A circular and dynamic model of the process of job design.
Many companies are introducing new technologies, new business processes and new ways of working (Holman, Wall, Clegg, Sparrow, & Howard, 2003). Particular initiatives such as call centre operations, e-business working, supply-chain partnerships, enterprise resource planning systems, dispersed teams, just-in-time working and the like, involve choices (whether made explicitly or implicitly) about the design of people's jobs. We believe that job design remains as significant a practical issue as ever it has been. Yet, at the same time, one could argue that the theory underlying mainstream job design thinking and practice has become rather set in its ways, if not to say moribund.
The job characteristics model offered by Hackman and Oldham (1976) remains, 30 years later, the dominant perspective in job design theory. The core proposition is that job characteristics (or job content) influence the motivation of the job-holder, which, in turn, has an effect on his/her performance and well-being. We believe that mainstream job design theoretical thinking can be characterized in a number of ways, which are summarized below:
* Job design is usually treated as a predictor variable (and only rarely as an outcome).
* Job designs are usually treated as if they are relatively fixed, at least in the short term, and common across individuals in the same job in the same organization. Furthermore, job redesign, when undertaken, is usually seen as a formal and deliberate attempt on the part of the organization to redefine who does what.
* The underlying explanatory model is largely motivational.
* The model is largely individualistic, focusing on the job-holder, often ignoring the roles and actions of other key actors in the job design process.
* Job performance is treated as an outcome variable.
* The core model is unidirectional in its causal logic (such that X influences Y which in turn influences Z).
(For examples of reviews that draw attention to aspects of the above, see Oldham, 1996; Parker & Turner, 2002; Parker & Wall, 1998, 2001.)
This is not to deny that a substantial body of useful work has been conducted in developing Hackman and Oldham's core model. For example, progress has been made in identifying some antecedents of job design (e.g. management style; see Parker, Wall, & Cordery, 2001), new moderators (e.g. context satisfaction; Oldham, 1996), additional job characteristics (e.g. cognitive demands; Parker & Wall, 2001), further outcomes (e.g. proactive behaviours; Parker & Turner, 2002) and contingencies influencing the applicability of the theory (e.g. uncertainty; Wall, Cordery, & Clegg, 2002). However, while these studies represent substantial progress, we interpret these as further embellishments of the core model. We believe the above characterization of mainstream job design theory remains valid in the main.
While the criticisms above are primarily aimed at what may be termed the job characteristics approach to job design, we should also acknowledge the contribution of sociotechnical thinking to our understanding of job design theory (see Cherns, 1976, 1987; Trist & Bamforth, 1951). A central idea in this tradition involves the adoption of semi-autonomous work groups (Herbst, 1962). However, while the importance of peers is accepted, as is the changing role of supervision, sociotechnical theory makes little attempt to delineate the role of such individuals in the job design process, except to say that people often work in teams, with the corollary that the role of the supervisor thereby changes. This tradition has also placed less emphasis on the motivating properties of jobs (Cummings, 1978). However the thrust of our criticisms above also apply to this tradition as it concerns itself with job design. In particular, the sociotechnical tradition also treats job design as a predictor variable, job performance as an outcome and the causal flow as unidirectional; furthermore, job designs are usually seen as relatively fixed in the short term (see, e.g. Beekun, 1989; Cordery, Mueller, & Smith, 1991; Cummings, Molloy, & Glen, 1977; Pasmore, Francis, Haldeman, & Shani, 1982).
As stated above, our objective in this paper is to offer a new model of the process of job design, that attempts to synthesize ideas emerging in the literature, as well as to develop some new insights. To achieve this, we draw on and further develop, five major sets of ideas in the area, and integrate them in the form of a new model. The five ideas are summarized below and elaborated upon later when describing our model.
The first idea is that job designs are flexible and adjustable in the short term. For example, Frese, Garst, and Fay (in press) argue that some people behave proactively to change their own job designs. Wrzesniewski and Dutton (2001) propose that job-holders can adapt their job content through a process of 'job crafting'. Various authors have identified the process of role innovation, through which job-holders introduce new behaviours into pre-existing roles (Nicholson, 1984; West, 1987). Graen and Scandura (1987) stress the influence of the job-holder in actively developing a role that is satisfactory to oneself and the role senders (such as supervisors and other colleagues). Several researchers have pointed out that the job-holder's supervisor can have an influence on job content in the short term, for example through delegation (e.g. Morrison, Upton, & Cordery, 1999). Thus, job design can be a recurring local process, not necessarily requiring explicit formal redesign initiatives.
Second, the job-holder's performance and perceptions of his/her competence by other key actors (most obviously the supervisor and peers) are important factors in the job design process. These may influence the extent to which the job-holder can negotiate the opportunity (and/or find the space) to change his/her job content (Bauer & Green, 1996; Lowin & Craig, 1968). In this way, job performance can act as a predictor of job design (as well as an outcome of it). This also implicates other local actors in the process of job design (Graen & Scandura, 1987; Salancik & Pfeffer, 1978).
Third, knowledge is a key factor in understanding the impact of job design. More particularly, changes in job content that expand the role and responsibilities of job-holders allow them to develop new knowledge and skills, and these help explain why performance may improve (Leach, Wall, &Jackson, 2003; Morrison et al., 1999). In this view, motivation is not the sole variable explaining the impact of job design (Frese & Zapf, 1994; Parker & Turner, 2002; Parker & Wall, 2001).
Fourth, changes in job content have an impact on self-efficacy, which has a critical role in predicting subsequent changes in behaviour and performance (Bandura, 1982; Burr & Cordery, 2001; Parker, 1998; Spector, 1995). Furthermore, there is a dynamic interplay between self-efficacy and several other variables central to the job design process. For example, Leach et al. (2003) postulate that knowledge acquisition leads to greater self-efficacy, which in turn can encourage further acquisition.
