1. Introduction
The COVID-19 pandemic caused sudden changes in work organisation (e.g., working from home, virtual teamwork) and affected workers (e.g., through social distancing, stress, and unemployment) [
1]. It has imposed the need to digitise work and forced a focus on the quality of life and health of employees. Studies have shown negative effects of COVID-19 on well-being [
2], job satisfaction, and family life [
3]. Throughout the pandemic, employee well-being has become one of the main priorities of employers [
4].
The responses of organisations to the pandemic have taken into account the need to respond to changes in employee needs. This has resulted in a heightened focus on corporate social responsibility (CSR) [
5] and organisations are placing increasing importance on the well-being of their employees [
6]. They have been introducing various work-integrated learning programmes or embedding a well-being approach in their organisational culture. Through this, they are not only contributing toward the improvement of employee moods but also providing working conditions conducive to high performance [
7]. As many as 80% of organisations operating globally (71% in Poland) declare that employee well-being is important or very important for their success [
8]. However, only 12% (10% in Poland) express full readiness to implement this approach. This limited interest in practice results in a research gap on the components of employee well-being and its relationship with digital work performance [
9,
10].
The dynamics of the pandemic environment have made it necessary for organisations to act and make decisions based on individual experiences. As they have been unable to experiment with different approaches, they have instead been selecting those with proven effectiveness. Lack of knowledge regarding optimal solutions can potentially lead to repeated disruptions in the organisation. The sudden, rapid need for the digitalisation of work from the COVID-19 pandemic resulted in challenges in the spheres of maintaining work efficiency, commitment, and work–life balance. There is therefore a pressing need for up-to-date research on employee well-being.
Although every activity, even those performed in the office without the use of technology, requires a degree of digital support, “remote work” involves the extensive use of new technologies in the performance of daily duties. For the purpose of this article, the terms “remote work”, “telecommuting”, and “digital work” are used interchangeably.
The present article aims to explore the relationship between employee well-being and digital working by adopting the following research hypothesis: there is a relationship between employee well-being and the level of digitisation of work performed, measured by the frequency of remote work. The analysis is based on the results of a survey of Polish workers on their opinions of employee well-being, which was conducted during the COVID-19 pandemic in January 2021. The article is structured in three sections: (1) a theoretical introduction on the essence and research models of employee well-being and the characteristics of remote working in Poland during the pandemic, (2) a section devoted to the methodology of empirical research, including a description of the aim, assumptions, and the research sample, and (3) a section presenting the results of statistical analyses, ending with conclusions.
3. Materials and Methods
The present research aimed to analyse the well-being of Polish employees during the COVID-19 pandemic. Data were collected in January 2021 through a survey conducted on a sample of economically active Polish workers (n = 1000) using the CAWI method based on a nationwide, accredited research panel. The representativeness of the sample was achieved using random sampling. The research was dominated by respondents with higher education (52%) and employed on the basis of an employment contract (77%) in the private sector (77%). The research sample was balanced in terms of gender (55% male and 45% female) and age (each of the four age groups covered between 18% and 28% of respondents). More than half (56%) of respondents had not worked remotely during the pandemic. The remainder had worked remotely from less than 1 day per week (13%) to full-time remote work (10%). Health care workers played a special role during the pandemic as they were classified as essential workers. The difficult epidemic situation resulted in longer working hours, increased stress, and anxiety. At the same time, limited opportunities to work remotely and contact with infected people may have had a particular impact on well-being. Therefore, work in health care, which made up 20% of the sample, was separated in the metrics for comparison to other industries.
Detailed characteristics of the research sample are presented in
Table 1.
The survey consisted of 22 questions relating to respondents’ views on particular aspects of employee well-being, engagement, and evaluation of remuneration fairness. Using the Angoff Method on Cutoff Scores and Judgment Consensus, questions related to engagement and pay equity were identified and excluded. Finally, for the purpose of this article, responses to nine questions related to employee well-being perceptions were analysed. The questions assessing employee well-being were adapted from the diagnostic tool “Gallup-Healthways, Well-Being 5 Index” [
40]. Four dimensions from the above tool were included in the research:
Purpose, i.e., liking what you do every day and being motivated to achieve your goals—examined through questions exploring job satisfaction and being able to do what you do best (questions 7 and 8 in
Table 2)
Relationships, i.e., having supportive relationships in one’s life—examined through items diagnosing the evaluation of the quality and partnership in the relationship with the supervisor (questions 2 and 3 in
Table 2)
Community, i.e., being satisfied with where you are and feeling safe in your own community—examined using questions identifying trust and team atmosphere (questions 1 and 4 in
Table 2)
Health, i.e., good health and enough energy to get things done on a daily basis—explored through questions about the adequacy of health to fulfil the job and hope for the future (questions 5 and 6 in
Table 2)
The Gallup-Healthways global survey shows that Polish people assess their well-being at a level similar to that of Western European countries. The exception is the financial dimension, which, for example, is assessed positively by 55% of respondents in neighbouring Germany, but only by 31% in Poland [
41]. Due to the atypical results of Polish employees in their assessment of well-being in the financial area, this dimension was not included in the analysis of employee well-being. Additionally, a question on the assessment of work–life balance was added (question 9 in
Table 2). Although this is not a component of the Gallup-Healthways tool, research suggests that it significantly forms an indicator of employee well-being [
42].
