Effects of the residential built environment on remote work productivity and satisfaction during COVID-19 lockdowns: An analysis of workers’ perceptions

Abstract

COVID-19 pandemic has forced people to stay home and switch to the remote working mode, which – reportedly - affect job satisfaction and productivity. The present study investigates the relationship between the residential environment and worker's job satisfaction and productivity in the remote working mode during the COVID-19 pandemic. A hypothetical structural equation model (SEM) of the influencing factors is constructed based on a literature review and experts' opinions. A survey-based respondents' opinions (n = 2276) were then used to test and analyze the model. The model results reveal that a residential built environment has an indirect effect on both remote work satisfaction and productivity. However, among all the factors, comfortable space (separate space and ergonomic furniture) is found to be the most important. This study presents the importance of adopting a residential built environment to respond to a crisis like a pandemic in achieving the desired comfort level of remote work. Although this study provides a holistic approach, it also proposes a base for the future country-specific analysis by providing some possible countries' differences.

1. Introduction

COVID-19 pandemic has abruptly altered people's lives globally by forcing them to spend most of their time at home to prevent the virus spread. Thus, the pandemic has altered the very definition of living space, as – for many – the dwellings become offices and classrooms, gyms, and more. This change has influenced residences' consideration of building sustainability [[1], [2], [3], [4], [5]], including building services [2,[6], [7], [8], [9]]. In addition, many have been affected in terms of studying [10,11] and working [12,13] due to the forced lockdowns that have switched modes to remote.

Many countries have adopted teleworking with the introduction of COVID-19 lockdowns [[14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25]]. The remote working mode brought positive as well as negative perceptions from professionals who switched from the traditional way of performing their job. For example, the benefits of working from home are reported to be less burnout, a better work-life balance, and lower depression among female parents of underage children [[26], [27], [28]]. The challenges of remote working are associated with nervousness about the coronavirus pandemic and childcare [26,29]. In addition to these, different living conditions of people (e.g., family size, presence of children) have also been individually assessed to identify the level of comfortable workspace setting [26]. It is well documented that the remote workers' well-being [26,30,31], health [13,30,32,33], and productivity [13] have been affected during the pandemic. As such, the factors of indoor environment quality become even more critical when people are isolated in their homes; consequently, it is resulted in lacking socialization and being forced to continue their daily routines (working or studying) at their homes. Several studies shed light on the importance of the services and factors in the indoor environment of dwellings as an influencing factor in working from homes, such as green area [34], light, noise, and space layout [31,32]. Another study reports on the importance of the home layout as a lack of separation between living and working spaces can impact productivity [31]. Other factors that affected productivity were noise, low level of natural light, and absence of good scenery from home windows. It was also found that natural light affects eye health, while noise and air quality issues lead to increased stress rates along with decreased focus [31]. Humidity problems can adversely affect people's nose, throat, and skin, experiencing prolonged exposure [33]. Indoor air quality is one of the essential indoor environmental factors [32]. It was also found that the detrimental effect on the physical and psychological state of those who were working from home was mainly linked to such factors as sports, communication with colleagues, children, and workspace comfort [30].

There have been several attempts to research the impact of indoor and outdoor physical environments on human psychology [34]. However, to the best of the authors' knowledge, the collective effect of the previously reported factors has not been investigated. The relationship between the residential built environment and the remote workers' productivity and satisfaction during the extended COVID-19 lockdowns have yet to be investigated and quantified. Thus, this article aims to quantify the effect of the built environment parameters on workers' job productivity and satisfaction who need to work from home during the COVID-19 pandemic period. The following steps were taken to achieve this research aim: (a) to conduct an extensive literature review on the topic; (b) to develop a hypothetical model upon the review of the context; (c) to develop a structural model of the parameters and test its validity and reliability; (d)to develop built environment-related strategies and recommendations to improve workers’ productivity.

2. Literature review

Indoor environments can initiate different physical and psychological issues among the residents. Continuously being at home can also affect working productivity and satisfaction through indoor environmental factors, such as health and safety, ICT, comfort, and ergonomics.

2.1. Identification of the critical factors

Health and safety. Health and safety in residential environments in the context of pandemics includes physical well-being, mental health, and protection from the viral transmission. Prevention of virus propagation and mental health were considered crucial characteristics of buildings during pandemics by the experts of medicine, academia, and industry [35]. Safety from virus propagation measures includes the use of new smart/innovative technologies that minimize personal involvement (e.g., touchless technologies, motion sensors, keycard swiping), self-cleaning spaces, and proper selection of indoor materials that do not facilitate viral and bacterial propagation and their increase in quantity [[36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46]]. Physical and mental well-being measures include household-level exercise spaces to improve both physical and psychological body states, availability of outdoor spaces in the building (e.g., balconies) to get some fresh air and feel being outside even during strict lockdowns, and access to common building spaces with sufficient safety and social distance for socialization [5,47,48].