Finally, underpinning much of the above, is the notion that the process of job design is dynamic and circular (as opposed to unidirectional). Thus, for example, job content is an outcome of these processes as well as a predictor, and the same holds for job performance. Aspects of this circularity have been recognized in parts of the job design literature. For example, Karasek and Theorell (1990) identify the possibility of long-term dynamic spirals of behaviour, one positive, the other negative. In the former, an active job setting allows successful learning which in turn promotes mastery and confidence and increased capacity to accept more challenging situations, thereby promoting more learning, and so on. James and Tetrick (1986) argue that job perceptions and job satisfaction are related reciprocally. Shea and Howell (2000) demonstrate empirically the reciprocal relations between serf-efficacy and job performance. Frese et al. (in press) argue that people can behave proactively to change their job content, and that this subsequently influences their opportunity and capacity for further proactivity. Furthermore, Leach et al. (2003) acknowledge the self-perpetuating link between knowledge and self-efficacy. As such, we are arguing for a general recognition of the circular and dynamic nature of the process of job design. In this we agree with Morgan's general axiom for organizational behaviour that we need to think in 'loops not lines' (Morgan, 1997, p. 274).
In the next section we present a new model of the process of job design that tries to integrate and further develop these ideas.
A new model of the process of job design
Our model of the job design process is presented in Figure 1. We first describe the relationships in the model, before considering the main contingencies and the nature of the time lags involved.
[FIGURE 1 OMITTED]
In outline, we propose that good job performance on the part of the job-holder is interpreted by supervisors and others as evidence of competence--this encourages them to trust the job-holder. At the same time, good job performance is interpreted by the job-holder as evidence of his/her own competence, and this encourages trust in oneself. This trust (by others and by self) enables some adjustment in the role of the job-holder. This may be the result of the job-holder's own actions (as in the job-crafting thesis). In essence, they 'back themselves' to take on more responsibility. Alternatively, or in parallel, role adjustments may be initiated by the supervisor (e.g. through delegation) or by peers (e.g. by sharing out the work differently). These role adjustments result in increases in job content, which in turn lead to increased job knowledge (as well as motivation and opportunity). Increased knowledge (and motivation) lead to improvements in performance, which again are perceived as indicators of competence, and so on. In addition, there is a dynamic interplay between job-holder self-efficacy and several other parts of the model. For example, good performance has a positive impact on self-efficacy, which in turn enhances performance.
The above summarizes the situation regarding good job performance and the virtuous circle it helps create and sustain. We propose that the model also holds for the converse situation. Thus, poor job performance is perceived by supervisors and others as evidence of low competence, as it is by the job-holder. This reduces their trust in the job-holder, and this results in role adjustments, for example in the form of tighter supervisory controls over the job-holder or perhaps in work simplification initiated by the job-holder. In turn, these result in relatively impoverished job content. This reduces motivation and the opportunities to develop new knowledge, which in turn lead to a decrease in job performance. Several of these processes reduce self-efficacy, which dynamically interacts with other variables in the model.
We now describe in more detail each of the key linkages in this model. We note here that the ordering of our description of the links implies no precedence--the model is circular and its description could begin at any point. We have chosen to start with performance (operating as a predictor variable in the first instance) to highlight the difference from mainstream thinking in this area.
Link 1: Performance [right arrow] Perceived competence [right arrow] Trust [right arrow] Role adjustment
Our initial proposition is that good performance on the part of the job-holder is interpreted as evidence of competence by the supervisor(s) and manager(s) concerned, and also by the peer group and other people with whom the job-holder interacts (such as internal or external customers and suppliers). By 'good performance' here we mean that the job-holder is fulfilling and perhaps exceeding the requirements and expectations of the role. How this is manifest will vary with the nature of the work. For example, a machinist may produce more parts than targeted and of especially high quality. This process may also include social comparisons; for example, a machinist may produce more than is the norm for the peer group and of higher quality. Thereafter, there is an attributional process whereby good performance is construed as evidence of competence (Green & Mitchell, 1979). The perception of competence means that the supervisor and others develop trust in the job-holder, such that they can delegate and/or share some role responsibilities (Bauer & Green, 1996; Lowin & Craig, 1968). Trust is a critical part of the process since delegation and work sharing usually involve handing over aspects of control to the job-holder. Trust involves taking a risk (e.g. the risk that the job will be not be undertaken in a satisfactory manner). Trust is particularly important under conditions involving uncertainty, interdependence and consequentiality (Ring, 1995; Zucker, 1986). Thus, for example, delegation involves high risks when the tasks are difficult to complete and critical to the overall performance of the work system. So far as we are aware, the inclusion of trust in a model of the process of job design is a novel contribution.
We also propose that an equivalent process operates for the job-holder him/herself. Thus, good performance is interpreted as evidence of one's own competence, and this in turn leads to the development of trust in oneself. This means that the job-holder is prepared to take the risk of having/taking more responsibility and control.
Trust of the job-holder (by others and/or self) creates the circumstances within which role adjustments may be undertaken. These may be initiated by the job-holder (as in the case of job crafting) and/or by the supervisor (as in the case of delegation) and/or by peers (for example, by sharing out the work differently). So far as job-holders are concerned, their good performance, the recognition of their competence and the fact that they are trusted, increases the scope they have to enlarge their own roles (see Wrzesniewski & Dutton, 2001). For example, their supervisors are more likely to turn a blind eye to self-initiated role adjustments, or indeed to encourage such changes. Peers are more likely to believe job-holders have earned the right to undertake their own adjustments. In such situations, job-holders are likely to take on new responsibilities and to take more control over their job (as proposed by Wrzesniewski & Dutton, 2001). This helps offer an explanation for the job-crafting argument. Thus, based on our model, personal job crafting is more likely to occur when the individual is performing well, is perceived as competent (by self and others) and is trusted (by self and others).
As far as the supervisor is concerned, the same holds. Supervisors are more likely to be active in role adjustments by delegating responsibilities when the job-holder is performing well, has displayed competence in an area (Bauer & Green, 1996; Lowin & Craig, 1968) and can be trusted (Schriesheim, Neider, & Scandura, 1998). We propose that this may be for several different reasons. For example, role adjustments in the form of delegation may be undertaken:
* to promote job-holder career and personal development (Bass, 1990; Morrison et al., 1999; Schriesheim et al., 1998);
* as a way of saving the supervisor's own resources, especially time and effort (Bass, 1990; Leana, 1987);
* as a reward for good performance (Graen & Scandura, 1987; Schriesheim et al., 1998);
* in response to short-term changes in work demands.
