Underpinning theory

Stress at the job because of the coronavirus pandemic

Stress is the mental or physical exhaustion caused by life circumstances and leads to reducing people’s life quality rapidly. Stress depends on socioeconomic circumstances, habitat, genetics or innate psychology6.

In fact, work stress is a big problem that businesses and organizations face in developed countries7. Nelson and Burke defined work stress is the emotional reaction of employees to the working environment at the organization due to the overload of work, pressures from the working environment, incorrect work assignments, etc 8. In addition, work stress will appear if employees are affected by a dangerous situation in the work process, threatening them and forcing them to find ways to adapt and respond to get rid of those negative emotions9. Petchsawang and Duchon affirmed that there are three main factors that lead to stress at work: environmental factors, organizational factors and employees themselves 10.

In the current complicated coronavirus epidemic, stress at the job due to the coronavirus has directly affected the attitudes and behaviors of employees in an organization. In other words, it is the attitude of stress caused by the pandemic factor. According to Hobfoll et al., war, terrorism, natural disasters and epidemics threaten human life, leading to anxiety, insecurity, and severe stress in life and work 11. The heavy consequence of the coronavirus epidemic has attracted domestic as well as foreign scholars to study work stress12, 13.

Satisfaction at the job

Satisfaction at the job is a workers’ positive or negative emotional status based on their work experiences 14. If employees’ job expectations are met, they will be satisfied.

Friday and Friday argued that employee satisfaction can be approached and assessed in two ways: general satisfaction and satisfaction with job dimensions 15. Each approach has its different advantages. However, in this study, job satisfaction was approached in a overall manner, and job satisfaction was studied based on the emotional or perceived of employees towards the job.

Commitment to the organization

Mowday and partners indicated commitment to the organization as a worker’s behavior or attitude toward an organization. Specifically, employees with a positive attitude are more likely to make them put in more effort, be willing to contribute to the organization, promote the implementation of the organization’s common goals and turn thoughts and words into actions to achieve the highest work efficiency. They also emphasized that organizational commitment made up of three aspects: identification, involvement and loyalty16.

Based on the study of Mowday and partners, Tran defined commitment to the organization as the strength of employee acceptance as well as involvement in one. Additionally, commitment to the organization is also constituted through 3 aspects: effort, loyalty and pride 17.

Like job satisfaction, organizational commitment is also approached in two directions. However, this study only looked at the overall organizational commitment based on the loyalty and trust towards the organization.

Intention to quit the job

Intention to quit the job describes an employee’s consideration of leaving the job at the organization 18, or it is the surest prediction of the employee’s impending actions based on the theory of planned behavior (TPB) of Ajzen19.

Barry et al. emphasized that the intention to quit is the proactive will of the employee after careful consideration of leaving the current organization20. The intention to quit is scattered in two directions: involuntary and voluntary21. However, this study only looked at the voluntary aspect of employees, which is the negative psychological state that makes employees appear willing to leave their jobs or organizations.

Hypothesis development and research model

Satisfaction at the job is inversely influenced by work stress22. Kang and Singh highlighted that work stress as a direct determinant of job satisfaction23. The more work stress increases, the more satisfaction decreases and vice versa. More importantly, the stress related to the medical workforce’s health and lives due to the coronavirus epidemic has become even more severe24. Due to the work characteristics of the healthcare workforce that must be in direct contact with coronavirus patients, the high risk of disease transmission build up work stress and lead to job dissatisfaction. Hence, the first hypothesis is recommended as follows:

H1: Work stress due to the COVID-19 pandemic will be negatively associated with satisfaction at the job.

Chan et al. found that job stress has an inverse impact on commitment to the organization25. Workers are more committed to the organization if they feel comfortable in their job or if the work stress is reduced to a minimum. Conversely, increased work stress signals that employees have less commitment with the organization26. Along with the rapid expansion of the coronavirus pandemic, it severely influences the psychology of the healthcare workforce, leading to depression, fatigue, and exhaustion, making work stress more apparent and reducing organizational commitment13. Thus, the second hypothesis is recommended as follows:

H2: Work stress due to the COVID-19 pandemic will be negatively associated with organizational commitment.

Liu et al. argued work stress is positively associated with the intention to quit the job 27. More specifically, employees are more likely to quit if the level of work stress increases. In fact, healthcare workers who intend to quit or get a new job because of the psychological sequelae created by the coronavirus epidemic. The work of Aguiar-Quintana et al. affirmed that work stress because of the coronavirus epidemic is a core reason why workers intend to quit28. So, the third hypothesis is is proposed as follows:

H3: Stress at the job because of the coronavirus pandemic will be positively associated to quit.