ICT. The main technological facilities used by remote workers are personal computers, the internet, and phone [49]. Adequate hardware facilities are essential for remote workers. Nevertheless, there could be a particular need for other work-related equipment, such as headphones, a microphone, a camera, or others—usually, companies who adopt remote working invest in provision with technological facilities [49]. Nevertheless, during pandemic lockdowns, the reliance on robust and adequate speed internet is growing, as most of the services (e.g., medical, product ordering, teaching and meetings) are switched to online, too [[50], [51], [52]].

Indoor environment working comfort and facilities. Working conditions, comfort, and ergonomic facilities are the critical determinants of the quality of the services provided by the residential built environment. It includes such criteria as light, noise, temperature, humidity, indoor air, comfort, and aesthetical indoor environment characteristics [4,5,32,[53], [54], [55], [56], [57], [58]]. Therefore, to provide the building residents with comfort and good mental and physical health, it is vital to keep those indoor environmental factors on the desired levels. Other essential aspects highlighted in the literature include housing form, the facilities it has (e.g., a table, robust WiFi, an office space separated from the living area), and housing prices (like those with the office facilities are generally more expensive) [59].

Moreover, the workplace is strongly desired to be visually private [60]. Ergonomic furniture is another important feature of a comfortable environment for working – proper design can even prevent the rise of pain symptoms through a prolonged period of using ergonomic furniture [61] and is essential for supporting a productive working process [62]. Other research studies on working from home experience showed that workers desired better ergonomic spaces in their homes [26] and improved furniture [60]. Greeneries (indoor gardens and green views) could improve the mental state of residents [51,63]. Green space includes indoor gardens, green views from the windows, and small gardens on balconies [44]. These could help people decrease their level of anger, provide relaxation, and decrease the chance of stress-related diseases such as cardiovascular illnesses and depression [33]. Availability of appropriate technologies and support from the workplace are claimed to improve the motivation and engagement in remote working [12].

Remote work productivity. Having an increase in productivity while working from home is still arguable. For example, one study in France revealed that only around twenty percent of the respondents indicated more productivity during remote working than in the office environment [49]. Other studies showed that people remotely working were more productive (around 80%) even when they got ill [64], and the workers’ productivity enhanced over prolonged remotely work [65]. Women and workers of older age tend to be more productive during remote working [13]. In other available studies on remote working, productivity levels were not changed with the shift from office to home [13] or even decreased [66,67].

Nevertheless, better psychological and physical states were observed, and higher productivity levels were achieved among the workers [13]. Other factors that improved productivity were indoor temperature, the absence of small children, and a comfortable workspace [13]. Various methodologies and metrics were used in the literature studies to evaluate employees' productivity. For example, the work environment (physical and non-physical) effects on employees’ productivity were investigated [68]. The parameters they used included Timeliness, Quantity, Quality, Attendance, and Ability to work in teams [68]. Other studies used indicators including, but not limited to, efficiency and effectiveness of work, creativeness, initiative, opinions expressed and generating new ideas [[69], [70], [71], [72]]. Although the terms “performance” and “productivity have been used interchangeably, the current study uses “productivity” which is defined as “a measure of how efficiently resources are utilized to achieve desired outputs”. However, it could be interpreted in different ways depending on the application context. In the remote work context, it can be defined as “successful and efficient execution of a project (by remote employee) and surpass of set goals in any pleasing space”. The present study adopts this definition to address the measured parameters identified in the theoretical framework.

Remote work satisfaction. The workers’ desire to switch to remote mode is also unclear, as a small number of those find reduced commuting time and increased work-life balance as incentives to prefer remote working [73,74]. Nevertheless, it can be described by the fact that the workers adopt a stance that their management would not allow remote working; thus, they do not consider it [49]. In the United States, many workers would like to continue in the remote mode after the pandemic, too [75]. Remote workers tend to lose their satisfaction compared to traditional workers before the COVID-19 pandemic [75]. Factors that encourage the development of remote working are suggested to be categorized by the following: (1) the nature of the performed tasks; (2) the awareness of the advantages and drawbacks of remote working; and (3) reconcilability with the work culture [49]. Blurred time and the spatial boundary between work and personal life increase anxiety among remote workers, working hours, and focusing on professional tasks [26]. Moreover, increased autonomy during teleworking leads to an increased feeling of loneliness and, as a consequence, stress [76].

Nowadays, many factors might influence an employee's satisfaction with one's job [77], ranging from more objective parameters, such as salary, quality of supervision, and work & life balance, to more subjective ones, including personal values, sense of fulfillment and purpose, the realization of one's progress and sense of belonging [77]. Numerous academic and commercial studies were conducted to examine employees' job satisfaction. A study by Swarnalatha & Sureshkrishna (2012) showed that the job satisfaction of automotive industry workers in India used commitment, compensation, responsibility, achievement, supervisory support, workgroup cohesion, and quantitative workload to evaluate employees' satisfaction [78]. Other studies, such as Girma (2016) and Martins & Coetzee (2007), used some other metrics that include communication, diversity, fairness, job satisfaction, opportunities for growth, productivity management, respect for employees, respect for management, teamwork, work/life balance [79,80].

