3.1. Descriptive Analysis of the Telework-Related Literature
As indicated above, the sample analysed in this research comprises a total of 539 documents. The selected articles from WoS were published between 1984 and 2021. As can be seen in
Figure 1, the rate of publications increased gradually from the beginning of the 1990s until 2019. This growth is common in almost any subject area, as the overall number of journals and publications has been increasing markedly in recent years [
71,
72]. However, in 2020 and 2021, the volume of scientific papers related to telework increased very strongly. In 2020, 64 papers were registered in WoS related to telework, which is more than twice as many as in 2019 (26 papers). In 2021, taking into account that the import of files from the WoS database took place in July of this year, 51 documents were already recorded, so it is foreseeable that the number will exceed one hundred.
The Web of Science database used for our research offers a classification of publications according to the area of knowledge from which they originate. The areas of knowledge from which telework dynamics were investigated are many (
Table 2). Although we mainly found areas focused on
business and economics, other disciplines have addressed this issue, such as
engineering;
psychology;
sociology;
social sciences;
public administration;
transportation;
computer, information and library science;
and women’s studies.
The 539 documents from the WoS database that make up the sample of this research were produced by a total of 1081 authors. The diversity of the areas of knowledge that have dealt with telework has an impact on the presence of authors with greater or lesser scientific production in this field.
Table 3 lists the authors with the largest number of telework-related papers. The dates of their first and last publication in this field (included in WoS) and the number of citations received from any database are also added. The following authors stand out in this list: Timothy D. Golden (USA), who has been researching in this area for two decades, showing special interest in professional isolation, conflict between work and family and relationships between co-workers, among other topics; Manuela P. Pérez (Spain), with research on telework, ICT and human resource management, flexibility in organisations and the reconciliation of work and family life; and Pascale Peters (Netherlands) with research on flexible working, work–life balance and strategic human resources management.
It is worth highlighting the research carried out by Kossek, E.E., in which we found the most cited paper in the sample, “Telecommuting, control, and boundary management: Correlates of policy use and practice, job control, and work-family effectiveness” [
73], with 378 citations received. This research addressed the concept of flexibility and work–family conflict and the individual’s psychological perception of this flexibility. Also noteworthy is the manuscript by Lapierre, L.M. and Allen, T.D. Although they were not among the authors with the largest number of papers on this topic, their article, “Work-supportive family, family-supportive supervision, use of organizational benefits, and problem-focused coping: Implications for work-family conflict and employee well-being” [
74] has been cited 281 times, where they address the work–family conflict in employees who practice teleworking.
Due to the multidisciplinary approach of telework studies, published research also exposes other points of view and uncovers a wide range of issues related to telework dynamics: job satisfaction [
75]; employee relations [
76]; virtual management [
77] (Cascio, 2000); and worker health and ergonomics [
78], among others.
The journals that have published the largest number of telework-related research papers in the research sample are shown in
Table 4. The 539 papers analysed were published in a total of 262 journals. Most of the journals dealing with telework are related to Business and economics and human resources. The quartile of each journal according to the
Journal Citation Reports (2020) classification is also shown. In the event that a journal is classified in more than one subject area, all the corresponding quartiles are marked. Most of the journals that contributed the most documents to the sample used correspond to the Q1 and Q2 quartiles. The journal with the largest presence in the sample of this research was “New Technology Work and Employment”, with more than 10% of the documents analysed.
The documents analysed come from 54 countries. In order to avoid errors in the interpretation of the countries that contributed most to this research topic, only the country of location of the university or institution of the first author of the article is reported. In this way, the countries of origin of the articles are counted more correctly, and cases where several authors from the same country appear in a single article are avoided. In this relation, among the countries with the highest number of articles, we found the USA and England with 113 and 70 documents, respectively, followed by Canada (33 documents), Spain (30 documents), Netherlands (27 documents), Germany (21 documents), Australia (18 documents) and Brazil and Italy with 15 documents, each published in WoS.
3.3. Longitudinal Analysis
Once the evolution of the keywords is known, we set out to carry out a longitudinal analysis of telework-related research. To do this, we used longitudinal maps (
Figure 3). These provided us with different perspectives on how the topics evolved in each of the periods. Firstly, we can see with which topics the first research started (
Figure 3—left) and what the current trends are (
Figure 3—right). On the other hand, the size of the spheres representing the clusters related to a topic provides an idea of the importance of that cluster. In our case, the size of the spheres represents the number of documents published. Finally, the evolution of the topics is symbolised by the lines connecting the different clusters. Having made this brief explanation of the procedure, we note that, from a longitudinal point of view, it can be observed that the main research contribution both quantitatively (number of documents) and qualitatively (impact of publications) is offered by the clusters related to conflict, management and labour flexibility policies (time axis Conflict–Management–Flexibility).