Some confirmation of the above is provided by Leana (1987), who found that delegation was positively related to supervisory ratings of subordinate job capability and trustworthiness, supervisor workload and subordinate job performance. These findings are entirely consistent with our model. Furthermore, Schreisheim et al. (1998) predict that 'delegation is more likely to occur when managers see subordinates as competent relative to task demands and as sufficiently trustworthy to allow the managers to be confident undertaking the risks associated with delegation' (p. 300).
Similar arguments hold for the job-holder's peers, and more generally for other people with whom the job-holder interacts (such as internal or external clients and customers). Thus, we propose that colleagues are more likely to approve of, and encourage, role adjustments when the job-holder has been performing well, is perceived as competent, and is trusted. Again, such changes may be interpreted as fair reward (based on notions of equity, see Adams, 1965; Greenberg, 1982) and as a rational way of getting the job done effectively. Furthermore, peers are more likely to be influential in such processes to the extent that the work is organized in teams, and to the extent that the team has some control over work allocation (as is the case in semi-autonomous group working).
We propose that the converse arguments pertain for poor performance. Thus, poor performance is construed by supervisors, peers and the job-holder as evidence of low competence, and this reduces the trust placed in the job-holder. This may well lead to a tightening of job control (in particular on the part of the supervisor), such that the jobholder is given less scope. Similarly, the job-holder may adjust the role by simplifying it. This is more likely to happen when the work is demanding and where supervisors and peers feel stretched, in such contexts a lack of trust is especially important.
In our view these processes may be initiated and led by the supervisor (as is implied in the leadership literature) and/or by the job-holder (as is proposed by Wrzesnieski & Dutton, 2001, in their job-crafting thesis). Our argument is that either or both can happen, and we discuss later the contingencies that may affect this. For now we propose that there will be, at the very least, some circumstances where both parties are involved, and our models in this area should try to integrate these ideas. To take the case where the supervisor initiates role adjustments, then clearly, the job-holder must accept the changes in the sense that they take on new responsibilities. In the case of changes initiated by the job-holder, we accept that there may be some circumstances where adjustments are made without the knowledge and approval of the supervisor or peers (as proposed by Wrzesnieski & Dutton, 2001). However we also believe there will be circumstances where the supervisor, at the very least, tacitly approves of the changes being made. As such we propose that role adjustments (including job crafting, delegation and work sharing) are, at least in some circumstances, social endeavours involving negotiation. This is a key departure from the job-crafting thesis. This is not to argue that jobholders are either completely passive (in that they are given their job designs by the supervisors) or that they are solely individually active (in that they craft out whatever job designs they wish). Our position is that they may be more or less active depending on the situation, and we return to this issue later.
We note here that this implies (at least in some circumstances) some form of negotiation between the various stakeholders, whether this process be formal or informal, overt or covert (see also Graen & Scandura, 1987; Miller, Johnson, Hart, & Peterson, 1999). This may on occasion be straightforward, but it also opens up the possibility that role adjustments may carry with them political and emotional issues. For example, we have worked in factories where the decision over who sets the pace of work held massive symbolic importance and was keenly contested between supervisors and the shopfloor employees, who each sought control over what they regarded as a (or even the) key work issue. As Wrzesniewski and Dutton point out, key work tasks are implicated in work meanings and identities. In addition, they may have an impact on payment and on perceptions of the effort-reward bargain. This helps us understand their political and emotional importance, and why, on occasion, such role adjustments become significant social negotiations.
We stress that there is no intention here to suggest that these processes occur mechanistically. Thus, we are not claiming that high job performance is always construed as an indicator of competence, which in turn always leads to the development of trust and subsequent role adjustments. Rather, we see these as more likely to happen--the argument is probabilistic.
Link 2: Role adjustment [right arrow] Job content [right arrow] Knowledge [right arrow] Performance
This part of the model matches most closely existing mainstream job design thinking and some of the emerging ideas referred to earlier. Our proposition is that role adjustments of the kind described above have a direct impact on the job-holder's job content. Thus, acts by the job-holder to initiate expansions to their role (Frese et al., in press; Wrzesniewski & Dutton, 2001), delegations of responsibility by supervisors (Morrison et al., 1999) and work allocations made by peers, are all predicted to change job content for the job-holder. This reflects the close interdependence between these actors in the work process (Graen & Scandura, 1987), as well as the flexibility inherent in many jobs (Ilgen & Hollenbeck, 1991). In the case of role adjustments involving expansions to the job holder's role, this will usually lead to changes in job content (or job characteristics), for example involving increases in job-holder autonomy, variety and cognitive demands.
Such changes in job content influence the opportunities for the job-holder to develop and acquire new knowledge (see e.g. Parker, Wall, & Jackson, 1997). In part, this is an inevitable aspect of taking on new responsibilities--to meet the new role requirements, the job-holder has to learn new things and extend his/her capabilities (Karasek & Theorell, 1990; Leach et al., 2003). This is a central tenet of action theory, in which increases in job control are held to provide the opportunity for job-holders to develop and select good and flexible strategies for dealing with work problems (Frese & Zapf, 1994). Control over the job promotes the development of new knowledge, especially when the job is challenging. Our next proposition is that such new knowledge leads to improvements in job performance. Thus, over time, the acquisition and application of new knowledge is reflected in improved job performance.
An example illustrates the argument (see Leach et al., 2003). In a manufacturing setting, maintenance tasks that previously had been undertaken by engineers and supervisors were taken on by machine operators. In time, this meant that they acquired new knowledge about the way the machines behaved, and the operators were able, as a result, to anticipate and prevent machine breakdowns and problems. Over a period of time, role adjustments (in this case in the direction of increased responsibilities) led to changes in job content (more autonomy), the acquisition and application of new knowledge, and subsequent improvements in performance (fewer breakdowns). (See also Wall, Corbett, Martin, Clegg, & Jackson, 1990.)
We note here that this is not to deny the role of motivation in this process. Thus, we agree with others (e.g. Parker & Turner, 2002), that such changes in job content are likely to enhance knowledge and motivation, as well as the opportunity for job-holders to apply both. We have stressed the knowledge component here because it is relatively under-recognized in the job design literature more generally.