2.2. Theoretical framework

The initial model, defining the relationships between the selected parameters, is proposed based on the extensive literature review. Health and safety, comfort and ergonomics, and ICT and other Enablers are identified as the primary physical parameters of the residential built environment impacting remote work productivity and satisfaction.

Safety from virus propagation, Mental health, and Physical health are chosen for further evaluation of Health and Safety in the residential buildings of the remote workers. Light, Noise, Humidity, Temperature, Indoor air, Comfortable working space, Ergonomic furniture, and Accessible greeneries are the indicators selected for assessing Comfort and Ergonomics [4,5,26,32,[53], [54], [55], [56], [57], [58], [59],61,63]. They potentially help to evaluate the level of a comfortable and ergonomic environment of those who have worked remotely during the COVID-19 pandemic. The following indicators are chosen to assess ICT infrastructure among the remote workers; Adequate hardware, Other work-related equipment, Internet connection, and Company/organizational support [2,49,51,52]. It is decided to focus on seven key productivity indicators to evaluate remote working productivity; Timeliness, Quality, Quantity, Impact, Efficiency, Engagement, and Team communication [13,49,65,[68], [69], [70], [71], [72],81]. These indicators are chosen based on their relative ease of measurement from employees' perspectives and more objective metrics. Collectively, these indicators may effectively capture any key changes, should such occur, in employees' productivity. It is also decided to use four indicators to evaluate employees' job satisfaction during remote work: work-life balance (with regards to having enough time to sleep, exercise, and be with family), employees' preferred working mode (office vs. remote work), tolerance to salary reduction to be able to work from home, and overall job satisfaction. These indicators are assumed to be sufficient to build a general understanding of job satisfaction differences between office and remote work [49,[73], [74], [75],[78], [79], [80]]. The third parameter, tolerance to salary reduction, is unique since it aims to measure employees’ willingness to sacrifice monetary benefits to keep the option of working from home, giving us an idea of the overall perceived value of working from home.

Following the literature review and initial selection of the parameters as discussed above, the identified factors and parameters were further discussed during a workshop that was conducted to finalize the model. The co-authors of the research represented different countries, such as Kazakhstan, Slovenia, Turkey, Romania, Poland, South Korea, the United Kingdom, Indonesia, and Malaysia. It achieved a rich discussion on developing the structural equation model and survey instrument. Thus, a conceptual structural equation model is developed and presented in Fig. 1 .

Fig. 1.

Hypothetical model of factors influencing remote working.

Fig. 1 shows the main inputs – Health and Safety, Working conditions, Facilities, ICT, and other enablers' availability in Built Environment, which leads to Remote work productivity and Remote work satisfaction. The list of latent and observable variables and their corresponding questions can be seen in Table 1 . As this research study investigates the link between the residential built environment and professional workers’ productivity and satisfaction in the framework of remote mode, nine hypotheses have been developed. Each connection (arrow) in Fig. 1 represents a hypothesized relationship between two factors, and overall the model contains nine main hypotheses, the direction/sign of each is assumed to be positive.

H1

Residential Built Environment Facilities influence Remote Work Satisfaction.

H2

Residential Built Environment Facilities affect Remote Work Productivity.

H3

Residential Built Environment Health and Safety impacts Remote Work Satisfaction.

H4

Residential Built Environment Health and Safety affects Remote Work Productivity.

H5

Residential Built Environment ICT & other enablers influence Remote Work Satisfaction.

H6

Residential Built Environment ICT & others improve Remote Work Productivity.

H7

Residential Indoor Environment Working Comfort affects Remote Work Satisfaction.

H8

Residential Indoor Environment Working Comfort impacts Remote Work Productivity.

H9

Remote Work Productivity affects Remote Work Satisfaction.

Table 1.

Latent and observable variables.