Respondents rated statements on a five-point Likert scale, with 1 being ‘strongly disagree’ and 5 being ‘strongly agree’. Calculations were performed using the statistical package R (version 4.0.2).
Exploratory factor analysis (EFA) with Promax rotation and with Kaiser normalisation was conducted to isolate the factors responsible for employee well-being based on the questions asked in the survey. The KMO coefficient was 0.904, meaning it was close to 1, and Bartlett’s test of sphericity (approximate chi-squared = 4126.47; 36 degrees of freedom) was statistically significant. Both measures indicate the validity of performing an EFA on the dataset.
As a result of the exploratory factor analysis, it was concluded that there was a solid basis for differentiating between three components of employee well-being, respectively named:
“Workplace relationships”—atmosphere, relationship with supervisor, camaraderie, and trust
“Physical and mental health”—physical condition appropriate to perform the job, hope for the future, job satisfaction, and the opportunity to do one’s best
“Work–life balance”
The factor structure explained more than 68.4% of the variation in the entire study construct and the limiting value of the component loadings was 0.3.
Table 2 presents the statements included in the individual extracted components and the levels of the factor loadings.
4. Results
To determine the impact of telecommuting on the identified employee well-being factors, three logistic regression models were estimated—a separate one for each of the three components. The response variables in each model were the employee well-being factors: “workplace relationships”, “physical and mental health”, and “work–life balance”. The explanatory variable was “remote working”. Three control variables were included in each model: “sector”, “heath care”, and “size of the company”, relating to the type of organisations in which the respondents were employed.
First, the descriptive statistics of the individual employee well-being factors examined were counted (
Table 3).
The analysis was conducted using logistic regression to make it possible to examine the influence of explanatory variables on the dichotomous response variable. To be able to analyse the logistic regression models, it was necessary to determine a uniform index within each of the components of employee well-being. Given that the scores for all three factors were derived from the Likert scale, they were divided into two groups in relation to the median values. The following measures of individual employee well-being factors were adopted:
“Workplace relationships” = 1 for values ≥ 3.75, 0 for values < 3.75
“Physical and mental health” = 1 for values ≥ 4.0, 0 for values < 4.0
“Work–life balance” = 1 for values ≥ 4.0, 0 for values < 4.0
For each of the examined variables, a reference value was adopted, i.e., one against which comparisons were made. In the case of “remote working”, the reference value was the declaration by the employee that he/she did not work remotely. Situations in which remote work was carried out with varying frequency, from less than once a week to exclusive remote work, were referred to through the lack of remote working. For “sector”, the public sector was taken as the reference value, with the private sector and the category combining foundations with associations being referred to. For the variable defining work in “health care”, the reference was the exercise of work in this industry and for the “size of the company”, the smallest organisations, i.e., up to 50 employees, were taken as the reference value. Their share in the surveyed sample was the highest (42%).
Based on the results obtained, we can conclude that “remote working” has a significant impact on the first of the identified employee well-being factors—“workplace relationships” (
Table 4). The odds ratio for the “workplace relationships” factor to be greater than or equal to the median level in the survey (3.75) was less than 1 for both working remotely 1–2 days per week (0.66) and when working fully remotely (0.5). This implies that working remotely 1–2 days per week or working remotely full-time decreases the probability of well-being in the “workplace Relationships” dimension compared to not working remotely.
Of the control variables, a significant relationship linked “health care” with employees’ ratings of “workplace relationships”. Based on the analysis, the chance that the value of the “workplace relationships” factor would indicate a level greater than or equal to the median in the survey was approximately 1.5 times higher when working in other industries compared to health care.
Due to a lack of statistical significance, no relationship could be established between the factor “physical and mental health” and the factor “remote working” based on the findings (
Table 5). Of the control variables, work in the “health care” industry was the only statistically significant relationship linked with this factor. Based on the analysis, we can conclude the chance that the score of the factor “physical and mental health” will indicate a value greater than or equal to the median in the study (4.0) is approximately 3.2 times higher for jobs in industries other than health care, assuming a constant value of the other parameters of the model.
The results indicate that there is a statistically significant relationship between “work–life balance” assessment and “remote working” (
Table 6). The coefficient for the chance that the “work–life balance” score would indicate a value greater than or equal to the survey median (4.0) was less than one (0.58), i.e., the chance of work–life balance well-being was decreasing for fully remote work relative to office-based work.