The Sociotechnical Approach to Work Organization
The Sociotechnical Approach to Work Organization
- David E. GuestDavid E. GuestSchool of Management and Business, King's College London
The sociotechnical approach, developed by psychologists at the Tavistock Institute of Human Relations in the 1950s, proposes that the design of work should seek to optimize both the social and the technical systems within organizations, offering a counter to ideas of technological determinism. It further suggests that organizations should be viewed as open systems, subject to sometimes unpredictable external and internal influences leading to a need for adaptability. The work group is viewed as the most relevant unit of analysis resulting in advocacy of autonomous work groups offering group members high levels of control over their work. Workers should participate in the design of their work and receive training and support to enable their involvement. This influential concept stimulated a large body of research in many countries. Despite some notable positive examples, outcomes were often mixed, reflecting the challenges of managing and sustaining significant change. The concept of joint optimization has also proved problematic, with psychologists tending to focus on the social system, while engineers give greater emphasis to the technical system. The advent of digital technologies is providing a new impetus to the need to design work to optimize both the social and technical systems, provoking renewed interest in the approach.
- Organizational and Institutional Psychology
Simply described, the sociotechnical approach proposes that the organization of work should seek the joint optimization of the social and the technical systems in the workplace to the mutual benefit of both the employer and employees. Within a wider interdisciplinary framework, it offers the potential to apply a set of psychological principles to the organization of work. The idea of a sociotechnical approach to the organization of work first emerged in the 1950s and was extensively elaborated and applied in organizations in the following decades. However, academic evaluations revealed mixed results, including an overemphasis on the social at the expense of the technical system, leading to a number of adaptations that have been applied in contemporary organizations and are described in the section “Shifting the Focus: Giving Primacy to the Technical System.” This review analyzes the nature, impact, evolution, and continuing viability of the sociotechnical systems approach as well as the challenges of introducing what can be viewed as quite radical changes in ways of working into complex organizational systems.
The Origins of the Sociotechnical Approach
The Coal Mining Study Reported by Trist and Bamforth (1951)
The idea of a sociotechnical approach to work organization can be traced to an article by Trist and Bamforth (1951) published in the journal Human Relations. This reported a study in the U.K. coal mining industry where a new and more efficient approach to extracting coal, the so-called longwall method, was replacing the traditional hand-got, small group-based approach. This reflected a significant technological advance and promised considerable improvements in productivity as well as opportunities for closer supervisory monitoring of work. However, this resulted in dissatisfaction among the miners because it had considerably narrowed roles and broken up the established work groups. In the context of the dangerous work at the coalface, these groups had consisted of workers who trusted each other and could be relied on to ensure safety and support. Alongside this, there was increased absenteeism, labor turnover, and industrial unrest and a failure to make the anticipated productivity gains. The researchers found one mine in the coalfield where a different approach had been adopted, which they termed the composite longwall system. This retained the intact work groups and many of the individual skills and provided the groups with considerable autonomy. While this did not appear to reflect the most efficient use of the technology, it resulted in higher worker satisfaction, better health indicators in a context where stress and ill health was a chronic problem among miners, and improvements in productivity. It is therefore an example of mutual gains for both management and workers.
Trist, Bamforth, and their colleagues at the Tavistock Institute of Human Relations drew several lessons from this study. First, they noted that organizations comprised both technical and social systems and that maximizing the technical system while ignoring the social system resulted in costs that were likely to limit any gains that the technical system offered. What was therefore required was an approach that optimized both the technical and the social system, namely a sociotechnical system. As Trist and Bamforth (1951, p. 38) put it, organizations should provide “responsible autonomy to primary groups throughout the system and ensuring that each of these groups has a satisfying sub-whole as its work task and some scope for flexibility in work pace.” In short, sociotechnical work organization challenged technological determinism by requiring consideration of both the demands of technology and attention to social factors such as the division of labor, skill utilization, and responsible worker autonomy, with the work group as a primary unit of analysis.
The Background to the Emergence of Sociotechnical Work Organization
To understand why the sociotechnical concept seemed novel at the time, it is helpful to consider the wider context. Within industry, scientific management, advanced by Taylor, held sway (Taylor, 1947). This proposed that productivity gains could be achieved by advances in technology allied to simplification of work and by the organization of work to ensure that workers were subjugated to the demands of the technology. Techniques such as time and motion study were introduced with the aim of ensuring “a fair day’s work” for “a fair day’s pay” based on a careful analysis of the work of a “first class man.” The aim was to arrive at the most efficient division of labor and organization of work. It was assumed that workers were extrinsically motivated and that financial incentives would therefore provide the best form of motivation to ensure high performance. The role of supervisors was to monitor performance. Taylor’s assumption was that this would result in a fair effort-reward bargain. However, views about fairness differed and use of financial incentives often provided workers with the control to determine output leading to the potential for restriction of output and other forms of industrial conflict in an attempt to negotiate a better rate for the job. Accounts emerged of a range of ways in which workers controlled output and used group pressures to ensure a degree of workgroup conformity (Lupton, 1963; Roy, 1952; Whyte, 1955), challenging the efficacy of the scientific management approach.
In parallel with, and partly in reaction to the application of scientific management, psychologists and other social scientists developed an alternative view of work organization with what came to be known at the human relations movement. They emphasized the importance of the social system as the major influence on workers’ behavior, producing evidence to show that a supportive social system resulted in superior performance to the approach associated with scientific management. Most notably, the Hawthorne Studies (Roethlisberger & Dickson, 1939) appeared to demonstrate that the work group had a major influence on individual worker performance, confirming the role of the group as a key unit of analysis that could influence increases in performance but also restriction of output. An influential experiment by Lewin et al. (1939) revealed the benefits of participative leadership, while Coch and French (1948) reported that group participation could help to overcome resistance to change. However, these studies were not without their critics. For example, a follow-up study by French et al. (1960) in a Norwegian factory failed to replicate the original study by Coch and French, partly because it went against the grain of traditional negotiations with unions about change that were perceived as more likely to ensure protection of pay levels. Perhaps more tellingly, a study by Strauss, reported in Whyte (1955) provides an account of how a group involved in paint spraying of dolls were given the autonomy to determine the speed of their production line. Soon their productivity soared and they were earning more than skilled workers. However, they were part of a larger production line and created problems for those who came before and after them in the production process to such an extent that the experiment was abandoned. As a result, 75% of the group left the company within a month. The important lesson is that group autonomy must be viewed within the context of the wider production and social system. This is one of the central tenets of the sociotechnical systems approach.
The Evolution of the Sociotechnical Approach
A group of social scientists, mainly based at the Tavistock Institute of Human Relations in London, was largely responsible for the development and initial application of the sociotechnical systems approach. They developed a number of general principles that informed the approach and a set of more specific guidelines and assumptions on which to base interventions in organizations.
The General Principles of Sociotechnical Work Organization
The principle of organizational choice
This challenges technological determinism and argues that there is always a choice of design of work arrangements. This principle is elaborated by Trist et al. (1963) in their book Organizational Choice. There had been some consideration about whether greater weight should be given to the economic system but this was dropped on the grounds that it was not central to the organization of work. The core choice therefore centered around the relationship between the technical and the social system reflecting the view that work systems should be designed to achieve both high performance and high worker well-being.