Latent variables	Observable variables	Measuring question/statement
Health and Safety (BE H&S)	HS1. Safety from virus propagation	My home is well protected against virus propagation
	HS2. Mental health	My home environment keeps my mental well-being in a good state
	HS3. Physical health	My home environment keeps my physical well-being in a good state
Indoor Environment Working Comfort (BE WC)	C1. Light	The level of natural light at my home is comfortable and sufficient for working at home
	C2. Noise	The noise level at my house is comfortable for working at home
	C3. Humidity	The humidity level at my home is comfortable for working at home
	C4. Temperature	The temperature level at my home is comfortable for working at home
	C5. Indoor air	The air at my home is healthy and comfortable
Facilities (BE F)	C6. Comfortable working space	a. There is a personal table at my home for comfortable working;
		b. There is an office space separated from living space at my home;
	C7. Ergonomic furniture	The ergonomic design of the furniture in my home does not cause me any pain or discomfort (e.g., pain in my neck, shoulder, back, eyes)
	C8. Accessible greeneries	I have access to greeneries (e.g., green views, green plants)
ICT and other enablers availability (BE ICT & others)	I1. Adequate hardware	I have sufficient hardware at my home (e.g., computer, laptop, tablet, phone)
	I2. Other work-related equipment	I have sufficient communication devices at my home (phone, microphone, camera)
	I3. Internet connection	I have adequate internet access and speed at my home
Remote Work Productivity (RW Prod)	P1. Quantity	I complete more tasks when remotely working than I do in the office
	P2. Quality	The quality of tasks I perform during remote work is better than in the office
	P3. Effectiveness	I had improved the impact of my work when I switched to remote work
	P4. Efficiency	I spend less amount of energy on the completion of a task during home working
	P5. Engagement	I engage more to work activities and meetings during home working
	P6. Ability to work in teams	I am more capable of working and communicating with a team during home working
Remote Work Satisfaction (RW Sat)	JS1-JS4. Work-life balance	I get sufficient work-life balance while working at home
		I have enough time to sleep
		I have enough time to exercise
		I spend enough time with my family
	JS2. Preferred working mode	If I could choose between working in the office or working from home, I would prefer to work from home.
	JS3. Overall satisfaction	Overall, I am more satisfied with working from home.

3. Methodology

3.1. Survey instrument and data collection

The survey instrument was developed based on the reviewed literature and expert opinions obtained during the online workshops in October 2021. This workshop included representatives of academia (faculty staff) from different fields, such as civil engineering, mechanical engineering, materials engineering, ergonomic engineering, economics, and transportation, from several countries (Kazakhstan, Romania, South Korea, UK, Turkey, Slovenia, Poland, New Zealand). The brainstorming resulted in a mind map, which can be found at the following link: https://miro.com/app/board/o9J_lq9Xg-I=/. Once the researchers agreed on the final structure and content, it was submitted for the approval of the Nazarbayev University Research Ethics Committee. The survey consists of 23 questions related to the proposed research model, each of which is asked in the Likert scale format. Besides the model questions shown in Table 1, the survey contains 11 socio-demographic questions related to the respondents’ remote working experience during the COVID-19 lockdown, such as; how long do they work remotely, where do they live, and what type of residence do they live, the number of housemates they have, and whether there any children living with the respondent, age, and gender. In the present study, productivity is measured based on self-assessment of the workers. The questionnaire was translated into eight languages widely used in the regions covered by this research and made available at the link https://nukz.qualtrics.com/jfe/form/SV_bIBwWADmmpZBgAm. The data was collected via the online instrument Qualtrics from November – to December 2021. No specific target group was aimed during data collection, so all the online working experience respondents were welcomed.

3.2. Structural Equation Modelling (SEM) and validity checking

For the evaluation of the reliability, validity, and further analysis of the model proposed in Section 2.1, the approach of Partial Least Square Structural Equation Modelling (PLS-SEM) has been utilized as a multivariate statistical tool for exploratory analysis of hypotheses and identification of the path weights (represented in Fig. 1) with the utilization of SmartPLS program [[82], [83], [84]]. SEM is a statistical tool for measuring and further analyzing a model that represents relationships between observable and unobservable variables. Thus, through analysis of input manifested variables, latent variables and the relationship between latent and observable variables are measured. Each of the latent variables is measured through at least three observables. For more precise analysis, a minimal quantity of manifested variables is better. PLS-SEM is a method used to evaluate compound relations, reasons, and consequences in path models with manifested and unobservable variables. Thus, this method is suitable to the scope of the study as it lets estimate the relationship between residential built environment factors and work from home satisfaction and productivity via manifested variables.

As per the PLS procedures, SEM reflective measurement model should be checked for its validity by checking the proposed model's (1) outer loadings, (2) Cronbach's Alpha (CA), (3) Dillon-Goldstein's rho (rho_A), (4) composite reliability (CR), and (5) Average Variance Extracted (AVE). Outer loadings represent the relationship between the latent indicator variable and its reflective construct, showing a strong relationship when equal to or greater than 0.7 [85]. While CA, rho_A, and CR are the unidimensionality checks that show how latent variables are consistent internally [86]. Similar to outer loadings, the minimal acceptable value for CA and CR is 0.7. AVE confirms that each latent variable converges while its minimum acceptable value is 0.5.

4. Results & discussion

In total, 2276 responses were received; among them, 1918 were suitable for further analysis. The following criteria were used in the selection: the respondent answered positively that he was working remotely during the COVID-19 pandemic, and the response contained 70% of answers to questions on observable variables. According to Hair et al. [82], the minimum sample size should be ten times larger than the number of observed variables; therefore, our dataset fits the minimum sample size rule for the analysis of the SEM model. The respondents are from 35 countries. Fig. 2 shows the Euro-Asian distribution graph with countries’ contributions.

Fig. 2.

Number of responses from contributing Euro-Asian countries.

The occupation types of the respondents are also presented in Fig. 3 . The majority (around 30%) is occupied in the education sector. Other most prevalent in terms of responses working sectors are Business, consulting, management; Information Technology; Accounting, banking, and finance; and Engineering and manufacturing.