We propose that the opposite case also holds. Thus, role adjustments where jobholders are given or take on reduced roles (e.g. as a result of being subjected to tighter controls in their work), lead to more impoverished job content, and this, in turn, reduces the opportunities to develop and apply new knowledge. At the same time, this is likely to reduce job motivation and the opportunities the job-holder has to apply their knowledge and motivation. Subsequently, this leads to reduced job performance.
Link 3: Role of self-efficacy
Self-efficacy has a critical role at various stages in our model. In particular, we propose that good job performance has an impact on self-efficacy, which in turn, improves performance. The better the job-holder performs, the more confidence she/he develops, the better she/he performs (see also Bandura, 1986; Lindsley, Brass, & Thomas, 1995; Shea & Howell, 2000). Furthermore, self-efficacy is reinforced when others, most obviously supervisors and peers, perceive the job-holder to be competent, place trust in them and then demonstrate these perceptions behaviourally, for example, by delegating responsibilities (i.e. by role adjustment). Such role adjustments have a symbolic value for the job-holder, serving to increase their confidence. Furthermore, there is empirical evidence that increases in the job-holder's job content promote self-efficacy (Burr & Cordery, 2001; Frese et al., 2003; Parker, 1998). In turn, the development of self-efficacy on the part of the job-holder is likely to be perceived as further evidence of competence by supervisors and peers. In addition, we agree with Leach et al. (2003) that self-efficacy gives job-holders the confidence to seek out, acquire and apply new knowledge, and further, that this in turn generates further confidence.
Similar arguments hold for poor performers. Thus poor performance reduces self-efficacy, which further decreases performance. This cycle is reinforced by perceptions of low competence, low trust and by role adjustments which tighten controls and reduce job content, again leading to lower self-efficacy. Lower self-efficacy leads to decreased acquisition and application of knowledge and lower motivation, which serve to further reduce self-efficacy.
As with any other model, we recognize that its operation is contingent upon prevailing circumstances. Here, we consider contingencies of two main kinds by addressing two separate questions: 'Under what circumstances does the model prevail?' 'And, under what circumstances are the different social actors likely to be more or less important an influence in the process of job design, and in particular, role adjustment?' We consider each in turn.
Under what circumstances does the model prevail?
Here, we identify two sets of issues, focusing on variability in role adjustment and performance. We propose that there are likely to be some circumstances where jobs are relatively fixed, and as such, have less scope for role adjustment. Thus, we need to identify the circumstances under which jobs are relatively fixed 'givens' (perhaps when working on assembly lines, in carefully scripted and monitored call centres or in highly bureaucratized organizations) and those where there is more scope for adjustment and manoeuvre, as implied in our model. We need to understand better the nature of the constraints (if any) on the opportunities for, and practice of, role adjustment. Such constraints can themselves be construed as organizational choices to organize work in certain ways and according to certain assumptions. This raises the challenge, inter alia, of trying to understand better the relationship between job design and technology. Our prediction, based largely on our experience of working in organizations, is that while there are constraints (most obviously in the form of technologies and organizational systems), there are always choices regarding job designs. We would predict there is more scope for role adjustment than is often perceived. For example, even in the case of assembly lines, often viewed as one of the most constrained of work systems, there are choices over cycle times, work flow speeds, the numbers and types of tasks making up a job, whether operators work in fixed locations or move with the line, the organization of breaks, the ordering and organization of materials, who undertakes quality control, and so on. In this regard we agree with Wrzesniewski and Dutton (2001) who also argue in favour of flexibility (see also Ilgen and Hollenbeck, 1991). However, we do acknowledge that this is a matter of degree.
At the same time, we propose that there are also circumstances where role adjustment becomes a particularly attractive option for supervisors and peers, and indeed, for job-holders. For example, role adjustment initiated by any of the parties is, we propose, more likely to take place under circumstances where the work is demanding (e.g. when people are subjected to difficult performance targets that must be met, and where the supervisors and peers feel stretched). In these situations, the process of role adjustment becomes a rational approach to getting the work done more effectively. Of course, we note these are the very circumstances where risks are involved, and thereby, where trust is especially important (Ring, 1995; Zucker, 1986). Put another way, role adjustment is more likely to occur under conditions where trust is especially important--i.e. under conditions of uncertainty, interdependence and consequentiality (Ring, 1995; Zucker, 1986). When trust is present, role adjustments are in the direction of enlarged roles, when absent, in the direction of simplified ones.
Thus we predict that:
* Even in the most closely delineated work systems (either technically or bureaucratically), there will often be scope for role adjustments.
* People with jobs already high on autonomy have more opportunity for role adjustment through personal crafting (Nicholson, 1984; West, 1987). In general, for example, people in professional roles have more scope for manoeuvre (Mintzberg, 1979).
* People in jobs that are harder to measure (e.g. in what historically were often described as 'indirect' as opposed to 'direct' labour) are more likely to have the opportunity for personal job crafting. The cleaners described by Wrzesnieski and Dutton (2001) are a case in point. Historically, organizations have found it easier to apply Tayloristic principles of measurement and control to direct labour processes (such as shopfloor production work in a factory) than they have to indirect work (as entailed in support functions such as maintenance engineering, quality control, production control and cleaning). In part, this is because jobs that are easier to measure are also easier to micro-manage. (This is not to argue that such jobs should be micro-managed, rather that jobs that are easy to measure often are in practice).
* Similarly, people in jobs that are dispersed and less visible are more likely to have the opportunity for personal job crafting--e.g. people working away from close supervision.
* Role adjustments initiated by supervisors, peers or job-holders are more probable when the work is demanding and where they feel stretched. Under these circumstances, role adjustments may represent a rational way of getting the job done more effectively.
* Trust is especially important under conditions of uncertainty, interdependence and consequentiality. When trust is present under such conditions, the roles are likely to become enlarged, when absent, they are more likely to be simplified.
We now turn to constraints acting upon the variability of performance. If performance is Invariant, then so will perceptions of competence and trust, and our model is likely to be less useful and predictive. For example, consider a job involving machine minding where the opportunities for performing well (or indeed badly) are in large part beyond the control of the job incumbent. If there is little or no variability in performance, then there is less opportunity for variability in subsequent competence and trust, and so on.
The arguments above have focused on constraints. They do however also carry with them the corollary that the model is likely to be especially useful when there are genuine opportunities and motivations for role adjustment, and where performance can vary. Again, we predict that this is likely to be the case in most situations.