As for the first period (1984–2009), it contains a set of six clusters. The most important clusters were Conflict, Performance and Organisation. These three clusters, as we will see later in the analysis of strategy maps, are considered to be driving themes, i.e., they are themes that drove research in the period under analysis. Among them, the cluster with the highest relevance was Conflict, with 26 papers and the highest impact, 2985 citations, but Organisation was the cluster with the highest centrality or external cohesion (level of connections with other themes) (
Table 5). The impact of the Conflict cluster in the first period is due to the proliferation of highly cited papers related to this theme. A dozen publications obtained more than 100 citations, focusing, among others, on research related to the management of telework boundaries, employee health and performance, flexible work arrangements or the impact of occupational isolation and the intention to quit. Some of these authors were Kossek, Lautsch and Eaton [
73] with 378 citations; De Croon et al. [
78] with 214 citations; Hornung, Rousseau and Glaser [
80] with 213 citations; and Hill, Ferris and Martinson [
26] with 191 citations. On the other hand, the Performance (14 papers and 867 citations) and Organisation (15 papers and 1535 citations) clusters evolved in the second period (2010–2016) alongside Conflict, joining the Management cluster and then Flexibility in the third period.
In the case of the second period (2010–2016), there were five clusters:
Management,
Social Factors,
Technology,
Choice and
Context. The most relevant, both in terms of its centrality and the number of documents (28) and number of citations (2042) was
Management. This cluster, which is the result of the evolution of
Conflict in the first period, includes topics that converge in the interest of research on the flexibilisation of work, the balance between work and family life, job satisfaction or the impact of telework on burnout and work engagement. The main authors were Kelliher and Anderson [
75] with 255 citations; Staats and Gino [
81] with 92 citations; Morganson et al. [
82] with 83 citations; or Sardeshmukh, Sharma and Golden [
83] with 62 citations. Other relevant clusters next to Management were Social Factors (17 papers and 925 citations) and Technology (17 papers and 803 citations).
Finally, the clusters formed in the third period (2017–2021) were
Flexibility,
Technology,
Job-demands,
Family,
Strategies and
Behaviour. The
Flexibility cluster leads in both the number of papers (14) and citations (670). It is important to note that, despite the fact that the largest number of documents (214) was registered in this period, the fact that the cluster with the highest centrality and internal density or cohesion (internal strength of all the links between the words describing the topic) only has 14 documents, shows that there is a great dispersion of topics. This implies that these topics are not sufficiently related enough to create clusters with the co-occurrence established for the different periods. The most cited topics in this cluster of this period regarded research interested in the evaluation of telework growth, flexible working policies, stress, job satisfaction of teleworkers, work engagement, teamwork or aspects relating telework to its use in the period of the pandemic (COVID-19) that broke out in 2020. The authors who obtained the highest impact in their publications on topics related to the
Flexibility cluster were Felstead and Henseke [
84] with 70 citations; Chung and Van Der Horst [
85] with 44 citations; Charalampous et al. [
86] with 36 citations; Suh and Lee [
87] with 30 citations; and Waizenegger et al. [
39] with 24 citations.
In conclusion, regarding the evolution of telework-related research, from a longitudinal point of view, the results suggest that, although various clusters have been generated that bring together different perspectives of the study such as communication, analysis of the importance of commuting, the environment, social factors, technology or work demands, among others, the cluster that leads the research trend is the one marked by the longitudinal axis of Conflict–Management–Flexibility. This has brought together the largest number of papers and the highest level of impact.
3.4. Analysis of Strategy Maps by Period
The longitudinal analysis detailed which clusters or themes were the most present in each period and what evolution they underwent. The next step was to detect the importance of each cluster for the objectives of this research.
Figure 4—left shows the “Strategic Map” for the period 1984–2009. This figure shows which clusters are the main ones according to their centrality or external cohesion and their density or internal cohesion. Once this strategic map was visualized, the main themes were clearly exposed, which are those located in the upper right-hand corner of the map; these clusters are called “Driving Themes”.
Figure 4—right, shows the “Thematic Network” of the driving theme of this period 1984–2009, in this case, the one corresponding to the Conflict cluster. In this thematic network, it can be seen which topics were the most investigated and what relationship there is between them. With this dynamic, two important findings were obtained: on the one hand, we detected which was the main cluster that drove research in a given period, and, on the other hand, with the thematic network, we observed which topics are related to this cluster and what link there is between them.
To find the themes that led telework-related research in the period of 1984–2009, we focused on the driving themes of that period. These were
Conflict,
Organisation and
Performance (
Figure 4—left). Their centrality and density levels were 31.78/20.25, 32.51/18.35 and 26.08/17.31 (
Table 5). These are the highest with respect to the rest of the clusters that were configured in this first period, which means that the research focused mainly on analysing aspects that had to do with the internal network of these three clusters (
Table 6). The most prolific themes within these clusters were:
Flexibility,
Work–life-balance,
Conflict and
Gender, within the
Conflict cluster;
Home,
Time-use,
Management and
Organisation, within the
Organisation cluster; and
Performance,
HRM,
Workplace and
Challenges in the case of the
Performance cluster.