Fig. 3.

Working sector representation of respondents.

Table 2 represents the socio-demographic characteristics of the respondents. The majority of the respondents (53%) stated that they had experienced remote working over a more extended period than six months, while only a minor group (10%) had it for less than a month. Most of the respondents are living in urban areas (63%), in apartments (54%) with a total area of 50–75 sq.m. (33%). More than half of the respondents do not have underage children, while the age of the majority is between 20 and 30 (39%). Gender division is almost equal – 52% females and 47% males.

Table 2.

Socio-demographic statistics.

The remote working period of the respondent
Less than one month	200	10%
1–3 months	391	20%
4–6 months	294	15%
More than six months	1031	53%
Living area
Highly rural	238	12%
Rural	204	11%
Suburban/Metropolitan	260	14%
Urban	1214	63%
Type of residence
Dormitory/shared room	45	2%
Apartment	1032	54%
Attached house	182	9%
Detached house	657	34%
The total area of residence
Less than 50 sq.m.	338	18%
50-75 sq.m.	644	33%
75-130 sq.m.	576	30%
More than 130 sq.m.	358	19%
How many people respondent shares his home with
Lives alone	145	8%
1-2 people	694	36%
3-4 people	764	40%
With five and more people	313	16%
Presence of underage children
No	986	51%
Yes, 1 child	421	22%
Yes, 2–3 children	456	24%
Yes, 4 and more children	53	3%
Age
Less than 20	168	9%
20–30	747	39%
31–40	565	29%
41–50	278	15%
More than 50	158	8%
Gender
Female	1005	52%
Male	898	47%
Prefer not to say|Other	13	1%

4.1. SEM results, validity, and implications

The majority of outer loading scores in the constructed SEM are higher than the limit of 0.7, except for C8, HS1, and I4; nevertheless, their values are close. It leads us to conclude that the manifested variables are valid in their relations to the latent variables. Nevertheless, C8, HS1, and I4 values are close to 0.7. All other unidimensional values (CA, rho_A, CR, and AVE) also fall within the acceptable limits, see Table 3 . Thus, the model assessment shows that all SEM factors are validated and are suitable for further analysis. Fig. 5 represents the developed structural equation model.

Table 3.

Outer model results and construct reliability and validity (Acceptance criteria: CA>0.7, AVE>0.5, rho_A>0.7, and CR > 0.7).

	Outer loadings	Mean	Standard Deviation	CA	rho_A	CR	AVE
C1	0.755	4.348	0.904	0.846	0.846	0.890	0.619
C2	0.738	4.044	1.087
C3	0.845	4.281	0.911
C4	0.798	4.450	0.809
C5	0.793	4.384	0.838
C6a	0.740	4.223	1.124	0.743	0.766	0.838	0.566
C6b	0.776	3.187	1.558
C7	0.836	3.587	1.307
C8	0.645	3.986	1.231
HS1	0.686	4.166	0.940	0.723	0.744	0.845	0.647
HS2	0.868	4.240	0.913
HS3	0.847	4.074	1.050
I1	0.856	4.513	0.814	0.778	0.778	0.872	0.694
I2	0.871	4.483	0.846
I3	0.768	4.211	1.053
JS1	0.792	3.723	1.227	0.862	0.869	0.896	0.589
JS2	0.743	4.022	1.138
JS3	0.755	3.717	1.237
JS4	0.737	4.042	1.112
JS5	0.788	3.452	1.428
JS6	0.788	3.426	1.381
P1	0.779	3.642	1.218	0.902	0.904	0.925	0.673
P2	0.858	3.526	1.195
P3	0.871	3.460	1.199
P4	0.754	3.621	1.254
P5	0.836	3.377	1.281
P6	0.819	3.263	1.277

Fig. 5.

Developed structural equation model in Smart PLS.

Discriminant validity (Table 4 ) is another important characteristic needed to be checked for the proposed reflective measurement model [82]. Discriminant validity shows how a construct is different from other constructs, which is seen by correlating it with other constructs, thus, seeing the extent of how many observable variables characterize a single construct. Since all the values in Table 4 are different from each other, this model is validated.

Table 4.

Discriminant validity of constructs.

	BE F	BE H&S	BE ICT & other	BE WC	RW Sat	RW prod
BE F	0.752
BE H&S	0.484	0.804
BE ICT & other	0.429	0.360	0.833
BE WC	0.592	0.602	0.497	0.787
RW Sat	0.420	0.377	0.346	0.394	0.768
RW prod	0.438	0.351	0.315	0.349	0.692	0.821

As the main aim of this research study was to identify whether residential built environment conditions influence remote work satisfaction and productivity, it is important to test the model for the set hypotheses. It is done through analysis of the path-values (need to converge to 1) and p-values (need to be within 5% limit) [82]. As shown in Table 5, Table 8 out of 9 hypotheses are supported.

Table 5.

Hypothesis test results.