Under what circumstances are the different social actors likely to be more or less important an influence in the process of job design?
We also need to understand better the circumstances under which the different actors have critical roles in the job design process, a gap in our understanding acknowledged by Wrzesniewski and Dutton (2001). Thus, under what circumstances does the jobholder exert control over role adjustment? When are the supervisors and peers critical? What are the variables that affect these distinctions? It is early days for theorizing in this area but nevertheless we speculate that:
* Peers are especially important in the job design process when the work is organized in self-managing teams. This results from their local control over the work allocation process.
* Peers are important in the process when the work is dispersed. For example, people undertaking repairs in the field (in businesses or in customers' homes) may help one another solve the problems they face, with the potential thereby to craft new roles. For example, Orr (1996) describes the informal network created by photocopier repairers working in the field who meet to swap stories and vignettes as a means of developing and sharing knowledge and expertise. They spontaneously crafted a form of work organization that helped meet their needs, and in the process made them more effective repairers.
* Supervisors have a significant role when work is organized on strong hierarchical lines.
* Work systems that are operating effectively are more likely to afford employees the opportunity to exert control over their own job designs (itself promoting a virtuous circle), while ineffective systems are more likely to be controlled hierarchically (thereby potentially promoting a vicious circle).
* Newcomers in jobs are likely to have less scope for role adjustment, in part because they will have had less time to develop and demonstrate competence or trust. We note, of course, that not all newcomers in jobs will respond in the same ways. For example, Ashford and Black (1996) found empirical support for the argument that newcomers with high desires for control are more likely to try to negotiate role changes in the first 6 months of their job than are fellow newcomers with lower needs for control.
* Customers (internal to the business or external) are likely to have an impact on role adjustment when the job incumbent has a strong customer-facing role. For example, when working in a supply chain with a powerful customer, the job-holder may come under pressure to take on roles and responsibilities to help solve the customer's problems, such as making sure some parts are delivered on time, even when this is not within their official company role. Similarly, financial advisors or people working in the holiday and travel business, who interact directly with customers, often by telephone or computer, may get 'drawn into' new roles as demanded by their customers in order to provide a better service. We recognize that some people in such roles are unable to respond in this way (usually for organizational reasons associated with their roles and the knowledge/information they may hold), but some organizations do allow such flexibility. Understanding the potential impact of customers on the job designs of customer-facing employees opens up further lines of enquiry in this field.
* Finally here, as above, personal job crafting is more likely to take place when individuals already have high autonomy, when the jobs are less closely measured and when the work is less visible.
Our model is circular and dynamic. We feel this reflects the nature of organizational life and relationships. As with some other organizational theorists, we believe we should be trying to address time lags in a more thoughtful way than usually pertains in our theorizing. For example, Wall and Martin (1987) argued that
To our knowledge no established framework of work design attempts to specify even the order in which predicted effects occur, let alone the time it takes for them to be observable, or their duration.... The inclusion of such predictions would make theory much more testable, and where empirically supported, would have immense practical value. (p. 78)
We agree and, to the best of our knowledge, there has been no change in this state of affairs with respect to theories of job design.
In our view, the specification of models of the kind elaborated in this paper allows us to begin to predict both the order in which changes occur and the nature of lags in the system. The model itself comprises a sequence of events. In our view, at least initially, we should try to specify what lags are likely to pertain in particular circumstances. Thus, in any particular instance, we should use the model to ask how long would it take:
* to demonstrate competence to self and others;
* for such competence to result in trust;
* for job-holders and other key actors to undertake some significant role adjustment;
* for such adjustments to result in changed job content;
* for the individual to develop new knowledge;
* for such new knowledge to result in improved performance;
* for improved performance to be interpreted as evidence of competence;
* for performance, role adjustment, job content and new knowledge to enhance self-efficacy.
We acknowledge that our understanding of such issues is limited. Furthermore, we believe the answers to such questions are likely to vary with the situation under investigation. Thus, while we predict the causal ordering according to our model, the timings in effects are likely to vary across situations.
However, an example serves to illustrate the ways in which some more specific ideas on lags may be generated. From the list above, consider the links between knowledge, performance, perceived competence, self-efficacy, trust and role adjustment using the case of a skilled machinist working on batches of complex engineering products in a factory. We predict it would take several months for a trained machinist to develop the particular knowledge (of materials, machining processes, tools, computer programmes and the like) necessary to perform well on a consistent basis. Over several months of consistently good performance the machinist will come to be regarded as highly competent by supervisors, peers and him/herself. As a result the machinist becomes trusted, again, a process which takes time. With greater trust, comes the opportunity for role adjustments. The supervisor may delegate more responsibilities, for example giving the machinist more complex batches of product to machine or perhaps new work for a valued customer. The machinist may take on more quality control duties. Clearly, there will be individual differences but the essential point is that, depending on the complexity of the products and manufacturing processes involved, this sequence is likely to take months (rather than hours or days).
Consider now the circumstance where this machinist performs less well (perhaps making a machining error leading to the scrapping of an expensive part). As a result there may be some decrements in perceived competence, but these are likely to be short term, localized and quickly repaired. There is an underlying stability which is likely to be resilient, partly as a result of the long-standing nature of performance and competence. (In part this is captured by the sporting aphorism that 'form is temporary class is permanent'). The opportunities for role adjustment are unlikely to be affected.
Contrast this scenario with that for a less skilled machinist. In this case, good performance is likely to lead to an increase in perceived competence in the short term, with enhanced levels of trust by supervisors, peers and selves. This will allow role adjustments, but to a limited degree. But the same responsive logic holds true for poor performance. Thus, poor performance results in reduced perceptions of competence and trust, along with reduced opportunities for role adjustments. In this case, the system is not stable, there is more variability and the lags in change in the variables are likely to be much shorter. The system is less resilient.
We propose that we need to develop our understanding of such time lags on a case-by-case bottom-up basis, enabled by an understanding of the particular circumstances pertaining, and these would include, inter alia, managerial attitudes. This approach has the benefit of forcing the theoretician to specify and try to answer difficult questions, to make the assumptions explicit, and in this way, to begin to address the gaps in our literature regarding the order of events and the nature of time lags. This seems to us to have real potential. However, it also raises the important question of how we empirically investigate such processes, and we return to this issue in our discussion.