Organizations as open systems
Advocates of the sociotechnical approach were influenced by the work of Von Bertalanffy (1950) who proposed an open system in the analysis of physics and biology. Adapting this to the organizational context, organizations were viewed as systems with inputs, processes and outputs. The significance of this was that changes to external circumstances would be likely to influence internal processes (Emery & Trist, 1965). It was therefore inappropriate to view the organization and its internal systems as closed or independent. It also meant that organizations needed to be adaptable since neither the technological nor the social system were immutable.
The principle of joint optimization
While organizations operate in open systems with a range of external and internal influences requiring adaptability and scope for organizational choice, a major goal for achieving the best outcomes is to ensure optimization of both the technical and the social systems. This may mean that neither receives primacy and that some compromises have to be made based on the assumption that the benefits of joint optimization outweigh the costs of giving primacy to one system, usually the technical system, at the expense of the other.
Work groups as the primary unit of analysis
While it was accepted that analysis could be conducted at a number of different levels within an organizational system, it was proposed the key level of analysis was the work group subsystem where the main interface between the technical and social systems was likely to occur. One telling example of this was provided in a series of studies in Norway by Emery and Thorsrud (1976). Their initial investigation of the appointment of worker representatives to boards of companies (Emery & Thorsrud, 1969) found that this had little or no impact on the experience of workers on the shop floor justifying their shift in focus to the work group and work unit.
Action research and participation
Researchers at the Tavistock Institute had been heavily influenced by the work of Lewin (1947) who had conducted experiments showing how social groups could influence behavior through what was described as action research. Specifically, he had shown how eating habits could be changed through behavioral commitment and how leadership could influence group behavior. The study of Coch and French (1948) had appeared to show how group participation could help to overcome resistance to change. The lessons that were taken from these studies were that the role of a change agent could facilitate change and that ownership of change by a group could be a powerful way of generating commitment to new ways of working.
Principles of Sociotechnical Design of Jobs and Work Units
Drawing on the experience gained from early studies of sociotechnical design and other studies conducted by the Tavistock Institute, Emery set out more specific principles for the design of jobs and work groups (Emery & Thorsrud, 1976). These started from the premise that workers had certain “needs” that should be met at work to avoid alienation and provide scope for fulfilling work. These consisted of:
The need for a job to be reasonably demanding/challenging.
The need to learn and go on learning on the job.
The need for some area of decision-making.
The need for a minimal degree of helpfulness and recognition.
The need to be able to relate work to social life.
The need to feel the job leads to some sort of desirable future.
While this is a plausible list and provides some potentially useful guidelines, it does not reflect any established notion of psychological needs. An important feature of the list is the emphasis on an optimal amount of responsibility, decision-making, and so on; not too much and not too little. There are parallels here with the job demands—resources models of Karasek (1979) and Bakker and Demerouti (2017) in seeking to find a balance and at the same time account for individual differences and competences.
The set of needs were translated by Emery and Thorsrud (1976) into specific principles for work design. First, with respect to individual jobs:
There should be an optimum variety of tasks.
There should be a pattern of tasks that provides a sense of a meaningful whole.
There should be an optimum length of a work cycle.
Workers should have some scope to determine the quantity and quality of the work.
The job should include and carry some responsibility for preparatory and boundary tasks.
The tasks and the job should command some respect in the community.
The job should make some perceptible contribution to the utility of the product for the consumer.
Again, the emphasis is on an optimum level of activity rather than any risk of overreaching through levels of variety or responsibility that could lead to excessive demand. Since the major interest lay in the group as a central unit within the social system, the job design principles were extended to the design of activities for what Trist and others described as primary work groups and came to be known as autonomous work groups. In practice, advocates acknowledged that almost no work groups were entirely autonomous so that the more correct term was a semiautonomous work group. The principles set out by Emery and Thorsrud (1976) for their design included:
There should be provision for “interlocking” tasks, job rotation, and physical proximity where there is a necessary interdependence of jobs, where individual jobs entail a relatively high degree of stress, and where individual jobs don’t make a discernible contribution to the end product.
Groups should, as an extension of the principle for individual job design, have an overall task that makes an identifiable contribution, have scope for setting standards and receiving feedback on results, allocate roles within the group, and have some control over boundary tasks. At the heart of the concept of the autonomous work group is therefore the concept of self-regulation by the group members.
Cherns (1976, 1987) set out an overlapping set of sociotechnical principles for work design that address concerns primarily at the organizational level. His 1987 paper was a development and essentially a clarification and extension of his earlier paper based on experience in the intervening decade. His principles included:
Compatibility of design with the aims of the system.
Minimal critical specification to allow scope for adaptation and learning.
Variance control as close as possible to the potential source of the variance.
Appropriate boundary location. This calls for the establishment of boundaries between work groups and between groups and management as well as boundary management. This could be a role for members of the group within the boundary or a role for supervisors at the boundary.
Information flow to ensure that relevant information reaches those who need it.
Support congruence to ensure that other organizational systems and notably human resource systems, such as rewards, are compatible with the organization design.
Power and authority should operate to ensure that there is access to, and responsibility for materials, resources, and performance at the appropriate level and in the appropriate units or groups in the organization.
Incompletion, which can be coupled with a new 1987 principle of transitional organizations, to reflect a view that design of organizations is a continuous process.
A concern for quality of working life as one of the goals of the system.
This set of principles goes well beyond what might be viewed as the territory of sociotechnical systems. Indeed, it is directed in part at consultants and change agents drawing on Cherns’s own experience and that of others such as Davis (1977), an engineer who developed an interest in sociotechnical principles.
It can be seen from analysis of these lists of principles for sociotechnical design set out by its advocates that they are fluid and dynamic and reflect experience as action researchers who have engaged in consultancy to contribute to both the design and evaluation of sociotechnical systems at different levels within organizations. The different experiences result in varying emphases which from time to time have been drawn together, more particularly by the team of Tavistock action researchers (Trist et al., 1963; Trist et al., 1993). Over time, different action researchers have extensively adapted or extended the sociotechnical principles to meet different conditions.
Having set out the rather disparate principles for sociotechnical design, the next section reviews its impact, first at the workplace level and then at the organizational level.
Outcomes of the Sociotechnical Design of Workplaces and Organizations
It is difficult to provide a systematic evaluation of sociotechnical interventions due to the variety of contexts, interventions, and measures. Since many of the widely known interventions are presented as extended case studies, this section will describe and evaluate some of these “demonstration” examples before drawing together the lessons that emerge.
