FACT OR FICTION
Master thesis, MscBA, specialization Change Management University of Groningen, Faculty of Management and OrganizationABSTRACT
Purpose. This study explores the dependency of the perceived freedom of choice of a change approach on both business complexity and socio-technical uncertainty, including the moderating effects of age and tenure. Moreover, the effect of exercising influence on the situation by the change agents to increase their freedom of choice is investigated.
Design/methodology/approach. A quasi-experimental design is used to test the hypotheses developed. Interviews are conducted to enrich the statistical findings and to study the effects of the influence exercised on the situation by change agents.
Findings. Except from confirming the association between the perceived freedom of choice of a change approach and the type of change approach selected and the moderating effect of tenure, the experiment did not lead to any significant results. Contrary to contingency logic, choices were not determined by situational variables; business complexity and socio-technical uncertainty. Analysis of the interviews identified expert power, need for change and resistance to change as antecedents for freedom of choice. Moreover, it was confirmed that change agents frequently exercise influence on the situation leading to an increase in their freedom of choice.
Research limitations/implications. This study underlines the importance of incorporating choice in change management theories. Our knowledge concerning the boundaries of the freedom of choice of a change agent is broadened. However, the results on where and when this freedom of choice occurs remain inconclusive.
Originality/findings. This research shows that the perceived freedom of choice of a change approach is not dependent on business complexity and socio-technical uncertainty. In addition, it was argued that change agents frequently influence the situation as to increase their freedom of choice. This was done with a multilateral approach consisting of both an experiment and interviews.
Key words. Freedom of choice, change approach, business complexity, socio-technical uncertainty, contingency theory.
Acknowledgments: First and foremost I want to thank Dr. C. Reezigt for his critical but constructive feedback
TABLE OF CONTENTS
INTRODUCTION ... 4
THEORY ... 4
Three perspectives on the scope of choice... 5
Choice in change management ... 8
Conceptual model ... 15 METHODS ... 15 Data collection ... 16 Measures ... 17 Data analysis ... 20 RESULTS ... 23 Experiment ... 23 Interviews ... 25
CONCLUSION AND DISCUSSION ... 30
Conclusions ... 30
Theoretical and managerial implications ... 36
Suggestions for further research and limitations ... 37
REFERENCES ... 38
APPENDIX A: THE EXPERIMENT ... 42
Case A1: high business complexity ... 42
Case A2: low socio-technical uncertainty ... 44
Case A3: high socio-technical uncertainty ... 44
Case A4: low socio-technical uncertainty ... 45
APPENDIX B: INTERVIEW QUESTIONS LIST ... 46
APPENDIX C: EXPLORATORY FACTOR ANALYSIS ... 47
APPENDIX D: SUMMARY TABLE OF THE INTERVIEW RESULTS ... 48
INTRODUCTION
The majority of the existing research on change management is either explicitly or implicitly based on contingency theory. Contingency theorists believe that, taking into account the specific situational variables, there always is an one best way for each situation to organize organizational change. Subsequently, it is argued that the only rational course of action is to accept the validity of the situation and to organize the change accordingly (Burnes, 2009). Hence, it is believed that the context or environment should determine how a (change) manager responds (Grint, 1995). However, it is up for debate if the people that lead a change initiative support this deterministic view that projects them as being virtual prisoners of organizational contingencies.
On the contrary, amongst others Burnes (2009) and Child (1972; 1997) argue that the degree of choice of, for example a change agent, is significantly higher than contingency theory suggests. In his book Burnes (2009) supports this statement by introducing the concept of equifinality coined by Child (1972). Burnes (2009: 596) defines equifinality as: “this concept states that different sorts of internal arrangements and structures can be perfectly compatible with identical contextual or environmental states”. Put simply, this means that there is more than one way for organizations to structure themselves in order to achieve their goals. So, instead of being obliged to follow the ‘one best way’ prescribed by a contingency model, a multitude of ways can be selected that all have the potency of success. Hence, the scope of choice of the decision maker increases. Moreover, Burnes (2009) also recognizes the possibility that influence is exercised on situation as to increase the freedom of choice and in order to make the situation fit to the organizational preferences.