As for the rest of the clusters in this period, we found Communication, which is right in the centre of the matrix, where, a priori, it generates some doubts about its role within the research related to telework in this period; however, later, we will see how it is integrated into one of the driving themes of the following period (2010–2016). Its centrality and density were 22.87/15.82. Finally, the Travel and Environment themes fall into the so-called emerging or decadent themes, which, due to their centralities and densities of 5.01/10.42 and 2.48/13.10, it was uncertain whether they would evolve into other categories within the matrix or eventually disappear.
In order to gain a more detailed understanding of the themes that led the research in the first period, the network that integrates the
Conflict cluster, the main driving theme, was analysed. To this end, we took into account the weight of the density of the relationships of the existing dyads between the themes that make up this cluster (
Table 7). Among the fifteen dyads with the highest level of density in the network, the most important in terms of weight is
Conflict/Job-satisfaction (0.19). In addition,
Conflict is related to
Work–life-balance (0.14),
Support (0.13),
Family (0.12),
Stress (0.11),
Gender (0.09),
Flexibility (0.09),
Spillover (0.09) and
Outcomes (0.08). On the other hand, other important dyads were configured, such as
Work–life-balance with
Spillover (0.15) and
Flexibility (0.10);
Support with
Leadership (0.14) and
Job-satisfaction (0.09);
Job-satisfaction with
Family (0.09); and
Flexibility with
Gender (0.09).
For the period 2010–2016, the driving themes were
Management,
Social Factors and
Technology (
Figure 5—left). Their centrality and density levels were 77.94/26.70, 39.90/14.63, and 41.97/8.65, respectively (
Table 5).
In terms of the number of documents (
Table 8), the most common themes within these three clusters were
Work–life-balance,
Performance,
Flexibility, and
Conflict within the
Management cluster. The
Management theme, which lends its name to the cluster, is in fifth place in this ranking. In the
Social factor cluster, the most prolific themes were
Job-satisfaction,
Social factors,
Workplace and
Employees. Finally, in the
Technology cluster,
Technology,
Boundary-management,
Virtual-organisations and
Stress were the most popular.
With respect to other clusters in the period, as in the previous period, there were no peripheral or basic themes; in this case, Choice and Context were emerging or decadent themes, lacking low centrality and density (11.56/7.83 and 5.83/7.62, respectively), as well as fewer documents.
Regarding the internal analysis of the thematic network of the Management cluster, as the main driving theme, the most important dyad is the one that relates to the themes of Work–life-balance and Time-use (0.25) (
Table 9); in addition, among the main dyads, Work–life-balance is related to Conflict (0.22), Flexibility (0.18), Family (0.17), Home (0.10) and Management (0.08); Flexibility with Conflict (0.17), Home (0.13), Family (0.10) and Management (0.10); Management with Home (0.13) and Organisation (0.07); and, finally, Conflict with Family (0.11), Performance (0.08) and Support (0.07).
Finally, in the period of 2017–2021, the driving themes were
Flexibility and
Job-demands; however,
Technology, although located at the border of the matrix with the peripheral themes (
Figure 6—left) had a higher level of density, documents and impact than
Job-demands. Therefore, it can be considered a cluster that led the telework literature in this period. Its centrality and density levels were 83.70/33.18; 37.83/12.71 and 33.42/12.87, respectively (
Table 5).
In terms of the number of documents (
Table 10), the most frequently discussed topics within the
Flexibility cluster were
Work–life-balance,
Performance and
Flexibility, repeating from the previous period, and in fourth place was
Job-satisfaction. In the
Job-demand cluster,
Job-satisfaction,
Stress,
Resources and
Office were in first place. Finally, in the
Technology cluster, the most prolific themes were
Technology,
Social factors,
Communication and
Professional-isolation.
As for the other clusters in the period, Family was the basic theme with 40.34/4.01, and Behaviour, with a centrality and density of 7.30/4.44, was the declining or emerging theme. As for Strategies, it is on the borderline between the emerging decadent themes and the peripheral themes. Its centrality and density are 2.54/9.33.
With regard to the internal analysis of the themes of the
Flexibility cluster (
Table 11), the most important dyad is the one formed by the themes of
Performance and
Job-satisfaction (0.24); in addition to this relationship,
Performance is related to
Flexibility (0.16) and
Policy (0.12);
Employees is related to
Organizational-commitment (0.20) and
Flexibility (0.11);
Flexibility, which lends its name to the cluster, has a significant number of relationships with the most important ones; among them, it is related to
Management (0.20),
Performance (0.16),
Work–life-balance (0.13),
Job-satisfaction (0.12),
Conflict (0.09) and
Policy (0.09); on the other hand,
Work–life-balance is related to
Gender (0.17) and
Conflict (0.14);
Job-satisfaction to
Conflict (0.14);
Conflict to
Gender (0.13); and, finally,
Gender to
Time-use (0.10).