The Coal Mining Studies
The original coal mining studies in the Yorkshire coalfield (Trist & Bamforth, 1951) that had led to the idea of a sociotechnical system with autonomous work groups were followed up when more examples were found of similar work arrangements in other U.K. mines which are described in Trist et al. (1963) and summarized in Trist (1993). In the Nottingham coalfield, partly at the initiative of a local area manager, autonomous work groups of 20 to 25 miners carried out all the tasks on a shift. All were multiskilled and all received the same pay. Trist reports that “productivity and work satisfaction were unusually and consistently high” (1993, p. 44). Steps were taken to extend the approach to other mines in the area. In the Durham coalfield, an example of large autonomous work groups was found. Over a four-year project, a complete colliery changed to the “composite” form of group working. Evaluations revealed that output was 25% higher with lower costs compared with conventional methods, while accidents, sickness, and absenteeism were halved (Trist et al., 1963). It was also possible to reduce layers of management, reflecting the system-wide implications. Despite the acknowledged success of autonomous group working in each of the mining studies, the researchers encountered what was to become a familiar problem. In both cases, the Divisional Boards blocked further development, partly, it was argued, because of fear of giving too much autonomy to miners and partly because of their belief that technology should have primacy. Following the Durham research, both the National Coal Board and the National Union of Miners, while acknowledging the success of autonomous group working, were giving priority to negotiations over pit closures and neither wanted to take the approach further.
Some years later, Trist participated in a study exploring the possibility of introducing autonomous group working in an American mine (Trist et al., 1977). The union, which had concerns about safety in the mine, supported an experiment to introduce autonomous work groups, and top management was also enthusiastic. The action research project involved extensive preparatory processes, two experimental autonomous work groups, and control groups. The evaluation showed that the first autonomous work group in particular was successful in achieving comparatively fewer safety violations and accidents, somewhat lower absence, lower maintenance costs, and higher productivity. Members of the group reported higher autonomy and decision latitude, a reduction in pressure from management, and generally less stress and tiredness at the end of the working day, all reflecting improvement in their quality of working life. Despite these indications of success, when it was proposed to extend autonomous group work to the entire mine, this was narrowly voted down by the local union members. In the following months the autonomous work groups disintegrated and the mine experienced several short strikes. The subsequent evaluation identified a number of reasons for this negative outcome. First, there was relatively low trust in management’s aims allied to suspicion that autonomous work groups would be used to derecognize the union. Secondly, there was jealousy of the privileges accorded in particular to the first group members who were able, for example, to attend meetings outside the mine. Third, there was resentment among experienced miners when newcomers were drafted into the second autonomous work group and received higher pay than their experienced colleagues elsewhere in the mine. Finally, there was opposition from middle management who perceived the innovations as a threat to their role and status. The lessons from this study concern the importance of being sensitive to, and taking account of the wider systems implications including traditional status hierarchies, issues of perceived fairness, and the history of labor relations.
The Ahmedabad Experiment
Rice (1953, 1955, 1963), another member of the Tavistock Institute of Human Relations, was invited to visit the Calico Mills in Ahmedabad, India where he raised the idea of a sociotechnical approach, and specifically the possibility that a group of workers should become responsible for the whole process for a sizeable set of looms in the automatic loom shed. The workers themselves developed the approach which was successfully implemented with management support and subsequently transferred to other loom sheds in the large 9,000-worker plant. The application of autonomous group working resulted in higher productivity and higher worker satisfaction as well as significant gains in efficiency and reduced maintenance costs. It should be noted that there were also increases in basic pay and financial incentives which one critical reviewer suggested could account for the performance gains (Roy, 1969).
Miller (1975) conducted a follow-up 15 years later. He found that workers in the original experimental shed had continued to operate as an autonomous work group throughout the period producing high performance and displaying a strong group identity and high morale. However, the workers in the other sheds had largely reverted to the original work system. Miller offers a number of explanations for this. First, he notes that the original experimental site had retained many of the original group members whereas the others had experienced high turnover drawing in new workers who were not familiar with group working. Second, the experimental group had continued to undertake the same kinds of weaving and yarn whereas other groups had seen constant changes in line with changes in fashion. Third, as the business entered new markets, there were demands for higher efficiency along with regular changes in market requirements. One consequence was that financial incentives were geared more strongly to output at the expense of good maintenance creating a vicious circle as machines broke down more often resulting in extra pressure, which management sought to control. Finally, and linked to these pressures, boundary management, which was an essential feature of autonomous work groups, was neglected at all levels within the organization, partly reflecting a loss of interest among senior management in autonomous group working. These points illustrate the importance of an open systems perspective and the comparison between the calm environment for the sustained success of the experimental group and the turbulent environment that led to the gradual collapse of group working at the other sites.
The Norwegian Studies
A set of Norwegian studies applied a sociotechnical systems approach with the broad aim of extending industrial democracy on the shop floor. They were intended to provide demonstration cases for possible wider national application and the program of action research had support at the national level from business and unions. They followed the evidence that appointing worker directors to company boards had no perceptible impact on industrial democracy for workers on the ground (Emery & Thorsrud, 1969). As a result, an action research approach based on sociotechnical principles, with considerable involvement at the local level, including workers’ representatives, was adopted. Four organizations provided the experimental context. The sociotechnical principles outlined by Emery and Thorsrud (1976) were applied with a core focus on identifying the tasks for autonomous groups, including management of variances, developing systems of self-regulation and managing boundary conditions.
The first study in the wire-drawing department of the metal manufacturing company, Christiania Spigerverk, was not a success. A change from one worker to one machine to a group of workers looking after a set of machines was proposed but a number of workers were reluctant to take part in the study, viewing it as something being imposed on them. Group working was eventually agreed and productivity role by 20%. However, this also meant an increase in the agreed bonus, which caused dissatisfaction elsewhere in the plant. Therefore, although workers directly involved were happy with the new arrangements, management felt unable to extend group working throughout the plant and the experimental group returned to the old ways of working. This highlights once again the systems implications and the importance of gaining commitment at all levels in the organization to support the changes. In this case, relevant features of the wider system include industrial relations and rewards, job security, and the levels of trust between workers, management, and sometimes between workers and their union, as well as suspicion of the role of the researchers. These lessons were taken forward into the three subsequent studies.
The second field experiment involved a thirty-worker section of the Hunsfos paper mill. The research team designed a group scheme involving greater local decision-making, training to upgrade skills and enable workers to undertake all the tasks in the section, greater internal communication and information provision, and a move from a complex individual reward system to a local departmental bonus scheme. Once again, the core principle was to create autonomous group working by a self-managed team. Following initial skepticism among some workers, partly reflecting concerns about the role of the researchers, implementation occurred once a local action committee was formed to take the project forward. The result was an improvement in quality and output levels and enthusiasm among the workers to continue working under the new system. The experiment had the strong support of the local union and management and over a number of years the sociotechnical approach spread successfully throughout the plant.
A third study was conducted at Nobo in a new department of 30 workers manufacturing electric panel heaters. The work was largely unskilled and organized along scientific management principles with piece-rate payment. Management invited the researchers to set up a sociotechnical system of working. Training enabled workers to become skilled to undertake several tasks, coordination roles were established, and some support tasks were incorporated enabling autonomous group working. Piecework was replaced by fixed basic pay plus a group bonus. After ten weeks, productivity had increased by 20% and continued to rise further. Most workers were happy with the new system of group working and it spread successfully to the rest of the 100-worker site.