Although Burnes (2009) embarks upon the discussion of the increased scope of choice, he does not elaborate upon the accompanying issues any further. Even more surprisingly, while there are theories on change approaches that offer room for a wider scope of choice, up to now no explicit research on this topic seems to exist. As a consequence, it still remains unanswered if the extensive use of contingency theory in, for example change management literature, is legitimized. Hence, we are still left with the relevant questions of where the boundaries of the freedom of choice of a (change) manager reside and when and where a change agent has choice and to which degree (Rond &Thiethart, 2007).
To explore the answers to these questions, in this thesis the freedom of choice of a change approach is addressed. This research does not elaborate upon the concept of equifinality any further, because in this thesis the focus does not lie on the amount of different options that can be selected in case of a certain environmental state. This study focuses on the selection of a change approach and on degree of freedom of choice that a change agent has in the selection process. Moreover, in line with Burnes (2009) his suggestion this research investigates the possibility of change agents influencing the situation to increase their freedom of choice.
To pursue these objectives, both a quasi-experimental design and interviews are conducted. However, before proceeding to the methodology, we start with a theoretical chapter in which a look is taken at what literature tells us about managerial choice and choice in change management.
THEORY
management is of central interest. The freedom of choice in the selection of change approaches, the room that theories hold for change agents to exercise influence on the situation and the potential effects of age and tenure in the choice process will be considered. Hypotheses are developed based upon the discussion of these variables. The hypotheses in turn form the basis for the conceptual model at the end of this first chapter.
Three perspectives on the scope of choice
Let us explore the concept of choice by taking a look at management theory in general. Three distinct groups of theorists will be reviewed that all have a different perspective on the freedom of choice that a manager has: hard determinists, the libertarians and the soft determinists.
Hard determinism. Rond & Thietart (2007: 542) define hard determinism, as: “subscribed to the view that all actions are causally determined and that their causes can be traced to events preceding our own existence, placing them beyond our control”. From this definition it follows that in this perspective choice is viewed as being severely limited, if not as being non-existent. Gopalkrishnan & Dugal (1998) divide the hard determinist in two groups.
Firstly, there is the group that takes the organization as level of analysis. Contingency theorists form a major part of this group. Their main argument is that the fit between environment and structure is the key to organizational effectiveness. It is argued that the context or environment should determine how a manager responds (Grint, 1995). Contingency theory is based upon three premises (Schreyögg, 1980: 308-309). First, it is assumed that there is only one best structural answer to a specific contextual situation. Second, the environment is considered as given. Third, it is believed that the organization has to achieve a certain level of economic performance to survive and that the criteria against which its performance is assessed are defined externally. A well known example of a contingency theory is the generic strategies model of Porter.
An interesting question is what this means for the conceptualization of change in this perspective. According to Astley & van de Ven (1983: 247) change viewed from the hard determinism perspective at the organizational level can be defined as: “dividing and integrating roles to adapt subsystems to changes in environment, technology, size and resource needs”.
Secondly, there is the group that views hard determinism at the industry level. The ecology school of organizational theory falls under this group. They argue that environmental resources are limited and that organizations continually need to adapt to take advantage of the changing base. It is believed that environmental factors determine successful organizational form, function and strategy and that only a selection of organizations is able to survive in the long run (Zammuto, 1988). In addition, it is argued that organizations do not have control over these environmental factors.
Because of this lack of control choice becomes severely limited. It can even be argued that the only choice left in this setting is the opportunity to select the types of environment in which a company operates. Consequently, looking at hard determinism at the industry level, change becomes: “a natural evolution of environmental variation, selection and retention” (Astley & van de Ven, 1983: 247).
Furthermore, at both levels the role of the manager is seen as being inactive, rational and reactive (Astley & van de Ven, 1983).
Finally, criticism on the hard determinism perspective can be summarized by the argument that choice is ignored. In the article of by Gopalkrishnan & Dugal (1998: 150) the following comments are listed:
Ø Leaders play a more significant role than the environment in effecting meaningful organizational outcomes. Ø Organizations are learning entities.
Ø Slack provides a means of organizational change.
Ø New organizational structures evolve to adapt to environmental changes because of managerial action.
An additional comment can be found in the article of Schreyögg (1980: 309-312)
Ø There is not only one pattern for securing ‘survival’ in a given environment (see also the explanation of the concept equifinality).
