to obtain a more fitted model. This suggestion will provide a new path by connecting covariations errors that are still in the same or different latent factors. The suggestions for modifying models are shown in Table 2.  Table 2, the covariant line used are e1 -e6, e1-e2, e1-e3, e3-e6, e4 -e6, e25 -e20, and e22-e20. The final full model structure after adding modification indices shown in Figure 2. Modification indices in structural equation modeling was used to identify specific modifications to the model that could improve its fit by reducing the discrepancy between the observed and expected covariance matrices. As in table 2, after potential modifications are identified, we can decide whether to revise the model based on theoretical grounds, statistical considerations, or substantive implications. As shows in Figure 2, after making modifications, it is necessary to re-estimate the model and evaluate the revised model's fit indices to ensure the improvements have been achieved. Feasibility assessment of the full model after modification indices is shown in Table 3.  Table 3. Improved results after adding the Modification Index to the full model are shown from CR values above 70% and AVE > 50% in all constructs used in this study. Furthermore, the evaluation of the Goodness of Fit Testing Model on the Full Structural Model with modification Index is shown in Table 4. Evaluation of the use of the modification index for the full model can be seen in Table 5. After the full model was repaired using the modification index, the Conformity Index results for the full model showed that the value of each Conformity Index was higher than the Cut-off Value. Significant improvement can be seen from the Significant Probability value which has reached greater than 0.05 (0.089) so that the model is declared Fit, as well as the results of other Conformity Indexes, so that the model can be concluded as feasible to use. The result of regression estimation of the structural model is presented in Table 5. The first hypothesis test shows that Job satisfaction has a positive and significant effect on employee performance. Table 5 shows a positive coefficient, Critical Ratio (CR) value, and a probability of 0.308, 2.365 and 0.018, respectively. Both of these values qualify to accept H1 because the CR value is 2.365, greater than 1.96, with a probability of 0.018, which is less than 0.05. Therefore, it can be concluded that hypothesis 1 is supported and job satisfaction positively and significantly affects employee performance. The second hypothesis, which is Job satisfaction has a positive and significant effect on work motivation, is accepted. Table 5 shows a CR value and probability of 7.241 and 0.0000, respectively. These two values qualify to accept H2 because the CR value of 7.241 is greater than 1.96, with a probability of 0.000, which is less than 0.05. The positive value from the coefficient (0.559) also confirms the hypothesis. Therefore, it can be concluded that job satisfaction positively and significantly affects work motivation.
The third hypothesis, which is job satisfaction has a positive and significant effect on organizational commitment, is rejected. Table 5 shows a CR value of -and probability of 0.178 and 0.859, respectively. Both values do not meet the requirements to accept H3 because the CR value is 0.178, which is less than 1.96, the probability of 0.859 is greater than 0.05, and the coefficient also shows a negative sign. This result means job satisfaction does not positively and significantly affect organizational commitment. The fourth hypothesis, that work motivation has a positive and significant effect on lecturer performance, is rejected. Table 5 shows a CR value and probability of 0.411 and 0.681, respectively. Both values do not meet the requirements to accept H4 because the CR value of 0.411 is smaller than 1.96, and the probability of 0.681 is greater than 0.05. Therefore, it can be concluded that work motivation has no positive and significant effect on lecturer performance. Finally, the fifth hypothesis shows that Organizational commitment has a positive and significant effect on employee performance. This is evident in the CR value and probability of 2.963 and 0.0003, respectively. Both of these values qualify to accept H5 because the CR value of 2.963 is greater than 1.96, and the probability of 0.000 is less than 0.003. Therefore, it can be concluded that organizational commitment positively and significantly affect lecturer performance.
The results of statistical tests prove that job satisfaction has a positive and significant effect on the performance of lecturers who work on a working-from-home basis. Job satisfaction is defined as the sensation of enjoying one's work in a working-from-home manner. The perceived satisfaction is related to the salary received and a pleasant boss. It is also related to a productive work results environment that supports lecturers' performance in completing their work. Therefore, the lecturers' job satisfaction is proven to have increased their performance. This result consistent with research from & Budiono, 2021), Hartono & Nurwati (2021); Iqbal et al. (2021); Sitohang & Budiono (2021); Syardiansah et al. (2020) and Wahjoedi (2021).