The final study took place at the Norsk Hydro fertilizer company. The impetus for the experiment lay in the challenging financial circumstances the company faced and the poor industrial relations at the plant. Initial discussions centered on pay and job security, both of which had to be resolved before any progress on the design of work was possible. A new pay scheme linking payment to the acquisition of skills provided an incentive for team members to become multiskilled. The multiskilling enabled job rotation and incorporation of many maintenance tasks facilitating the development of small autonomous work groups. A group bonus was linked to quantity and quality of output. The outcome was judged to be a success and the sociotechnical principles were then applied to the design of a new plant. The initial proposal, based on traditional scientific management lines, had been for a workforce of 94 undertaking a range of specialist tasks. The new proposal suggested that 56 workers would be sufficient, and 60 were eventually agreed. The new plant was successfully set up based on the sociotechnical design principles resulting in high levels of output and high worker satisfaction. This project was viewed as a success and it was agreed to diffuse the approach throughout the organization. But a range of external factors limited progress. These included changes in markets and in strategic priorities as well as the considerable complexity of trying to bring about this type of change in one of the largest companies in Norway.
These Norwegian examples indicate that the challenge lies not so much in the application of sociotechnical principles for the design of autonomous group working but in gaining commitment to change, in creating and sustaining ownership of the new system of working, in building in resilience when external factors present possible threats and in achieving diffusion to other sites. In all this, the role of external consultants and researchers is considered essential in facilitating change through their action research activity but also potentially problematic if they are perceived as a threat or if they become too central to the success of the whole exercise.
Assumptions about potential diffusion of the lessons from these field experiments was a central feature of support for them at the national level. The broad aim was to use these demonstration cases as examples to spread the application of industrial democracy and sociotechnical work organization throughout Norwegian industry. It largely failed in this endeavor. As Bolweg (1976) notes, despite considerable efforts by the joint union and management national-level Cooperation Council, only 32 companies tried to introduce changes along the lines suggested and only five managed to produce sustained change. Bolweg describes one of these, Norsk Medisinaldepot, which was particularly successful, but it was very much the exception to the rule. In a balanced assessment, Qvale (1976), who was involved in some of the projects, concludes that as demonstration cases to promote industrial democracy on the shop floor they were a modest success but as the basis for diffusion they were a failure.
Shell’s New Philosophy of Management
In 1965, the top management at Shell Refining Company in the U.K., part of the larger Shell Group, agreed to develop a new philosophy of management designed to address a range of industrial relations problems and, on a larger scale, help it to cope with an increasingly uncertain external environment. They invited social scientists from the Tavistock Institute to assist with this process. Emery and Trist (1965) had just published what came to be regarded as a major statement about the need to adopt an open systems approach to the analysis of organizations. They argued that management faced increasing challenges from a turbulent and uncertain external environment which lay largely beyond their control. This required organizational structures that were flexible and adaptable which in turn depended on the commitment of employees at all levels within the organization. The idea of the new philosophy of management was to set out agreed principles on which this might be achieved. The subsequent stage was to put it into practice applying sociotechnical principles of work design.
Hill (1971), the main internal Shell consultant involved in the project, has provided an account of the development and implementation of the new philosophy. A document was initially drafted with the help of Tavistock social scientists and over a period of four years, extensively communicated through a series of workshops to management and employees throughout the organization. At the Stanlow Refinery, where the local manager was enthusiastic about the new philosophy, three “demonstration” projects in work redesign were introduced, but none survived. A later project, drawing on the lessons from these early projects, used sociotechnical principles in the microwax department with greater success resulting in higher productivity and lower absence (Burden, 1975). The philosophy also influenced a more positive industrial relations climate, reflected in a sequence of productivity deals, and influenced decisions about the organization of work based on sociotechnical principles at a new Teesside refinery (Hill, 1971). A highly critical account of the application of the philosophy by Blackler and Brown (1980) challenged the wider commitment to the philosophy within the company, suggesting that many of the initiatives lost their impetus over time while those that did get successfully implemented depended on the enthusiasm of local managers rather than any wider commitment to the philosophy. Despite this critical analysis, this company-wide initiative to apply sociotechnical principles provides an interesting case study of the potential for their application at this level which has some similarities to the Norwegian Norsk Hydro example. One indication of its impact is that the approach was subsequently adopted within some success at Shell plants in Canada and Australia (Trist, 1981).
The Sheffield Studies
Although the various case studies largely inspired by researchers from the Tavistock Institute have been widely cited and highly influential in promoting interest in the concepts of sociotechnical systems, autonomous work groups, and in the wider quality of working life, they lack the kind of evaluative rigor that is normally associated with research in the field of work and organizational psychology (Cummings et al., 1977; Lawler, 1977). Researchers at the Sheffield University Institute of Work Psychology sought to remedy this. Their first study evaluated the introduction of autonomous work groups in an established confectionary factory (Wall & Clegg, 1981). Workers in sections manufacturing and packing boiled sweets reported low levels of satisfaction, mental health, and performance. Following extensive diagnosis, a reorganization removed the physical separation of the two sections and instead created two groups responsible for all processes. This involved the removal of supervisors and a chargehand. Data collected prior to the change, and six and eighteen months afterward revealed perceived increases in work group identity and autonomy and increases in intrinsic motivation and performance between the pre- and postchange periods, which were maintained at Time 3. Job satisfaction and mental health also improved, although the larger changes occurred between Times 2 and 3. Labor turnover reduced significantly compared with other sections of the plant. This study is therefore evidence of the successful application of autonomous work groups although it had some unanticipated systems implications. Engineering staff, and in particular maintenance workers, were put under pressure to improve the reliability of the machinery and, as production increased, marketing staff were under pressure to sell more sweets.
The next Sheffield study was a more ambitious quasi-experimental evaluation of the introduction of autonomous group working at a new confectionary manufacturing plant where the sociotechnical system was designed with the help of a Tavistock Institute consultant (Wall et al., 1986). The impact was monitored over three years and compared with a traditionally organized plant with extensive data collected at three time points. The results were very mixed. They revealed that perceptions of autonomy among members of autonomous work groups were high and there was an increase over time and by comparison with the other plant in both intrinsic and extrinsic satisfaction. However, there was no increase in intrinsic motivation or in well-being and there was higher labor turnover and more cases of dismissal under conditions of autonomous group working. It took almost two years to reach production targets, after which production levels were maintained. One indirect benefit was that labor costs were reduced by the absence of supervisors. While the authors offer plausible explanations for the higher labor turnover (a tighter local labor market) and the poorer disciplinary record (the work groups were reluctant to act against poor team-member behavior until it was too late and management had to step in) the overall results of this methodologically more rigorous study raise questions about the efficacy of autonomous work groups in relatively routine production contexts.