Libertarianism. Now that we have taken a closer look at the hard determinism perspective, the same will be done for the libertarian perspective, which can be defined as: “genuine freedom of choice” (Rond & Thietart, 2007: 536). It follows that in the libertarianism perspective it is believed that choice is unconstrained by contingencies. According to Gopalakrishnan & Dugal (1998) the libertarianism scholars can be divided into a group that focuses on the organizational level and a group that focuses on the level of collective organizations.
Firstly, at the organizational level it is argued that the dominant coalition within an organization make the choices. Contrary to the hard determinism perspective, it is argued that performance depends upon the management team instead of the environment. Child (1997:70) views management in this perspective as an ongoing political process with the following steps: information, evaluation, learning, choice, action, outcome and feedback of information. Moreover, it is believed that the management team has the opportunity to influence the conditions prevailing within the environment in which the organization is operating (Child, 1972).
Action theory, resource dependency theory, contemporary decision theory and strategic choice fall under the libertarian perspective viewed at the organizational level (Astley & van de Ven, 1983). An example of an accompanying model is the: “upper echelons model”. This model is based on the belief that the performance of an organization is ultimately a reflection of its top managers (Hambrick, 1987). In line with this model, resource based scholars contend that managers develop unique knowledge bases and inimitable skills that produce distinctive competencies and superior performance levels. Hence, the role of the manager in the libertarian perspective is seen as being proactive (Astley & van de Ven, 1983).
Again, an interesting question is what change means in this perspective. Astley & van de Ven (1983: 247) define change viewed from the libertarian perspective as: “environment and structure are enacted and embody the meaning of action of people in power” (Astley & van de Ven, 1983: 247).
Secondly, at the collective level it is believed that organizations interact to modify and construct their collective environment and the rules and options available within the environment through mobilization of action and resources. Furthermore, it is argued that organizations are natural systems where decision making is justified or rationalized ex-post.
Whittington (1988: 522) in his realist view describes: “humans as agents that are capable first of formulating unique independent objectives and then of pursuing both purposefully and effectively”.
Now, at the collective level, again the concept change begets a different meaning. Astley & van de Ven (1983: 247) define change here as: “collective bargaining, conflict, negotiation, and compromise through partisan mutual adjustment”.
Again, a remaining interesting issue is the view that is taken of the environment. From the libertarian viewpoint, the environment is seen in a subjective manner, as information flows that can be interpreted, made sense of and enacted (Gopalakrishnan & Dugal, 1998).
Finally, criticism on the libertarianism perspective can be found in the work of Donaldson (2005). He argues that strategic choice theory fails theoretically and empirically, because contingency strongly determines structure. As a consequence, he argues that there is no room for choice that is unconstrained by contingencies.
Soft determinism. Now that we have taken an in-depth look at the hard determinism and the libertarian perspective, let us proceed to the soft determinism perspective. Soft determinism can be defined as: “subjected to the view that freedom is the power to do what one wants, even if this ‘want’ is itself determined” (Rond & Thietart, 2007: 542). Hence, this third perspective reconciles hard determinism and libertarianism.
This reconciliation can be seen in the article of Gopalakrishnan & Dugal (1998). They argue that in reality the power, choices and abilities of managers to make a difference lies in between hard determinism on the one hand and libertarianism on the other hand. This view is supported by Burnes (2009: 430-431), who states: “we can see that organizations have considerable scope for changing or influencing the constraints they face but that, despite what the postmodernists claim, there are some constraints that appear beyond the scope of individual organizations to change.” So, although some constraints can be influenced and therefore choice is existent, other constraints cannot be influenced and therefore restrict choice.
An example of a model that is in line with the soft determinism perspective is the “strategic choice and environmental determinism framework” developed by Hrebinak and Joyce (1995). In their model a group of organizations is described that operates in one or more environmental niche(s). In these niches certain rules and immutable environmental conditions severely constrain outcomes or behaviours in some areas. However, in other areas room for choice does exist. Hence, this is an example of choice and determinism coexisting. These niches are argued to occur in highly regulated industries. Here, on the one hand product characteristics, capital requirements and legal constraints are highly regulated. But, on the other hand individual choice is high due to factors such as size, market structure, multiple means or methods of achieving desired outcomes, and low resource dependency on external sources. In literature various examples of these industries can be found. One example is the study of the airline industry by Peteraf & Reed (2008). The results of their study suggest that choice is employed in domains unconstrained by regulatory strictures to counteract the effects of constrained or pre-determined choices in other domains. They state that this is an adaptive mechanism that helps firms adjust to environmental change or manoeuvre over a rugged landscape. Studies with similar outcomes were conducted by Rodrigues & Child (2003) and Ivanova (2007).
environment. Therefore it can be concluded that both a subjective and an objective view of the environment is taken simultaneously.