In addition, we speculate that there may be differences in time lags dependent on whether the cycle is virtuous or vicious. Thus, we predict that virtuous cycles of activity, as described above, are likely to operate in the medium term, i.e. to take several weeks or months. Our experience is that it takes time for people to develop competence, to become trusted, to adjust roles. While there will be exceptions and, indeed, individual differences, we predict that virtuous circles of this kind are slow moving. On the other hand, we find it plausible to argue that vicious circles operate more quickly. Again, while there will be exceptions and individual differences, we would argue that losing confidence, being seen as relatively incompetent and losing trust (or not earning it) can happen quite quickly. Negative experiences may well have more immediacy than positive ones. Furthermore, the risks of failure may be too high when trust is doubted. The process here may take days and weeks rather than months. We would also predict that new knowledge may be slow to develop as roles and job content expand, and that old knowledge is also slow to deteriorate as roles and job content reduce.
It is also possible to use our model to make some other predictions regarding changes over time. Thus, for example, we predict that:
* Following the introduction of new technologies and/or working practices, jobholders' job content will change over time as they improve their performance, develop and demonstrate their competence, and earn the trust of their peers and supervisors.
* This process will happen differentially, such that those who perform better more quickly, will adjust their roles sooner than those who perform less well.
* Newcomers in a role are subject to the same processes--thus they will have reduced job content in comparison with existing (competent) employees, but this difference will reduce over time as they perform better, establish their competence and earn the trust of their colleagues.
Overall then, we believe that circular and dynamic models of this kind open up novel opportunities for thinking about the nature of time lags in the process of job design.
Summary and discussion
Our model is an attempt to further our theorizing about job design, and to extend and challenge existing dominant ways of thinking. In Table 1 we summarize what we believe are the key differences between our perspective and the dominant view in mainstream thinking.
We make five main claims regarding our model. First, we wish to extend the range of variables implicated in the process of job design, laying particular emphasis on the role of performance, competence, trust, knowledge and self-efficacy. While the relevance of most of these variables is recognized within the job design literature, we believe their role is underdeveloped in current theorizing, in particular, the role of trust. Second, we argue for the dynamic circularity ('loopiness') of the relationships between these variables. This circular specification serves to synthesize and develop some existing ideas in what we believe is a novel and more comprehensive format. Third, we include in our model some consideration of the role of the various local actors in the job design process, most obviously the job-holder, the supervisor and peers. We believe these are all potentially active agents in the process of job design, and their roles need inclusion in our thinking. We make a range of predictions regarding the circumstances under which the various actors are more or less important in the process of job design. This extends the job-crafting thesis which hitherto has limited its consideration to the activities of the job-holder. Fourth, we have tried to identify some of the contingencies affecting our model, particularly regarding the scope and motivation for variations in role adjustment and performance. Finally, we have tried to specify the causal order of events in the process and speculate on the nature of the time lags in our model. In these ways we believe that our model synthesizes and extends emerging ideas in the field of job design.
We now consider the opportunities that these ideas open up for further development of the area. Turning first to research, we believe the model creates a series of opportunities for empirical testing, regarding the model itself, the contingencies we have identified, the sequencing of events and the time lags that may be involved. We have made a number of model-based predictions in this paper and these should be subject to empirical testing. In addition, we need to test empirically whether or not the processes underlying good performance are the same as those implicated in poor performance, as we have proposed in our model. Are the linkages, contingencies and time lags the same, or different? We also need to investigate how and when such processes stabilize and/or become unstable. The logic of virtuous and vicious circles (deviation amplifying loops--see Maruyama, 1982) of the kind described above, is that they continue getting better or worse, but clearly there must be limits to such processes. We need to examine such systems in context to investigate what other factors act as stabilizing forces and under what circumstances (e.g. see Lindsley et al., 1995; Shea & Howell, 2000). At the same time, we need to examine the circumstances under which stable work systems and job designs are rendered unstable and become subject to change.
Given that circular relationships are likely to be hard to test, we have been challenged regarding the advantages of such formulations. Our first response is that such relationships are the norm in organizations. Feedback loops are commonplace as people and organizations respond to emerging circumstances. In principle, if this is the nature of 'reality', our theorizing should try to reflect this state of affairs. Second, such models encourage and enable us to be more specific about the ordering of events and nature of time lags in the system under examination, something at which we are, as a profession, notoriously bad. Such models seem to us to create excellent opportunities for pushing our thinking forward. Third, such thinking enables us to make some specific predictions of the kind described above.
One of the potential disadvantages of our formulation is that it may be difficult to test empirically. For this we envisage four complementary approaches. First and most obviously, we need to gather longitudinal data over theoretically pre-specified time periods with which we can test the model. This is common in economics, for example, where measures over time are used to try to identify how long it takes for investments of various kinds to have an impact on subsequent profitability and productivity. Second, we can use the model to generate particular hypotheses, which if confirmed would at least be consistent with the model. For example, we have used the model to predict that newcomers will have reduced job content in comparison with existing (competent) employees, but that these differences will reduce over time as they perform better, establish their competences and earn the trust of their colleagues. As with a number of other predictions above, this is specific and testable. Supporting evidence would not prove our model is 'correct' but it would be confirmatory in part. Third, we can heed the advice of Wrzesnieski and Dutton (2001), who noted that their instances of job crafting all arose through qualitative and interpretive studies. We believe that interpretive case studies of job design in action provide a rich and complementary source of theoretical development and empirical test. Fourth, in the longer term, we envisage the use of software-based intelligent agents to simulate people's behaviour according to our model (and other rival explanations) in which we specify the nature of the relationships, the time lags involved and generate simulated data. We can then examine the behaviour of the model to see if it replicates real-life situations. We believe there is increasing scope for work of this kind (Bonabeau, 2002; Jennings, 2000).