A Preliminary Assessment
The well-known examples of the application of the aforementioned sociotechnical systems approach had a considerable impact on the social sciences and on wider debates on the quality of working life (see, for example, Davis & Cherns, 1975). They stimulated a considerable number of studies in North America and Europe that purported to be influenced by sociotechnical systems thinking. By 1982, Pasmore et al. (1982) were able to review 134 “experiments.” They reported that 53% involved autonomous work groups but only 16% involved technological change. Authors claimed that most of the studies were successful with 60 per cent reporting productivity increases, 54% reporting more positive attitudes and 27% reporting cost savings. Changes involving the technical system were the least likely to report productivity increases. Pasmore et al. noted that there was a clear bias toward reporting positive results and an often poor research design. This point is reinforced in a review by Cummings et al. (1977) who, using stricter criteria for inclusion, reviewed 58 experiments in work redesign. They note that while most studies report positive results, most also have serious design limitations such as lack of control groups or appropriate statistical tests, questionable measures, and a tendency to select what were identified as positive groups for study. They outline ten steps to ensure future high-quality research. Finally, Kelly (1978) has provided a highly critical review of the main sociotechnical case studies, concluding that they allowed only a limited degree of autonomy, failed to address joint optimization effectively, and underplayed the role of financial incentives. However, he does acknowledge that there is evidence for the benefits of autonomous work groups. Pasmore et al. (1982) conclude that “the success of the interventions may have more to do with changes in the social system and in the qualifications of personnel than with joint optimization of social and technical systems” (p. 1195). This raises the important question of whether psychologists and other social scientists are best placed to introduce a sociotechnical system without the active collaboration of engineers or other technology experts.
The previous section has described and evaluated some of the best-known examples of the application of the sociotechnical systems approach to the redesign of work. It is notable that most of these studies took place over three decades following the seminal paper by Trist and Bamforth (1951). They have stimulated the large body of research reviewed by Pasmore et al. (1982) and Cummings et al. (1977). Trist et al. (1993) brought together many of the most important publications addressing sociotechnical systems theory and practice and offered another opportunity to evaluate its impact. The book received two full and contrasting reviews in Human Relations. Pasmore (1995), while acknowledging some of the challenges, describes sociotechnical systems as “a paradigm which had a greater impact on the ‘practical affairs of man’ than almost any other in the social sciences” (p. 15) and suggests that “clearly, socio-technical systems theory remains as valid today as it was in the 1950s” (p. 15). The second review by Scarborough (1995) is much more critical. He raises many familiar concerns such as the neglect of power and politics, the validity of the systems concept of organizations, the dominance of action research at the expense of rigorous research and theory development, and the ambiguity of some of the language. Indeed, he characterizes sociotechnical thinking as “a kind of intellectual Esperanto: a language championed by a brave and enthusiastic few, but actually spoken by no one else” (p. 32). More broadly, he believes that “as an intellectual project [sociotechnical analysis] seems at best stalled, at worst moribund” (p. 24).
Both reviews overstate their case. Pasmore overstates the influence and robustness of the sociotechnical systems approach while Scarborough is too negative. The impact of the approach and its link to concepts of industrial democracy and quality of working life owe much to the spirit of the times. There were extensive debates about alienation, rising expectations among an increasingly well-educated workforce, and associated concerns about low motivation and poor productivity (Work in America 1973; Walton, 1982). By the mid-1970s some of these concerns had been superseded in the eyes of managers by the first oil crisis and its consequences. A notable feature of the main examples provided above is that they all involved social scientists from the Tavistock Institute of Human Relations. They also involved a form of action research with a strong “normative re-educative” (Bennis et al., 1961) component and tended to depend on support from local champions. When the social scientists left and the champions moved on, the chances of sustainability declined, though there were exceptions to this, notably in the initial experimental plant at Ahmedabad.
Since the 1980s, research by work and organizational psychologists using the sociotechnical systems paradigm has greatly declined. Nevertheless, relevant insights that draw on the sociotechnical approach have appeared from time to time. For example, an important feature of sociotechnical analysis concerns variances and the argument that these should be handled at the point where they arise. Wall et al. (2002) proposed a contingency approach suggesting that workgroup autonomy is most effective under conditions of uncertainty. Griffin et al. (2007) define uncertainty as a contextual variable reflected in lack of predictability about the inputs, processes, and outputs. Cordery et al. (2010) tested the proposition in Australia with 17 wastewater treatment teams that faced different levels of complexity and uncertainty in their work. Prior to any changes, greater uncertainty was associated with poorer performance. However, after autonomous group working was introduced, uncertainty was associated with higher performance, a result partly attributed to the ability of the team to deal with variances on the spot rather than referring them up the hierarchy. While the focus on uncertainty and variances may help some of the different results of studies of autonomous team working, it is notable that the Cordery et al. study altered the social system rather than the technical system. As the cited reviews have noted, while these studies have merit as examples of work redesign, the neglect of the technical system is a common limitation in tests of the sociotechnical approach. The next section therefore turns to look at research that gives greater primacy to the technical system. One consequence is that the contribution from work and organizational psychology is less central.
Shifting the Focus: Giving Primacy to the Technical System
The relative weakness of the technical dimension in much of the initial application of sociotechnical systems is perhaps not surprising when the changes and their evaluation are led by psychologists and other social scientists who are more familiar with analysis of the social system. Arguably, if joint optimization of social and technical systems is to be achieved, production engineers are an important constituency to win over to the concept. An early example is Davis, an engineer by training, who had recognized that technological determinism made assumptions about workers that accorded them a minimal role in organizations, with consequent dysfunctions (Davis et al., 1955). He went on to champion use of sociotechnical systems design to promote the quality of working life (Davis, 1977; Davis & Cherns, 1975). An engineering perspective has subsequently played a prominent role in the application of a sociotechnical approach although the terminology sometimes differs.
The Case of Volvo
Engineers had a central role in one of the most famous examples of a new approach to the design of work provided by Volvo, the Swedish car manufacturing company. In the 1960s and 1970s, they faced the challenges of high absenteeism and labor turnover among workers engaged in routine production work. Following some initial activities in existing plants to increase worker participation and adapting the technology to provide more variety and autonomy for production workers, it was accepted that the existing technology limited the scope for radical redesign of work. Supported by the company president, Gyllenhammar, Volvo adopted a radical approach to the manufacture of vehicles at the new Kalmar plant. Instead of the traditional moving assembly line, what they sought to achieve, in the words of Gyllenhammar was
a manufacturing facility which, without any sacrifice of efficiency or profitability, will give employees the opportunity to work in groups, to communicate freely with each other, to shift between jobs, to feel genuine identification with the product and a responsibility for quality, and to influence their working environment (Lindholm & Norstedt, 1975, p. 63).
Groups of 15 to 20 worked on sections of a car such as the chassis, the engine, and the gearbox. The new plant was radically designed to provide cells for small work groups although it still required a final production system where groups had responsibility for sub-sections such as the electrics or the steering mechanism. A docking system provided buffers that removed the pressures of conventional assembly. While subsections of the assembly typically took three minutes, it was possible for a worker to learn how to do all the elements and therefore assemble the entire subsection of a vehicle over 30 minutes. Therefore, without using the language of sociotechnical systems, the Kalmar factory provides an example of how innovative production engineering can promote the quality of work and address concerns about worker dissatisfaction. In an initial evaluation, Lindholm and Norstedt claimed that the Kalmar factory achieved standard production efficiency targets within six months of its opening in 1974 while also providing high levels of job satisfaction.