Literature does not provide explicit statements on the role of the manager and the meaning of change in the soft determinism perspective. Nevertheless, because the soft determinism perspective is a combination of the hard determinism perspective and the libertarian perspective, the answers to these questions can be obtained by combining the insights from these previously discussed perspectives. Subsequently, the role of the manager comes partially reactive and partially proactive. In addition, change in this perspective would consist of partially adapting to the environment and of partially enacting the environment. Henceforth, choice viewed from the soft determinism perspective is unconstrained in some, while constrained in other areas.
Choice in change management
Up to this point three different perspectives on managerial choice have been reviewed. Now, choice in change management is addressed by looking at the freedom of choice in the selection of change approaches and by looking at the freedom of choice that results from the influence exercised on the situation by the change agent. A change approach can be defined as “an overall perspective on the phenomenon of change and how to bring it about, including implicit and explicit beliefs concerning both the content and the process of good change.” Accompanying the term change approach is the term change method. The methods used are subordinate to the change approach and can be defined as “the step-by -step models that tell us how to achieve the results set out to be achieved by the approach followed” (Werr, Stjernberg & Docherty, 1997: 288).
Firstly, theories will be reviewed that explicitly argue that that the selection of a change approach is restricted by a set of situational variables. Secondly, a look will be taken at theories in which it is argued that choice can be exercised in change management through exercising influence on the situational variables. Finally, a look is taken at the effect of age and tenure in the choice process.
Contingency approaches: the effect of situational variables. Suppose that a need for change is recognized and a manager in the organization decides to act upon it. The next relevant question is to what extent a manager has a choice in the process of conducting this change and making it a success. To find an answer to this question, in this section two different contingency theories are discussed with a focus on the question of how much freedom of choice they offer to change agents in their selection of the change approach to follow.
When reviewing these approaches it is useful to keep in mind they are rooted in the hard determinism perspective and that they fall under the modernism paradigm. Burnes (2009: 599) defines modernism as: “…a strong belief in progress, economic and scientific rationality, a search for fundamental rules and laws which govern both the natural world and human nature...”. Looking at this definition and remembering the previous discussion on the hard determinism perspective, we can expect that the degree of freedom of choice in the selection of a change approach in these contingency theories is low. Let us examine two of these approaches to find out if this suspicion is correct.
figure 1 underneath, ranging from: on the one side change that is implemented rapidly, has a clear action plan and low involvement and on the other side slow change with a
FIGURE 1: Choice continuum (Kotter & Schlesinger, 2008: 137)
The place taken on the continuum should be henceforth determine the approach that the
Ø The amount and type of resistance anticipated Ø The position of the initiator vs. the resisters
Ø The person who has the relevant data for designing the change and the energy for implementing it Ø The stakes involved.
Kotter and Schlesinger (2008)
change agents, it has to be remembered that low involvement and forcing change side effects. Hence, it is believed that a particular set of situational variables organize change. Therefore, accordin
severely constrained, if not non-existent. Secondly, we take a look at a approaches, see table 1 on the next page, these approaches has a number of
believe that the selection of an approach from one of Comparing table 1 (p.10) with the continuum
clear that the dimensions of directed change and planned change, although worded slightly different, seem to correspond with the dimensions from the continuum.
approach are additional to the theory of Kotter & Schlesinger (2008). Moreover, Buono & Kerber (2010) seem to mention different situational variables. They argue that the degree of business complexity and the degree of socio-technical uncertainty determines which of the three
the next page. Let us see what the authors
Firstly, business complexity is defined as
number of components and the extent of differentiation in the organization in which it is to be implemented” (Buono & Kerber, 2010: 8). Amongst
and the array of critical stakeholders”, are indicator
Secondly, socio-technical uncertainty is understood as: “referring to the amount and nature of information processing and decision making required for the change, based on the extent to which the tasks involved are determined, established,
indicators for socio-technical uncertainty
on the one side change that is implemented rapidly, has a clear action plan and low involvement and on the other side slow change with a less clear plan and high involvement.