Hypothesis		Path value	Original Sample	Sample Mean	Standard Deviation	P Values	Comment
1	BE F - > RW Sat	0.036	0.046	0.046	0.023	0.046	Supported
2	BE F - > RW prod	0.288	0.302	0.302	0.028	0.000	Supported
3	BE H&S - > RW Sat	0.070	0.072	0.072	0.025	0.005	Supported
4	BE H&S - > RW prod	0.144	0.148	0.147	0.027	0.000	Supported
5	BE ICT & other - > RW Sat	0.103	0.074	0.075	0.021	0.000	Supported
6	BE ICT & other - > RW prod	0.148	0.121	0.121	0.023	0.000	Supported
7	BE WC - > RW Sat	0.072	0.078	0.078	0.026	0.003	Supported
8	BE WC - > RW prod	0.016	0.021	0.023	0.028	0.450	Not supported
9	RW prod - > RW Sat	0.590	0.596	0.596	0.019	0.000	Supported

Table 8.

SEM variables’ scores for analysis by working sector.

	Education		Accounting, banking, and finance		Business, consulting, management		Engineering and manufacturing		Information Technology
	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod
BE F	0.110	0.328	0.292	0.614	0.121	0.151	0.068	0.302	0.195	0.280
BE H&S	0.063	0.121	0.010	0.163	0.093	0.200	0.149	0.088	0.095	0.249
BE ICT & other	0.010	0.076	0.179	0.148	0.044	0.198	−0.003	0.208	0.136	0.103
BE WC	0.062	0.048	−0.049	−0.160	0.126	0.012	0.085	−0.036	0.094	0.003
RW prod	0.56		0.343		0.575		0.633		0.451

The only unsupported hypothesis is the path from built environment working comfort to remote work productivity (H8). Thus, it is not proved that the working comfort variables of the built environment (light, noise, humidity, temperature, and indoor air) lead to better teleworking productivity for the given sample. In contrast, in the study of Awada et al. [13], it is claimed that satisfactory temperature, air quality, noise, and lighting level correlate with better productivity levels during remote work; still, the correlation is weak. H7 (0.072), which claims that built environment working comfort leads to better satisfaction during teleworking, is supported yet low. In contrast to a low score of H7, noise, air, and light are claimed to be very important in achieving workers’ satisfaction in available literature [87].

The strongest path value is H9, which demonstrates that staying productive during remote work leads to better satisfaction. This finding is similar to the findings of Toscano and Zappala, which claim that feeling productive during teleworking makes people feel more satisfied with their remote job [66]. In the offline working environment, increased productivity is also proven to lead to better satisfaction [88]. Moreover, H9 shows that although the direct effect of the residential built environment on remote work satisfaction is low (as H1-H7 path values are small), the effect is much higher indirectly – through remote work productivity.

The second strongest hypothesis is H2 (0.288), which connects built environment facilities and remote work productivity. Therefore, it shows that having an ergonomic workplace and greeneries is important for the respondents to be productive during teleworking. Indeed, the available literature also proved that dedicated working space and comfortable desks and chairs are associated with improved productivity during remote work [13]. The availability of plants is also claimed to enhance productivity levels [89]. In contrast, hypothesis H1 has the lowest path value (0.036), which means that built environment facilities have the lowest effect on remote work satisfaction. Thus, the observable variables of H1 - comfortable working space, ergonomic furniture, accessible greeneries - are slightly influencing satisfaction with the remote work. In another research, comfortable furniture is claimed to be crucial for the workers’ satisfaction [87], while greens tend to lead to happier workers [89].

Hypotheses on the effect of ICT have similar path values –H5 (0.103) and H6 (0.148). Thus, the effect of ICT conveniences on remote work productivity and remote work satisfaction is similar for the surveyed respondents. Similar to H5 and H6, other research also claims that adequate ICT resources positively impact productivity [90] and work-life balance during COVID-19 pandemic teleworking [91].

H3 (0.070) and H4 (0.144), which represent paths from health and safety residential facilities to work from home satisfaction and productivity, correspondingly, are also supported in the analysis of SEM. In the authors’ previous research, it was also found that health and safety facilities are important for remote study satisfaction [10]. These findings are resonant with the available literature studies: health (both mental and physical) are claimed to affect working productivity [13] and satisfaction [66]. During remote work in the context of pandemics, workers feel safer from the virus at home, which leads to better satisfaction, yet, social isolation decreases productivity [66].