We now consider the implications of this model for practice. Our view is that a more circular and dynamic view of job design may help practitioners in five major ways. First, it opens up the complexity of the area by challenging what still, in practice, appear to be widespread assumptions about organizing work using Tayloristic principles emphasizing command and control. As is clear in this model, one danger of such approaches lies in their self-fulfilling tendencies. Second, the model points to the range of interconnected variables that provide action levers for trying to improve performance at work. This perspective reinforces the notion of equifinality (Katz & Khan, 1966), whereby interventions can be undertaken at different parts of a system in order to effect change. Third, this model highlights the possibility of counter-intuitive managerial actions, for example whereby poor performance may be managed by reductions in supervisory control and increases in delegation. Fourth, the model allows us to attempt predictions about the likely sequence and timing of effects. Thus, we might use the model in a particular circumstance to predict that changes in role adjustments might take 6 months to begin to have a noticeable effect on performance. Such predictions could be of significant practical benefit, for example in persuading senior managers in a company to allow time for a change initiative to bear fruit. It may also identify the time at which training (i.e. the presentation of new knowledge) might have the maximum benefit.
Finally, these ideas open up the prospect of different approaches to the practice of job design. For example, one practical implication is that we try to design jobs around what people are good at. Thus, rather than design a job and fit people into it, the implication is that organizations select good and talented people, and allow and encourage them to design (or craft) their work to make the most of their talents. In practice, this means that we do not try to turn people into something they find it hard to be, but fully develop and extend what they are (and may become) good at. The managerial task thus becomes one of identifying, selecting, coordinating and developing the mix of talents they need to meet the full range of work and project requirements. Roles can also be adjusted in order to develop new competences, i.e. as a form of personal and career development. This approach is not uncommon in professional organizations, but there is no reason, in principle, why it should not be extended to other groups of employees. For example, Jenkins, Pasternak, and West (2005) describe working practices among Formula 1 motor racing teams who select highly competent people and let them get on with it. Thereafter the structure and relationships are emergent. The deferral of organizational design until practice establishes itself opens up new opportunities for work design. Our argument is that models of this kind create new ways of thinking about job design and the organization of work.
The authors would like to acknowledge the advice and help of a number of colleagues, in particular, Toby Wall, Des Leach, Nick Turner, Amy Wrzesniewski, Stephen Wood and Kate Hollis, as well as the critical comments of three anonymous referees and the journal editor.
Received 21 October 2005; revised version received 8 March 2006
Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology, (vol. 2, pp. 267-299). New York: Academic Press.
Ashford, S. J., & Black, J. S. (1996). Proactivity during organizational entry: The role of desire for control. Journal of Applied Psychology, 8, 199-214.
Bandura, A. (1982). Serf-efficacy mechanisms in human agency. American Psychologist, 37, 122-147.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bass, B. M. (1990). Bass & Stogdill's handbook of leadership. New York: The Free Press.
Bauer, T. N., & Green, S. G. (1996). Development of leader-member exchange: A longitudinal test. Academy of Management Journal, 39, 1538-1567.
Beekun, R. I. (1989). Assessing the effectiveness of sociotechnical interventions: Antidote or fad? Human Relations, 42, 877-897.
Bonabeau, E. (2002, March). Predicting the unpredictable. Harvard Business Review.
Burr, R., & Cordery, J. L. (2001). Self-management efficacy as a mediator of the relation between job design and employee motivation. Human Performance, 14(1), 27-44.
Cherns, A. B. (1976). The principles of sociotechnical design. Human Relations, 29, 783-792.
Cherns, A. B. (1987). Principles of sociotechncial design revisited. Human Relations, 40, 153-162.
Cordery, J. L., Mueller, W. S., & Smith, L. (1991). Attitudinal and behavioural effects of autonomous group working: A longitudinal filed study. Academy of Management Journal, 34, 464-476.
Cummings, T. G. (1978). Self-regulating work groups: A socio-technical synthesis. Academy of Management Review, 3, 625-634.
Cummings, T. G., Molloy, E. S., & Glen, R. (1977). A methodological critique of fifty eight selected work experiments. Human Relations, 30, 675-708.
Frese, M., Garst, H., & Fay, D. (in press). Making things happen: Reciprocal relationships between work characteristics and personal initiative in a four-wave longitudinal structural equation. Journal of Applied Psychology.
Frese, M., & Zapf, D. (1994). Action as the core of work psychology: A German approach. In H. C. Triandis, M. D. Dunnette, & J. M. Hough (Eds.), Handbook of industrial and organizational psychology (Vol. 4, pp. 271-340). Palo Alto, CA: Consulting Psychologists Press.
Graen, G. B., & Scandura, T. A. (1987). Towards a psychology of dyadic organizing. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 9, pp. 175-208). Greenwich, CT: JAI Press.
Green, S., & Mitchell, T. R. (1979). Attributional processes of leaders in leader-member interactions. Organizational Behavior and Human Performance, 23, 429-458.
Greenberg, J. (1982). Approaching equity and avoiding inequity in groups and organizations. In J. Greenberg & R. L. Cohen (Eds.), Equity and justice in social behavior (pp. 389-435). New York: Academic Press.
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16, 250-279.
Herbst, P. G. (1962). Autonomous group functioning. London: Tavistock Publications.
Holman, D., Wall, T. D., Clegg, C. W, Sparrow, P., & Howard, A. (Eds.), (2003). The New Workplace." A guide to the human impact of modern working practices. London: Wiley.
Ilgen, D. R., & Hollenbeck, J. R. (1991). The structure of work: Job design and roles. In M. D. Dunnette & L. Hough (Eds.), Handbook of industrial and organizational psychology (2nd ed., pp. 165-207). Palo Alto, CA: Consulting Psychologists Press.
James, L. R., & Tetrick, L. E. (1986). Confirmatory analytic tests of three causal models relating job perceptions to job satisfaction. Journal of Applied Psychology, 71, 77-82.
Jenkins, M., Pasternak, K., & West, R. (2005). Performance at the limit: Business lessons from Formula 1 Motor racing. Cambridge: Cambridge University Press.
Jennings, N. R. (2000). On agent-based software engineering. Artificial Intelligence, 117, 277-296.
Karasek, R., & Theorell, T (1990). Healthy work: Stress, productivity, and the reconstruction of working life. New York: Basic Books.
Katz, D., & Kahn, R. L. (1966). The social psychology of organizations. New York: Wiley.