FIGURE 1: Choice continuum (Kotter & Schlesinger, 2008: 137)
taken on the continuum should be dependent upon the following situational factors approach that the change agent should select (Kotter & Schlesinger, 2008: 138): The amount and type of resistance anticipated.
. the resisters.
The person who has the relevant data for designing the change and the energy for implementing it
(2008) argue that insofar these situational variables leave a choice open for it has to be remembered that low involvement and forcing change bares a high risk of negative
it is believed that a particular set of situational variables results in according to this theory a change agent’s freedom in selecting
existent.
Secondly, we take a look at a theory introduced by Buono & Kerber (2010). approaches, see table 1 on the next page, to implement organizational change. The authors
advantages and disadvantages. In addition, also Buono & Kerber (2010) of an approach from one of these three is dictated by situational variables.
with the continuum (figure 1) of Kotter & Schlesinger (2
clear that the dimensions of directed change and planned change, although worded slightly different, seem to correspond with the dimensions from the continuum. Nevertheless, the dimensions of the guided change to the theory of Kotter & Schlesinger (2008). Moreover, Buono & Kerber (2010) seem to situational variables. They argue that the degree of business complexity and the degree of technical uncertainty determines which of the three change approaches should be selected
. Let us see what the authors mean with these situational variables.
Firstly, business complexity is defined as: “referring to the intricacy of the system, in essence, the ts and the extent of differentiation in the organization in which it is to be implemented”
Amongst others: “size, geographical dispersion, the number of products and services and the array of critical stakeholders”, are indicators of business complexity.
echnical uncertainty is understood as: “referring to the amount and nature of information processing and decision making required for the change, based on the extent to which the tasks
tablished, or exactly known” (Buono & Kerber, 2010: 8).
technical uncertainty: amount of ways known to approach the situation, the degree to which on the one side change that is implemented rapidly, has a clear action plan
involvement.
FIGURE 1: Choice continuum (Kotter & Schlesinger, 2008: 137)
following situational factors, which (Kotter & Schlesinger, 2008: 138):
The person who has the relevant data for designing the change and the energy for implementing it.
argue that insofar these situational variables leave a choice open for bares a high risk of negative results in an one best way to selecting a change approach is
introduced by Buono & Kerber (2010). They list three main . The authors argue that each of also Buono & Kerber (2010) these three is dictated by situational variables.
& Schlesinger (2008), it becomes clear that the dimensions of directed change and planned change, although worded slightly different, seem to , the dimensions of the guided change to the theory of Kotter & Schlesinger (2008). Moreover, Buono & Kerber (2010) seem to situational variables. They argue that the degree of business complexity and the degree of selected, see figure 2 on
: “referring to the intricacy of the system, in essence, the ts and the extent of differentiation in the organization in which it is to be implemented” : “size, geographical dispersion, the number of products and services
an understandable sequence of steps can be followed and the degree to which an identifiable set of established procedures and practices are available. In figure 2 underneath the authors show how they believe that these two situational variables determine the choice of the change approach.
TABLE 1: Three approaches to manage change (Buono & Kerber, 2010: 7)
FIGURE 2: Three approaches to manage change (Buono & Kerber, 2010: 9)
Now, let us take a closer look at figure 2 and ask ourselves how much choice the change agent allegedly has. The change agent can select one of the three circles, taking into account the situational variables. Hence, only if the circles are overlapping, the change agent has a choice. In the article of Bono & the model is elaborated upon. However, concerning the roles of the both business complexity and socio-technical uncertainty there are some issues that remain questionable or unclear.
Firstly, the authors argue that a higher degree of socio-technical uncertainty leads to an increased reliance of the change agent on judgment, intuition and expertise. They argue that this effect occurs because the challenge and its solutions become less clear and the appropriate solution is more difficult to identify. As a consequence, it can be argued that this results in an increase in the freedom of choice in the selection of a change approach. Hence, if change agents have to rely on their specialist judgment, intuition and expertise, it is reasonable to expect that other organizational members leave the selection of the change approach to them, because they are not capable of making the necessary consideration themselves due to a lack of the required
Directed change Planned change Guided change
Characterstics
Top-down, hierarchical Linear, "road map " Iterative spiral
Change Goals (Ends)
Tightly defined, unchanging goal Clear goal, with some modification as needed
Loosely defined direction
Change Process (Means)
Thightly constrained Flexible, particip ative Experimental, improvisation
Change Leadership (Role)
Tell, order, command Devise a p lan to accomplish the goal Point the way , guide and watch over, instruct
Changemaker Dynamics
Persuasion Influence, cooperation Collaboration
Pace of Change
Urgent, fast, "just do it" Go slow during planning to go fast during implementation
knowledge. However, such an effect is not visible in the model; as the degree of socio-technical uncertainty increases, the freedom of choice does not increase.