Volvo took the concept of autonomous group working a step further at the Uddevalla car plant (Sandberg, 1995). Its distinctive feature was that workers had the ability to complete all elements in the manufacture of a car. The cycle times which in traditional car plants might be one or two minutes were between one and three-and-a-half hours. Typically, a group of nine workers would assemble a complete car. There was flexibility to work either in a small group or individually. For this to be successful, a key element was continuous learning at the individual level and also at the plant level. As workers became familiar with the process, production speeded up and quality improved. Reviews suggested that production levels were comparable with the main traditionally organized Volvo plant at Torslanda, Gothenburg. However just as production reached high levels, Volvo took the decision to close the plant in 1993, four years after it opened. One year later they also closed Kalmar. A number of external factors accounted for this including rising unemployment in Sweden, which had become a focus of trade union attention, and a decline in the market for Volvo cars. The company, now under a new president, decided it was simpler to shut these smaller plants than the large traditional plant. While those involved in the Volvo experience did not describe these initiatives in sociotechnical terms, they were, for all intents and purposes, examples of plant-level sociotechnical design and operation. Indeed, Buchanan (1997, p. 89) noted “Volvo’s retreat was thus damaging to the sociotechnical systems movement in that it had become an exemplar for organisations in many other sectors.” Despite some further initiatives in work design in the auto industry such as General Motors’ Saturn plant, the analysis by Womack et al. (1990) of lean production at Toyota reflected a resurgence of technological determinism and the decline of sociotechnical experimentation in the industry.
The Dutch Approach
In parallel with developments in Scandinavia, other European countries such as Germany and Italy had begun to support activities in work redesign that were heavily influenced by the sociotechnical systems approach. It is probably in the Netherlands that the most distinctive developments occurred. In the 1970s, Philips had begun a series of experiments in what they termed “work restructuring,” which bore similarities to sociotechnical design. While these had some success, they often failed to diffuse beyond the context in which they had taken place. Influenced by this experience as well as by the Norwegian emphasis on industrial democracy and the Swedish use of production engineering expertise, Dutch researchers developed their own adapted version of a sociotechnical systems application (de Sitter et al.,1997). They suggest it “can be viewed as a Dutch variant of sociotechnical systems design” (de Sitter et al., 1997, p. 497). At the same time, they are critical of certain features of sociotechnical systems analysis including what they view as an artificial and in practice unsustainable distinction between social and technical systems and an oversimplification of the concept of joint optimization that lacked guidance on how this can be achieved. They are also critical of the use of outside experts at the expense of embedded local knowledge. They suggest this partly accounts for the lack of diffusion of the sociotechnical projects both within and across organizations, a point also emphasized by Pava (1986) and Engelstad and Gustavsen (1993). The Dutch contribution is the concept of “Integral Organizational Renewal” (IOR) subsequently also labeled “modern sociotechnical thinking” (Kopp et al., 2019).
IOR is described as “a systematic approach to design which supports improvements in both the quality of work and what is called “the quality of the organization” (i.e., its ability to deal with a complex and continuously changing environment,” de Sitter et al., 1997, p. 498). In acknowledging the importance of the environment, they are supporting the analysis of open systems by Emery and Trist (1965) that highlighted the need to design organizations that can cope with a turbulent and uncertain external environment. A core feature of IOR is its emphasis on the organization as the key unit of analysis and the principle that organizations can be simple, but jobs are complex. There are a set of design principles such as a focus on integral design, on understanding controllability, on control structures, and on structural parameters that add up to a potentially complex set of requirements. Central to the approach is a participative strategy of change where experts aid in the process including providing intensive training in design so that the workforce, both workers and managers, can own the design with a principle of organizing a structure from the bottom up in a way that maximizes the quality of work as well as meeting the technical requirements for organizational design. Combined with the need for support from senior management and works councils, the aim is to embed an organizational design that can respond to external uncertainties, enhance productivity and also ensure the quality of work. Activities reflecting this approach have been coordinated through The Netherlands Institute for the Improvement of the Quality of Work and Organization (NKWO) and de Sitter et al. (1997) describe examples of its application across a range of contexts. Nevertheless, despite the greater focus of this approach on managerial priorities, Kopp et al. (2019) acknowledge that there are still only relatively limited examples of its application.
The Dutch approach with its greater focus on technology nevertheless retains a strong emphasis on participation. This is based on the view that changes in work organization are unlikely to be effective if they are imposed. Instead they have to be “owned” by those who are going to experience them. At the same time, they accept that if there is to be a fully open system with its variety of internal and external influences, there will be limits to the extent of employee participation. They therefore believe that their approach to sociotechnical systems design is incompatible with other approaches that give primacy of industrial democracy, as was the case in Norway, or to quality of working life, with its tendency to emphasize the social system (Davis & Cherns, 1975). This is not to claim that these alternatives are not legitimate goals, but more to argue that they are not compatible with the some of the core of features of sociotechnical systems design.
Information Systems and Ergonomics
There has been a longstanding interest in sociotechnical systems among researchers in the fields of information and computer technologies and ergonomics. This is important because much of the reported research has addressed production technologies while most employment in advanced economies is found in the service sector. Mumford (2006) pioneered an interest in the application of sociotechnical thinking to information systems and it is implicit in the work of Checkland (1981) on soft systems. Eason (2007) illustrates an application of sociotechnical systems thinking to the analysis of the U.K. National Health Service Information Technology Program.
Ergonomists, with their longstanding interest in human factors, have much to contribute to work design and Clegg (2000), adopting an ergonomic perspective, endorsed the view that there was a need to rethink sociotechnical analysis for an era of information and communication technologies and the resultant new forms of work organization. He argued that new technologies had often failed to advance productivity because they had neglected the social system. He wanted to see social scientists, including ergonomists, involved in the design of new technologies and outlined a set of principles and processes to achieve this. Subsequently, Baxter and Somerville (2011), for example, have argued the case for building sociotechnical analysis into design methods for systems engineering.
Conclusions and the Future of Sociotechnical Systems Design
Much of this article may have the appearance of an historical review. Sociotechnical systems analysis was developed by psychologists and other social scientists at the Tavistock Institute of Human Relations in London during the 1950s and stimulated extensive research by social scientists around the world over three decades. The evidence about its impact is very mixed. There are examples of successful applications but also many illustrations of the limitations imposed by wider systems constraints. These attest to the difficulties of bringing about organizational change when commitment to the change is highly conditional and often temporary. Over the years, the contribution of organizational psychologists to work redesign has continued, including some interest in autonomous work groups, while largely neglecting the technical system. The article by Van den Broeck and Parker (2017) provides a comprehensive overview of work redesign research. It has been largely left to engineers and interdisciplinary teams to pursue innovations in the organization of work that give full weight to the technical system within what might be described as a technical-social systems framework.