Additionally, we have received 428 comments from the respondents. Forty percent of the respondents emphasized the need for comfortable working space (C6), including the need for bigger space and moving to another home space. Twenty percent of them responded that they would like a better internet connection (I3) for more comfortable remote work. Ten percent needed more ergonomic furniture at home. Some of the remote workers expect their workplaces to support their workers by paying for the internet, hardware/software, and furniture for the comfortable working (10%), and through the increase of computer literacy of the workers, and having less stress from the office heads (2%). Five percent expressed that they prefer working in the office, and several respondents mentioned they need more social interaction during working. Five percent of the people that have experienced teleworking mentioned they had problems with light and noise in their residential environment. Two percent expressed their fears of worsening their health during remote working through increased screen time. Another 2.5% were craving better indoor air quality (C5), having more greeneries at home (C8), improved humidity (C3), and temperature (C4). Nine percent of the respondent would be more satisfied with remote working at home if they had better hardware – monitors or laptops (I1) - and other equipment – printers and scanners (I2). Due to blurred boundaries between work and home, three percent wanted improved work-life balance through fewer working hours and less work stress. The total percentage is higher than 100% because some of the comments expressed several points simultaneously.

4.2. Analysis by groups: by country, working sector, and gender

As the model has been validated, it was also interesting to analyze different groups (gender, country, and working sector) to find any deviating implications. Although the number of the collected dataset might not be sufficient to represent the views of the separate gender, country, and working sector, the research study still would like to find any possible features which might create a ground for further investigations.

Table 6 summarizes the SEM scores for different genders. For females to achieve satisfaction from remote work, built environment facilities, ICT, and working conditions are more important. The most substantial factor, among others, is comfortable working conditions. At the same time, males have better remote work satisfaction when their homes provide health and safety, ICT, and facilities. The most important factor, among others, is health and safety. To achieve remote work productivity, a built environment with comfortable facilities is the most important for both males and females. Moreover, females were found to be more productive than males, which is similar to the available findings [13].

Table 6.

SEM variables’ scores for analysis by gender.

	Female		Male
	RW Sat	RW prod	RW Sat	RW prod
BE F	0.085	0.315	0.085	0.289
BE H&S	0.047	0.143	0.113	0.152
BE ICT & other	0.080	0.134	0.046	0.103
BE WC	0.100	0.055	0.038	−0.025
RW prod	0.528		0.576

Table 7 summarizes the results for different countries. For the remote workers in Kazakhstan, all the built environment factors have almost a similar effect on the satisfaction from remote work. Regarding productivity, built environment facilities are the most important, while built environment working comfort has a minor effect. Polish workers’ satisfaction from working from home depends on built environment facilities and working conditions, while facilities and comfortable working conditions influence productivity. Slovenia is observed to have an equal effect on the latent variables on remote work satisfaction (except for health and safety, which do not influence at all). The productivity from remote work is also similarly dependent on the factors except for built environment working conditions, which do not affect productivity. Romanian workers find residential facilities and health and safety at home to not affect remote work satisfaction, while residential working conditions are the most important factor. In contrast, residential facilities and health and safety are the most important factors for productivity. Remote workers in Turkey find built environment facilities and ICT as the most influential factors in both remote work satisfaction and productivity.

Table 7.

SEM variables’ scores for analysis by country.

	Kazakhstan		Poland		Slovenia		Romania		Turkey
	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod	RW Sat	RW prod
BE F	0.097	0.353	0.149	0.148	0.095	0.211	−0.010	0.330	0.144	0.380
BE H&S	0.092	0.155	0.009	0.000	−0.009	0.198	−0.039	0.204	0.070	−0.119
BE ICT & other	0.077	0.147	−0.019	0.034	0.030	0.158	0.069	0.046	0.096	0.209
BE WC	0.065	0.002	0.115	0.230	0.025	−0.059	0.188	−0.050	−0.006	0.037
RW prod	0.523		0.563		0.615		0.679		0.443

Table 8 summarizes the analysis by working sector. For the workers involved in education during remote work, all four factors are similarly important for the satisfaction from remote work, while for productivity – facilities at home are the most influential. Accounting remote workers find residential facilities the most important factor for both satisfaction and productivity from working from home. Business sector workers, during remote work, find all factors have similar importance for remote work satisfaction. Compared to the Business workers, all four figures of the Education workers are smaller, which could also show that all four have little relationship with the residential environment. Whereas, for remote work productivity, BE F, BE H&S and BE ICT have more influence than others. The respondents representing the engineering sector are observed to find residential health and safety as the most influential factor in remote work satisfaction. The most affecting variables are productivity, built environment facilities, and ICT. For the IT workers, comfortable facilities and ICT are most influential on remote work satisfaction, while remote work productivity is most influenced by built environment facilities and health and safety.

Fig. 4 summarizes the general trends of the collected data from the respondents. The most satisfying remote working sectors are Recruitment and HR, Sales, Media and Marketing, and Transport and Logistics workers, while the most unsatisfying sectors are Public services and Administration, Law, and Education (Fig. 3-a). The percentage of those who feel dissatisfied with remote work decreases the longer people work, except for those who worked remotely for less than one month.

Fig. 4.

a) Remote work satisfaction by working sector, b) Satisfaction with working from home depending on gender and number of underage children living in the same residence, c) Satisfaction from working from home depending on the number of people the respondent shares his home with and living area type.