Job satisfaction is a prerequisite for organizational commitment29. Tran agreed with the above opinion and confirmed that organizational commitment is positively affected by employee satisfaction17. More recently, Nguyen and Uong have found similar findings to the aforementioned studies30. Hence, the fourth hypothesis is recommended as follows:

H4: Satisfaction at the job will be positively associated with commitment to the organization.

The study by Curivan indicated employees tend to intend to quit if they are unable to obtain job satisfaction or dissatisfied with their current job31. Thus, job satisfaction is one of the crucial factors that directly affects the intention to quit, but negatively. This finding is unraveled in the works of Vu and Nguyen4 as well as Nguyen and Uong32. So, the fifth hypothesis is recommended as follows:

H5: Satisfaction at the job will be inversely associated to quit.

Martin and Roodt affirmed in their research that organizational commitment and intention to quit have a negative nexus29. In other words, at that time, employees did not want to continue working in the organization because they felt that they themselves don’t fit in and involuntarily and actively participated in the affairs of the organization anymore. Following the above results, Nguyen and Uong explored the intention to quit is directly affected by organizational commitment, but negatively32. The sixth hypothesis is recommended as follows:

H6: Organizational commitment will be inversely associated to quit.

Based on the assumptions, we build a study pattern as follow (see Figure 1):

Constructs development

The first table present the propose scales inherited from diverse studies (see Table 1).

In order for the preliminary scale in the proposed research model to meet practical requirements in the healthcare sector in Vietnam, the authors conducted a group discussion with 20 healthcare workforce and 10 managers at public hospitals in Hanoi. Additionally, the authors carried out in-depth interviews with 05 human resource management experts to adjust the variables observed in the preliminary scale, considering the connection among research constructs in the proposed study pattern and solving problems during group discussions. The discussions and interviews took place online on Microsoft Teams platform in August 2022.

The qualitative study results showed that participants agreed with the constructs and relationships in the proposed research model. On the one hand, the “job satisfaction” and “organizational commitment” scales are similar to the original scales.

Furthermore, the scale of “work stress due to COVID-19 pandemic” suggests that an observational variable should be added to be consistent with the work practices faced by healthcare workforce in the duration of the coronavirus epidemic. Furthermore, the scale of “intention to quit” has an inverse observation variable. Therefore, experts believe that it is necessary to reverse it to avoid causing complexity in the survey process.

In addition, the authors calibrated some words to make the scale easier to understand and more suitable for the cultural background of Vietnamese people. The official scale is presented in Table 2.

Sampling and Data collection

The official scale uses a 5-point Likert scale with questions scaled from “strongly disagree” to “strongly agree”. A preliminary survey of the official scale involving 30 healthcare workforce working at Hanoi’s public hospitals was conducted. The survey outcomes revealed the scales reached reliability, as Cronbach’s Alpha exceeded 0.637.

To meet the reliability in the Exploratory Factor Analysis (EFA), we chose 345 survey forms and obtained 339 valuable survey forms, which ensured that the analyzed sample size is more than the minimum sample size (minimum sample size: 24 * 10 = 240), which is typically 5 or 10 times the number of manifest variables38.

We used a snowball sampling method to obtain data of the healthcare workforce at Hanoi’s public hospitals. The surveys are directly sent to healthcare workforce from October to November, 2022.

The research data were analysed through Cronbach’s Alpha, EFA, CFA and SEM to check the good-fit of the study pattern.

The characteristics of healthcare workforce at Hanoi’s public hospitals participating in the survey were 67.7% male. Among them, 87.5% of the healthcare workers were over the threshold of thirty five (middle-aged), 100% had higher education and 86.3% got married.

Reliability, convergent and discriminant validity

The factor loadings exceeded the threshold of 0.5 39. Therefore, the observed variables are reliable (see Table 2).

Table 2 reveals that Cronbach’s α values exceeded 0.6, so the scales achieved homogeneity reliability40. In addition, Table 2 also provided evidence they had convergent validity since the average variance extracted index exceeded 0.5, and the composite reliability index exceeded 0.7 39. Hence, the scales were suitable for further analysis.

In addition, the study found that the HTMT ratio did not exceed the threshold of 0.8541. Thus, the scales reached discriminant validity (Table 3). It proves that the results support the existence of discriminant validity for all tested constructs.

Structural Equation Model

The model fit statistics for CFA (Figure 2) was χ/df = 2.352; p = 0.000; TLI = 0.909; GFI = 0.903; CFI = 0.917; and RMSEA = 0.057. The revision index is notably low, and thus there is no scope for further improvement. The index χ/df ratio is below the threshold of 3.0. Our TLI, GFI, and CFI, which have the values of 0.909, 0.903, and 0.917 are the best since it has to be over threshold 0.9 and RMSEA is below the threshold 0.08 for the model to be assessed as a good fit 39.