In addition, Buono & Kerber (2010) do not make a similar argument for business complexity; they do not argue that due to an increase of the business complexity the reliance upon the judgement and expertise of the change agent increases as well. However, it is questionable if this effect only occurs for socio-technical uncertainty. For instance, if a company is stationed in a lot of different foreign countries, the expertise required for the change agent to handle the cultural differences is likely to increase as well. Subsequently, in a similar vein it can be argued that this could lead to an increase in the freedom of choice of a change approach. As a consequence, similar dependencies of the perceived freedom of choice of a change approach on both business complexity and socio-technical uncertainty should be apparent.
Furthermore, when viewing the variables in figure 2 (p.10) separately, it appears that the guided change approach can be selected regardless of the degree of business complexity. Hence, only taking into consideration the degree of business complexity, in each situation the change agent can choose between at least two approaches; only the composition of these two approaches changes. On the contrary, when looking at socio-technical uncertainty separately, in each situation the change agent can select only one approach; the change agent has no choice. Hence, it seems that the authors believe that business complexity itself separated from socio-technical uncertainty and vice versa has a different effect upon the freedom of choice of a change approach than these variables have in combination. Considering the former argument, this would be unlikely.
Finally, referring back to the perspectives on managerial choice that were introduced in the first chapter, the model opposes to the libertarian in which it was argued that choice is unconstrained by contingencies.
All in all, the legitimacy of the model appears to be questionable. Therefore, to explore the issues raised, the effects of both business complexity and socio-technical uncertainty on the selection of a change approach will be studied. For this cause, the following hypotheses are developed:
H1: The perceived freedom of choice of a change approach is dependent upon the degree of socio-technical uncertainty of the situation.
H2: The perceived freedom of choice of a change approach is dependent upon the degree of business complexity of the situation.
Subsequently, it can be argued that there should be change theories and approaches existent in which the room for the change agent to influence the situation is incorporated. To investigate this, in the remainder of this section a closer look will be taken at both a change theory and a change approach in which this seems to be the case; continuous change theory and post-modern organizational development (P-OD). Note that we are no longer concerned with the selection of a change approach, but with the freedom of choice of a change agent in the change process in total.
However, before we proceed, let us take a final look at the article of Buono & Kerber (2010). Reviewing the change approaches from table 1 (p.10) at close, it can be argued that the degree of choice in the planned and the directed approach is lower compared to the degree of choice in the guided approach. An improvised change process, with a loosely defined direction is likely to provide more freedom of choice to the change agent than a tightly pre-defined or participative change process. Subsequently, it can be argued that in the guided change approach it is more likely that the change agent can exercise influence on the situation faced. Hence, at the level of the specific activities that accompany a change approach, the theory of Buono & Kerber (2010) also seems to differ on the amount of options that are open to exercise influence on the situation as to increase the freedom of choice of the change agent. With this in mind, let us now proceed to a change theory and a change approach in which the possibility to exercise influence in the change processes seems to be even more evident.
Firstly, we will discuss continuous change theory which can be defined (Weick & Quinn, 1999: 375) as: “a grouping of organizational changes that tend to be ongoing, evolving and cumulative”. While this definition does not help to say something about the possibility to exercise influence within the accompanying change approach, Weick and Quinn (1999: 366) also describe the accompanying change process with the following three steps:
1. Freeze: make sequences visible and show patterns through maps, schemas, and stories.
2. Rebalance: reinterpret, re-label, and re-sequence the patterns to reduce blocks. Use logic of attraction. 3. Unfreeze: resume improvisation, translation, and learning in ways that are more mindful.