The statistical test results proved that job satisfaction has a positive and significant effect on the work motivation of lecturers who worked from home. The lecturers' job satisfaction is proven to have increased their work motivation.
The statistical test results proved that job satisfaction has no positive and significant effect on organizational commitment. This can happen because organizational commitment is interpreted as feeling proud and emotionally attached to the organization. Due to lecturers' restrictions, including their inability to meet students in person, verbal and non-verbal communication cannot be carried out efficiently, preventing emotional attachment. This also limits lecturers in transferring both implicit and explicit knowledge.
The statistical test results proved that work motivation does not affect lecturer performance..Meanwhile,.organizational commitment has a positive effect on lecturer performance..This indicates that the lecturers who work from home are emotionally attached to the institution..There is also a sense of pride and apprehension about leaving the institution. Furthermore, the benefits obtained from the institution will increase the lecturers' performance.

Mediating Effect
Work motivation and Organizational commitment are mediating variables between Job satisfaction and employee performance. The estimation results of the relationship were carried out using the Sobel test, and the results are shown in Table 6. Work motivation plays a crucial role in connecting job satisfaction to employee performance, as research by Forson et al., (2021) that found motivation is a meaningful construct that defines employee satisfaction at the workplace. However, this research found that work motivation fails to act as a mediation variable that connects job satisfaction with the performance of lecturers who works from home. This research only found direct effect from job satisfaction towards work motivation, but work motivation has no significance relation towards employee performance. This finding is in line with research by Malik et al., (2022), where changes in the work environment that occurred during the pandemic also caused workers to experience loss of motivation, which was caused by a lack of knowledge about prevention and protection was associated with lower motivation to work. In order to enhance employee motivation, in particular in pandemic situation, extrinsic motivation must be focused more (Nilasari et al., 2021).
In this research, organizational commitment also fails acts as a mediation variable that connects job satisfaction with the performance of lecturers who work in working-from-home. Significance direct effect found in relation between organizational commitment towards employee performance. However there is no direct effect in relation between job satisfaction no organizational commitment, this relation in line with result from Sutapa et al., (2022). In Sutapa et al., (2022) shows that female staff with positions as division heads who work from home, are more loyal to institutions, have higher improvement efforts, and perform better work efficiency than male staff. Also research from Anugrah & Priyambodo, (2022) which found that there was a significant positive relationship between organizational commitment and the performance of employees in Surabaya who worked from home during the Covid-19.

Conclusions
The Covid-19 pandemic significantly affected people's lifestyles. During this study, Covid-19 was still spreading, and the number of patients increased daily. Some employees have become accustomed to working from home, but many lecturers have experienced this for the first time. In its implementation, several obstacles to working from home were found, as follows: (1) Inadequate facilities, (2) limited communication and coordination, and (3) some types of work cannot be done optimally..Working from home affects the job motivation and organizational commitment of lecturers in completing their tasks.
The organizational commitment variable, hypothesized as a mediating variable between job satisfaction and lecturer performance, was not proven. The relationship between job satisfaction and organizational commitment has no significant positive effect. However, the relationship between organizational commitment and lecturer performance has a significant positive effect. In the case of working from home due to the pandemic, organizational commitment is not strong enough to encourage lecturer performance. The government has mandated that all lecturers must be conducted through a working-from-home scheme. Institutions must pay attention to lecturers' job satisfaction to improve their performance. These include satisfaction with salaries, environment, and the job itself. Further studies are recommended to analyze the condition of workers after the new normal period, where some still work from home. In addition, How the employees adapt to the new normal condition also needs further investigation.
Further research can explore the inclusion of additional variables that act as mediators between work motivation and employee performance, such as variables related to creativity. Additionally, independent variables like transformational leadership could be incorporated to examine their impact on self-efficacy. Furthermore, moderator variables like job levels can be introduced to investigate their moderating effects on the relationship between organizational commitment and employee performance.