Looking forward, it is premature to write off the potential contribution of organizational psychologists to sociotechnical analysis and work design. Growing concern about the impact of “Industry 4” (Frey & Osborne, 2017), the era of artificial intelligence (AI), smart factories, robotics, and extensive digitization have led to calls for the incorporation of sociotechnical systems thinking into workplace design (Kopp et al., 2019). They note that in Germany, “Work 4.0” is set alongside “Industry 4.0” to offer “a broader perspective on digital transformation as an enabler for people-centred new work designs” (p. 293) based largely on sociotechnical principles. Bednar and Welch (2020) have also highlighted the risks of neglecting the social system in “Industry 4.0.” Instead they advocate the rapid introduction of “Industry 5.0,” which recognizes the need to design sociotechnical systems that optimize the interests of both the technology and those employees who have to work with it. It is notable that the European Community has initiated “EUWIN,” the European Workplace Innovation Network, with the aim of simultaneously improving both productivity and well-being in the context of “Industry 4.” The relationship between new digital technologies and worker well-being offers a fertile focus of research by work and organizational psychologists. Indeed, Parker and Grote (2020) argue that in the digital age work design matters more than ever and psychologists with their expertise in this field should be making an important contribution. They outline some research priorities including working proactively alongside technologists to influence the design of work, highlighting in particular the importance of workers autonomy and local variance control. They also suggest the need to be involved in the education of technologists to highlight the importance of work design as well as the need to ensure that workers are educated to recognize the importance of adaptability, proactivity, and the need to engage in continuous development. Organizational psychologists should become involved in intervention studies to test some of the relevant approaches to sociotechnical design of work places (e.g., Clegg, 2000; Waterson et al., 2002). Finally, Parker and Grote (2020) recognize the need to reemphasize sociotechnical systems thinking as a major analytic framework.
Developments in sociotechnical analysis are extending beyond the workplace. One of the interesting contemporary applications of sociotechnical systems thinking can be found in policy analysis applied to complex problems. These include management of transitions such as the introduction of Industry 4 technologies but also the challenge of climate change. The core argument of policy analysts such as Geels (2004, 2018) and Edmondson et al. (2019) is that sustainable transitions require analysis of both the social and the technical systems and their interaction. For Geels (2004, p. 900) sociotechnical systems are “linkages between elements necessary to fulfil societal functions” including energy, transport, housing, food production, and consumption. Like the initial sociotechnical thinking, the core point is that technological determinism and advances in technology are not enough to ensure that change brings societal benefits and therefore change requires analysis of complex, open, dynamic, multilevel systems in which change is only likely to be effective if “the social” can be effectively engaged in the process. A somewhat similar view is taken by Davis et al. (2014) who acknowledge the potential of sociotechnical systems thinking if it can move beyond its focus on traditional notions of technology to a wider focus on complex systems. They illustrate its potential with an analysis of a crowd disaster at a football match (Challenger & Clegg, 2011).
At the start of the 1980s, Emery (1982), one of the main pioneers of sociotechnical analysis was also one of the first to identify the declining interest in its application. He used the idea of Kondratieff cycles to argue that interest would wax and wane over the long term according to wider contextual circumstances such as the economic changes precipitated by the oil crisis of the 1970s. A glance at Google Scholar reveals that the concept of sociotechnical systems analysis remains alive and well. The promise and the threat of Industry 4 for societal and individual well-being suggests that there is still a major role for work and organizational psychologists to play in exploring the interplay between social and technical systems at work and in society more widely.
- Boos, D., Grote, G., & Guenter, H. (2013). A toolbox for managing organisational issues in the early stage of development of a ubiquitous computing application. Personal and Ubiquitous Computing, 17, 1261–1279.
- Cohen-Rosenthal, E. (1997). Sociotechnical systems and unions: Nicety of necessity? Human Relations, 50(5), 585–604.
- Dankbaar, B. (1997). Lean production: Denial, confirmation or extension of socio-technical systems design? Human Relations, 50, 567–583.
- Grote, G., Ryser, C., Wafler, T., Windischer, A., & Weik, S. (2000). KOMPASS: A method for complementary function allocation in automated work systems. International Journal of Human-Computer Studies, 52, 267–287.
- Herbst, P. (1976). Alternatives to hierarchies. Martinus Nijhoff.
- Parker, S., & Jorritsma, K. (2021). Good work design for all: Multiple pathways to making a difference. European Journal of Work and Organizational Psychology, 30 (3), 456–468.
- Pearson, C. (1992). Autonomous work groups: An evaluation at an industrial site. Human Relations, 45(9), 905–936.
- Van Eijnatten, F., & Van der Zwann, H. (1998). The Dutch IOR approach to organizational design: An alternative to business process reengineering? Human Relations, 51(3), 289–318.
- Bakker, A., & Demerouti, E. (2017). Job demands—resources theory: Taking stock and looking forward. Journal of Occupational Health Psychology, 22(3), 273–285.
- Baxter, G., & Somerville, I. (2011). Socio-technical systems: From design methods to systems engineering. Interacting with Computers, 23, 4–17.
- Bednar, P., & Welch, C. (2020). Socio-technical perspectives on smart working: Creating meaningful and sustainable systems. Information Systems Frontiers, 22, 281–298.
- Bennis, W., Benne, K., & Chin, R. (Eds.). (1961). The planning of change: Readings in applied behavioral sciences. Holt, Rinehart & Winston.
- Blackler, F., & Brown, C. (1980). Whatever happened to Shell’s new philosophy of management? Saxon House.
- Bolweg, J. (1976). Job design and industrial democracy. Martinus Nijhoff.
- Buchanan, D. (1997). Review of Sandberg (ed.) op cit. Human Resource Management Journal, 7(4), 89.
- Burden, D. (1975). Participative management as a basis for improved quality of jobs: The case of the micro-wax department Shell UK Ltd. In L. Davis & A. Cherns (Eds.), The quality of working life (Vol II, pp. 201–215). Free Press.
- Challenger, R., & Clegg, C. (2011). Crowd disasters: A socio-technical systems perspective. Contemporary Social Science, 6, 343–360.
- Checkland, P. (1981). Systems thinking, systems practice. Wiley.
- Cherns, A. (1976). The principles of socio-technical design. Human Relations, 29(8), 783–792.
- Cherns, A. (1987). Principles of socio-technical design revisited. Human Relations, 40(3), 153–1602.
- Clegg, C. (2000). Sociotechnical principles for system design. Applied Ergonomics, 31(5), 463–477.
- Coch, L., & French, J. (1948). Overcoming resistance to change. Human Relations, 1, 512–532.
- Cordery, J., Morrison, D., Wright, B., & Wall, T. (2010). The impact of autonomy and task uncertainty on team performance: A longitudinal field study. Journal of Organizational Behavior, 31, 240–258.
- Cummings, T., Molloy, E., & Glen, R. (1977). A methodological critique. Human Relations, 30(8), 675–708.
- Davis, L. (1977). Evolving alternative organizational designs: Their socio-technical bases. Human Relations, 30(3), 261–273.
- Davis, L., Canter, R., & Hoffman, J. (1955). Current job design criteria. Journal of Industrial Engineering, 6, 5–11.
- Davis, L., & Cherns, A. (Eds.). (1975). The quality of working life (Vol 1). The Free Press.
- Davis, M., Challenger, R., Jayewardene, D., & Clegg, C. (2014). Advancing sociotechnical systems thinking: A call for bravery. Applied Ergonomics, 45(2),171–180.
- De Sitter, U., Den Hartog, F., & Dankbaar, B. (1997). From complex organizations with simple jobs to simple organizations with complex jobs. Human Relations, 50(5), 497–534.
- Eason, K. (2007). Local sociotechnical systems development in the NHS National Programme for Information Technology. Journal of Information Technology, 22, 257–264.