Females and males with four and more children are most satisfied with working from home (Fig. 3-b). These results are aligned with the earlier publications; for example, in the United States, females prefer more days working remotely [65]. Thus, some researchers claim it is important to assess gender roles in the environment that erases boundaries between office and living space, as women are generally more involved in unpaid domestic labor [59]. In addition, parents of underage kids are more likely to encounter problems during COVID-19 [75].

Those who live alone in rural areas are the least satisfied with remote working among all other groups. Respondents living in highly rural areas are the most satisfied with working from home, while those in suburban areas are the least satisfied. This trend is similar to the authors’ previous research, where students residing in suburban areas were also the least satisfied with remote education [10]. The highest dissatisfaction with remote working is observed in the highly rural areas. This trend correlates with the quality of internet access, which was reported to be the poorest there. The better the internet access, the more satisfied respondents are with remote work. Better internet access, in turn, correlates with the degree of urbanization.

5. Conclusion

The present study investigates the built-in effects of the residential environment on remote work satisfaction and productivity during the COVID-19 pandemic. A structural equation model (SEM) was constructed based on the literature review, hypothesizing a path relationship between the residential environment and remote work satisfaction and productivity. The SEM analysis indicated that the residential built environment indeed had an effect on remote work productivity and satisfaction, but its direct impact is not very large (the path values range from 0.016 up to 0.103). However, the indirect effect of the built environment on satisfaction through productivity was more substantial (path value 0.590). In more detail, factors such as ‘Health and Safety’ (safety from virus propagation, mental and physical health), ‘Working Comfort’ (light, noise, humidity, temperature, indoor air), ‘Facilities’ (separate from living and ergonomic working space, greens) and ‘ICT’ (equipment for work and internet) affect remote work satisfaction. Although the dataset is limited, this study also provided some possible gender-, country-, and working sector-specific features, which might be a basis for a more thorough and data-rich study in the future. In general, for different sex, countries, and working sectors, the most important factor was comfortable working facilities, which include comfortable working space, ergonomic furniture, and greeneries. These aspects have been further emphasized in the comments of the respondents.

COVID-19 has been a lesson for the whole world on living and working under total isolation beyond the accustomed ways of living and working. Thus, based on the SEM analysis findings and the comments from the surveyed, the following implications can be suggested for managers and teleworkers themselves to promote a better remote working experience. First, creating comfortable facilities for remote working is highly recommended. It includes separate working spaces with comfortable furniture and plants, as hypothesis H2 (effect of built environment facilities on remote work productivity) is the strongest among others, representing the built environment's effect on remote work satisfaction and productivity. Second, providing adequate ICT resources is important, including adequate hardware and robust internet. Moreover, respondents have noted that providing training on the use of ICT resources would be required for teleworkers. The third priority is to facilitate health & safety and working comfort for the teleworkers. These include providing a safe environment against virus propagation and having adequate mental and physical health, light, noise, humidity, thermal environment, and indoor air quality. Overall, SEM results imply that focusing efforts on workers' remote work satisfaction and productivity in these three areas would likely provide the highest return on investment of resources. In contrast, the hypothesis about built environment working comfort effect on remote work productivity is not supported during SEM analysis. Thus, light, noise, humidity, thermal, and air comfort are found to be non-priority factors in pursuing better teleworking productivity.

The present research provides insights regarding the effect of the residential built environment on remote work satisfaction and productivity. The constraints of the study include a limited number of respondents from certain countries and the bounding of the study by selected variables – built environment, remote work productivity, and satisfaction. We suggest conducting a rigorous country analysis with more responses collected in future studies. It is possible that in this study, people provided opinions depending on their conditions, as we collected the data during wintertime, while the responses might be different during other seasons. In future studies, the researchers could also do a similar study in another season. In addition, age can be used as a moderative parameter for SEM.

Funding

The authors acknowledge the financial support from the Nazarbayev University Faculty Development Competitive Research Grant Program (Funder Project Reference: 280720FD1904).

CRediT authorship contribution statement

Aidana Tleuken: Writing – original draft, Visualization, Software, Investigation, Formal analysis, Data curation. Ali Turkyilmaz: Writing – review & editing, Validation, Methodology, Conceptualization. Magzhan Sovetbek: Writing – original draft, Investigation. Serdar Durdyev: Writing – review & editing, Writing – original draft, Resources, Conceptualization. Mert Guney: Conceptualization. Galym Tokazhanov: Resources, Conceptualization. Lukasz Wiechetek: Writing – review & editing, Resources, Conceptualization. Zbigniew Pastuszak: Writing – review & editing, Resources, Conceptualization. Anca Draghici: Writing – review & editing, Resources, Conceptualization. Maria Elena Boatca: Resources, Conceptualization. Valerij Dermol: Writing – review & editing, Resources, Conceptualization, Resources, Conceptualization. Nada Trunk: Resources, Conceptualization. Serik Tokbolat: Resources, Conceptualization. Tamar Dolidze: Writing – review & editing, Resources. Lin Yola: Writing – review & editing, Resources. Egemen Avcu: Resources. Jong Kim: Writing – review & editing, Supervision, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.