The model fit statistics for SEM were χ/df = 2.563; p = 0.000; TLI = 0.913; GFI = 0.906; CFI = 0.920; and RMSEA = 0.063. The revision index reveals no scope of improvement. The index χ/df ratio is below the threshold 3.0. Our TLI, GFI, and CFI, which have the values of 0.913, 0.906, and 0.920 are the best since it has to be over threshold 0.9. RMSEA is below the threshold of 0.08, indicating that the model is a good fit 39.

Futhermore, the p-value of the correlation between the constructs does not exceed 0.05, indicating that the SEM has statistical significance 39. Table 4 summarizes the hypothesis test outcomes.

Discussion

The analysis results presented in Figure 3 and Table 4 showed that “work stress due to the COVID-19 pandemic” has a negative impact on organizational commitment and job satisfaction of the healthcare workforce at Hanoi’s public hospitals (β = -0.55 and β = -0.34) while it has a positive association with the intention to quit with β = 0.07. In addition, employee satisfaction also plays a crucial role in the organizational commitment of the healthcare workforce (β = 0.46). Moreover, commitment to the organization and satisfaction at the job have a negative effect on the intention to quit of the healthcare workforce (β = -0.07 and β = -0.63). If they have commitment and satisfaction, they are less likely to quit. Thus, the hypotheses H1 to H6 are accepted. These findings are consistent with the results of the study of Saleem et al. 24, Truong et al. 12, Aguiar-Quintana et al. 28, Nguyen & Uong30, Nguyen & Uong32. However, the extent of the influence of the factors was not the same when studying with healthcare workforce at Hanoi’s public hospitals.

To specifically identify each factor of the impact model for the intention to quit of healthcare workforce at Hanoi’s public hospital, we present 6 important results as follows:

Firstly, H1 hypothesis states that work stress due to the COVID-19 pandemic has a negative impact on job satisfaction (β = -0.55, p < 0.05). This result is consistent with research by Saleem et al. 24. This means that in the context of the COVID-19 pandemic, if the healthcare workforce has a lot of work stress, job satisfaction tends to decrease significantly.

Next, to investigate H2 hypothesis, it was identified that work stress due to the COVID-19 pandemic has a negative impact on organizational commitment (β = -0.34, p < 0.05). This finding is consistent with the conclusions of Truong et al. 13. Nevertheless, this result is inconsistent with Bhatti et al. (2016), who argued that work stress due to the COVID-19 pandemic has a higher impact (β = -0.223, p < 0.05) on the organizational commitment of the employee.

Regarding the H3 hypothesis, work stress due to the COVID-19 pandemic is likely to increase but not significantly affect the intention to quit of healthcare workforce (β = 0.07, p < 0.05). This is consistent with research by Aguiar-Quintana et al.28) that work stress due to the COVID-19 pandemic will become a crucial contributing factor to increased intention to quit the healthcare workforce. This result also indicates that if Hanoi’s public hospitals want to reduce the employee turnover, they must give them less pressure in work. Nevertheless, this result contradicts Liu et al. 27 that work stress has a higher impact on intentions to quit (β = 0.0944, p < 0.05) of the healthcare workforce.

Related to the H4 hypothesis, job satisfaction has a positive impact on organizational commitment (β = 0.46, p < 0.05). This result supports Nguyen & Uong’s study30. If the healthcare workforce has high satisfaction, they are more likely to be committed to the hospital.

Continuing with the H5 hypothesis, job satisfaction is likely to significantly reduce intentions to quit of healthcare workforce (β = -0.07, p< 0.05). This assertion is supported by Nguyen & Uong32. That means the higher the employee satisfaction, the lower the intention to quit.

Finally, the H6 hypothesis suggests that organizational commitment is also likely to reduce the intention to quit of healthcare workforce at Hanoi’s public hospitals but with a lower impact than job satisfaction (β = -0.63, p < 0.05). This finding has similarities to Nguyen & Uong32.

In conclusion, our research differs in the study context as well as the location compared to Nguyen and Nguyen3, Vu and Nguyen4, and Wen et al. 5 since it is studied in the context of Hanoi’s public hospitals of Vietnam.

Moreover, the highlight of this study is the research methodology used. The snowball sampling method and structural equation modeling have made important contributions to the unique emphasis on research methodology compared to the previous studies. In which, The snowball sampling method correctly identified health care workers who showed signs of stress due to COVID-19, leading to the intention to quit at Hanoi’s public hospitals. Given the delicate nature of the research topic on the intention of healthcare workers to quit, the snowball sampling method is a goodfit in this context as colleagues can recommend and willing to share their thoughts.