Leach, D.J., Wall, T. D., & Jackson, P. R. (2003). The effect of empowerment on job knowledge: An empirical test involving operators of complex technology. Journal of Occupational and Organizational Psychology, 76, 27-52.
Leana, C. R. (1987). Power relinquishment versus power sharing: Theoretical clarification and empirical comparison of delegation and participation. Journal of Applied Psychology, 72, 228-233.
Lindsley, D. H., Brass, D. J., & Thomas, J. B. (1995). Efficacy-performance spirals: A multi-level perspective. Academy of Management Review, 20, 645-678.
Lowin, A., & Craig, J. R. (1968). The influence of level of performance on managerial style: An experimental object-lesson in the ambiguity of correlational data. Organizational Behavior and Human Performance, 3, 440-458.
Maruyama, M. (1982). Mindscapes, management, business policy and public policy. Academy of Management Review, 7, 612-619.
Miller, V. D., Johnson, J. R., Hart, Z., & Peterson, D. L. (1999). A test of antecedents and outcomes of employee role negotiation ability. Journal of Applied Communications Research, 27, 24-48.
Mintzberg, H. (1979). The structuring of organizations. Englewood Cliffs, NJ: Prentice Hall.
Morgan, G. (1997). Images of organization. London: Sage.
Morrison, D. L., Upton, D. M., & Cordery, J. L. (1999). The role of supervisor behavior in facilitating opportunities for skill development and utilization. Human Factors and Ergonomics in Manufacturing, 9(1), 49-67.
Nicholson, N. (1984). A theory of work role transitions. Administrative Science Quarterly, 29, 172-191.
Oldham, G. R. (1996). Job design. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology (pp. 33-60). Chichester, UK: Wiley.
Orr, J. E. (1996). Talking about machines: An ethnography of a modern job. Ithaca, NJ: Cornell University Press.
Parker, S. K. (1998). Enhancing role breadth self-efficacy: The roles of job enrichment and other organizational interventions. Journal of Applied Psychology, 83(6), 835-852.
Parker, S. K., & Turner, N. (2002). Work design and individual work performance: Research findings and an agenda for future inquiry. In S. Sonnentag (Ed.), The psychological management of individual performance." A handbook in the psychology of management in organizations (pp. 69-93). Chichester, UK: Wiley.
Parker, S. K., & Wall, T. D. (1998). Job and work design: Organizing work to promote well-being and effectiveness. San Francisco, CA: Sage.
Parker, S. K., & Wall, T. D. (2001). Work design: Learning from the past and mapping a new terrain. In N. Anderson, D. S. Ones, H. K. Sinangil, & C. Wisvesvaran (Eds.), Handbook of industrial, work and organizational psychology (Vol. 1, pp. 90-109). London: Sage.
Parker, S. K., Wall, T. D., & Cordery, J. L. (2001). Future work design research and practice: Towards an elaborated model of work design. Journal of Occupational and Organizational Psychology, 74, 413-440.
Parker, S. K., Wall, T. D., & Jackson, P. R. (1997). That's not my job: Developing flexible employee work orientations. Academy of Management Journal, 40, 899-929.
Pasmore, W, Francis, C., Haldeman, J., & Shani, A. (1982). Socio-technical systems: A North American reflection on the empirical studies of the seventies. Human Relations, 35, 1179-1204.
Ring, P. S. (1995). Trust. In N. Nicholson (Ed.), Encyclopedic dictionary of organizational behavior (pp. 584-586). Oxford: Blackwell.
Salancik, G., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23, 224-253.
Schriesheim, C. A., Neider, L. L., & Scandura, T. A. (1998). Delegation and leader-member exchange: Main effects, moderators, and measurement issues. Academy of Management Journal, 41(3), 298-318.
Shea, C. M., & Howell, J. M. (2000). Efficacy-performance spirals: An empirical test. Journal of Management, 26, 791-812.
Spector, P. E. (1995). Self-efficacy. In N. Nicholson (Ed.), Encyclopedic dictionary of organizational behavior (p. 511) Oxford: Blackwell.
Trist, E. L., & Bamforth, K. W. (1951). Some social and psychological consequences of the long-wall method of coal-getting. Human Relations, 4, 3-38.
Wall, T. D., Corbett, J. M., Martin, R., Clegg, C. W., & Jackson, P. R. (1990). Advanced manufacturing technology, work design and performances: A change study. Journal of Applied Psychology, 75, 691-697.
Wall, T. D., Cordery, J. L., & Clegg, C. W (2002). Empowerment, performance and operational uncertainty: A theoretical integration. Applied Psychology." An International Review, 51(1), 146-169.
Wall, T. D., & Martin, R. (1987). Job and work design. In C. L. Cooper & I. T. Roberston (Eds.), International reviews of industrial and organizational psychology (pp. 61-91). Chichester: Wiley.
West, M. (1987). Role innovation in the world of work. British Journal of Social Psychology, 26, 305-315.
Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning employees as active crafters of their work. Academy of Management Review, 26, 179-201.
Zucker, L. G. (1986). Production of trust: Institutional sources of economic structure, 1984-1990. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior (Vol. 8, pp. 53-122). Greenwich, CT: JAI Press.
Chris Clegg * and Caroline Spencer
University of Leeds, UK
* Correspondence should be addressed to Chris Clegg, Leeds University Business School, University of Leeds, Leeds, L52 9JT, UK (e-mail: firstname.lastname@example.org).
Table 1. Comparing mainstream approaches to job design with the new model Mainstream New Job design as predictor Job design as predictor and outcome Job designs are relatively Job designs are flexible in fixed in the short term the short-term Motivational explanation Knowledge is also a key mechanism Individualistic model More social model (including focused on job holder other actors) Performance as outcome Performance as predictor and outcome Unidirectional process Circular and dynamic process Underspecified sequence Specification of causal chain of events of events No explicit consideration Opportunity to specify and of time lags investigate lags
|Printer friendly Cite/link Email Feedback|
|Author:||Clegg, Chris; Spencer, Caroline|
|Publication:||Journal of Occupational and Organizational Psychology|
|Article Type:||Author abstract|
|Date:||Jun 1, 2007|
|Previous Article:||A typology of the part-time workforce: differences on job attitudes and turnover.|
|Next Article:||Perceived similarity and complementarity as predictors of subjective person-organization fit.|