- Edmondson, D., Kern, F., & Rogge, K. (2019). The co-evolution of policy mixes and socio-technical systems: Towards a conceptual framework of policy mix feedback in sustainable transitions. Research Policy, 48(10), 103555.
- Emery, F. (1982). Sociotechnical foundations for a new social order? Human Relations, 35(12), 1095–1122.
- Emery, F., & Thorsrud, E. (1969). Form and content in industrial democracy. Tavistock Publications.
- Emery, F., & Thorsrud, E. (1976). Democracy at work. Martinus Nijhoff.
- Emery, F., & Trist, E. (1965). The causal texture of organizational environments. Human Relations, 18(1), 21–32.
- Engelstad, P., & Gustavsen, B. (1993). Swedish network development for implementing national work reform. Human Relations, 46(2), 219–248.
- French, J., Israel, J., & As, D. (1960). An experiment on participation in a Norwegian factory. Human Relations, 13(1), 3–19.
- Frey, C., & Osborne, M. (2017). The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254–280.
- Geels, F. (2004). From sectoral systems of innovation to socio-technical systems: Insights about dynamics and change from sociology and institutional theory. Policy, 36(6–7), 897–920.
- Geels, F. (2018). Disruption and low carbon systems transformation: Progress and new challenges in socio-technical transitions research and the multi-level perspective. Energy Research and Social Science, 37, 224–231.
- Griffin, M., Neal, A., & Parker, S. (2007). A new model of work role performance: Positive behaviour in uncertain and interdependent contexts. Academy of Management Journal, 50(2), 327–347.
- Hill, P. (1971). Towards a new philosophy of management. Gower Press.
- Karasek, R. (1979). Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Sciences Quarterly, 24(2), 285–308.
- Kelly, J. (1978). A re-appraisal of sociotechnical systems theory. Human Relations, 31(12), 1069–1099.
- Kopp, R., Dhondt, S., Hirsch-Kreinsen, H., Kohlgruber, M., & Preenen, P. (2019). Sociotechnical perspectives on digitalisation and Industry 4.0. International Journal of Technology Transfer and Commercialisation, 16(3), 290–309.
- Lawler, E. (1977). Adaptive experiments: An approach to organizational behaviour research. Academy of Management Review, 2(4), 576–585.
- Lewin, K. (1947). Frontiers of groups dynamics. Human Relations, 1(5), 5–41.
- Lewin, K., Lippitt, R., & White, R. (1939). Patterns of aggressive behaviour in experimentally induced social climates. Journal of Social Psychology, 10, 271–299.
- Lindholm, R., & Norstedt, J. (1975). The Volvo report. Swedish Employers’ Federation.
- Lupton, T. (1963). On the shop floor. Pergamon Press.
- Miller, E. (1975). Socio-technical systems in weaving, 1953–1970: A follow-up study. Human Relations, 28(4), 349–386.
- Mumford, E. (2006). The story of socio-technical design and reflections on its successes, failures and potential. Information Systems Journal, 16, 317–342.
- Parker, S., & Grote, G. (2020). Automation, algorithms and beyond: Why work design matters more than ever in a digital world. Applied Psychology: An International Review.
- Pasmore, W. (1995). Social science transformed: The socio-technical perspective. Human Relations, 48(1), 1–21.
- Pasmore, W., Francis, C., Haldeman, J., & Shani, A. (1982). Sociotechnical systems: A North-American reflection of empirical studies of the seventies. Human Relations, 35(12), 1179–1204.
- Pava, C. (1986). Redesigning sociotechnical systems design: Concepts and methods for the 1990s. Journal of Applied Behavioral Science, 22(3), 201–221.
- Qvale, T. (1976). A Norwegian strategy for democratization of industry. Human Relations, 29(5), 453–469.
- Rice, A. K. (1953). Productivity and social organization in an Indian weaving shed: An examination of the socio-technical system of an experimental automatic loomshed. Human Relations, 6(94), 297–329.
- Rice, A. K. (1955). Productivity and organization in an Indian weaving mill: A follow-up study of the experimental reorganization of automatic weaving. Human Relations, 8(4), 399–428.
- Rice, A. K. (1963). The enterprise and its environment. Tavistock.
- Roethlisberger, F., & Dickson, W. (1939). Management and the worker. Harvard University Press.
- Roy, D. (1952). Quota restriction and gold-bricking in a machine shop. American Journal of Sociology, 57(5), 427–442.
- Roy, S. K. (1969). A re-examination of the methodology of A.K. Rice’s India textile mill work reorganisation. Indian Journal of Industrial Relations, 5, 170–191.
- Sandberg, A. (Ed.). (1995). Enriching production: Perspectives on Volvo’s Uddevalla plant as an alternative to lean production. Ashgate.
- Scarborough, H. (1995). The social engagement of social science: A Tavistock anthology. Human Relations, 48(1), 22–33.
- Taylor, F. (1947). Scientific management. Harper.
- Trist, E. (1993). Introduction to Volume 11: Origins of the concept. In E. Trist, H. Murray, & B. Trist (Eds.), The social engagement of social science: A Tavistock anthology: Volume 11: The socio-technical approach (pp. 36–60). Pennsylvania University Press.
- Trist, E. (1981). The evolution of socio-technical systems. Ontario Ministry of Labour.
- Trist, E., & Bamforth, K. (1951). Some social and psychological consequences of the longwall method of coal getting. Human Relations, 4(1), 3–38.
- Trist, E., Higgin, G., Murray, H., & Pollock, A. (1963). Organizational choice. Tavistock Publications.
- Trist, E., Murray, H., & B. Trist (Eds.) (1993). The social engagement of social science: A Tavistock anthology: Vol.11: The socio-technical perspective. University of Philadelphia Press.
- Trist, E. Susman, G., & Brown, G. (1977). An experiment in autonomous group working in an American underground mine. Human Relations, 30(3), 201–236.
- Van den Broeck, A., & Parker, S. (2017). Job and work design. In Oxford research encyclopedias: Psychology. Oxford University Press.
- Von Bertalanffy, L. (1950). An outline of general system theory. British Journal for the Philosophy of Science, 1, 134–165.
- Wall, T., & Clegg, C. (1981). A longitudinal study of group work redesign. Journal of Organizational Behavior, 2(1), 31–49.
- Wall, T., Cordery, J., & Clegg, C. (2002). Empowerment, performance and organizational uncertainty: A theoretical integration. Applied Psychology: An International Review, 51(1), 146–149.
- Wall, T., Kemp, N., Jackson, P., & Clegg, C. (1986). Outcomes of autonomous work groups: A long-term field experiment. Academy of Management Journal, 29(2), 280–304.
- Walton, R. (1982). Social choice in the development of advanced information technology. Human Relations, 35(12), 1073–1084.
- Waterson, P., Older Gray, M., & Clegg, C. (2002). A sociotechnical method for designing work systems. Human Factors, 44(3), 376–391.
- Whyte, W. (1955). Money and motivation: An analysis of incentives in industry. Harper and Bros.
- Womack, J., Jones, D., & Roos, D. (1990). The machine that changed the world. Simon & Schuster.
- Work in America. (1973). Work in America. MIT Press.