In this process, Weick & Quinn (1999) describe the role of the change agent as a sense maker. In an article of Ford, Ford & D’amelio (2008: 363) sensemaking is defined as: “an active process that involves the interaction of information seeking, meaning ascription and associated responses. Including extracting particular behaviours and communications…., interpreting them to give meaning and then acting on the resulting interpretation”. Based on this description, with terms as improvisation and interpretation included, it can be argued that continuous change theory seems to offer a change agent a vast amount of room to influence the situational variables; to shape reality. Hence, in continuous change theory the change agent has a considerable higher degree of choice in terms of what to change, how to change and when to change as compared to contingency theories, in which the best way to organize the change is pre-described and in which the single existing reality has to be accepted and considered as given.
Kotter & Schlesinger (2008). Hence, both in the guided change approach and in continuous change theory it can be argued that the room for influencing the situation and the shaping of reality is apparent and that this arguably can lead to an increased degree of freedom of choice for the change agent.
Secondly, we will take a look at post-modern OD (P-OD). Originally organizational development (OD) was defined as: “an approach to change developed in the USA based on the work of Kurt Lewin and concerned with improving the effectiveness of the human side of the organization through participative change programmes” (Burnes, 2009: 599). However, in an effort to describe a new form of OD Bushe & Marshak (2007) wrote an article concerned with P-OD, which they describe by the following four characteristics (Bushe & Marshak, 2007: 5).
Firstly, P-OD change processes emphasize changing the normal, everyday conversations that take place in the system. Secondly, P-OD may or may not have a “data collection” phase, but when it does, the purpose is to surface, legitimatize, and/or learn from the variety of “realties” that exist in the system. There is no attempt to “objectively” diagnose the system, per se. Third, the aim of P-OD is to generate new images, texts, narratives and socially constructed realities that impact how people in the system think and act. The focus is not so much on changing behaviour but on changing intersubjective reality and cognitive maps with the assumption that this will lead to new behaviour. Finally, P-OD is consistent with OD’s values of collaborative decision-making, giving people the opportunity to freely make informed choices and using the change process to develop and build capacity in the system.
From these characteristics it can be seen that P-OD, compared to contingency theory, does not prescribe an one best way to change and does not assume that there is only one reality which has to be accepted and adhered to. On the contrary, during the change process realities are socially constructed. Bushe & Marshak (2007: 5) explain that when implementing P-OD: “it has to be explicitly recognized that reality is sometimes created and maintained trough actions and negations involving power processes”. As a consequence it can be argued that the change agent begets the possibility to influence the situation through facilitating the creation of new images, texts, narratives and realities. Therefore, change agents can partially decide when and how to change, leading to an increase in their freedom of choice. In addition, in the conversations with the participants during the change process it can be decided what has to be changed. Subsequently, it can be argued that these conversations can be influenced, steered and assigned interpretation to, again creating room for the change agent to influence the situation leading to an increased freedom of choice. Compared to for example the planned approach of Buono & Kerber (2008), where regardless of the preferences of the change agent, it is pre-described when the change has to occur and what has to be changed, the possibilities to exercise influence on the situation and therefore the scope of choice in P-OD seems to be considerably wider.
In the above discussion of the guided change approach, continuous change theory and P-OD, the existence of change theories and approaches in which the possibility to influence the situation is incorporated has been proven. As a consequence it is expected that the perceived freedom of choice of the change agent in these theories and approaches increases. To further investigate the existence and the workings of this effect in practice, the following hypothesis is formulated:
Upper echelons models: the effect of age and tenure. In this final section, based on the discussion on “upper echelons models” in the first section of this chapter, a look is taken at the effect of both age and tenure on the choice process.
The basis for upper echelon models is the belief that organizational outcomes, strategic choices and performance levels, are partially predicted by a manager’s personal background and characteristics (Hambrick & Mason, 1984). In their article Hambrick & Mason (1984) developed propositions on the effect of a series of observable characteristics on the choices that managers make. For the purpose of this thesis we will take a look at two of these characteristics.
Firstly, Hambrick & Mason (1984) discuss the association between age and the way in which managers make choices. The authors paint a picture of youthful managers as attempting the novel, the unprecedented and as taking more risks. On the contrary, they expect older managers to be more conservative. The older manages are argued to have less physical and mental stamina and to be less able to grasp new ideas and learn new behaviours. Moreover, older managers are expected to attach greater importance to financial and career security. Finally, older managers are said to have a greater psychological commitment to the organizational status quo.
Subsequently, following this line of reasoning, it can be argued that the freedom of choice of a change approach, dependent upon the situational variables, will be less for elderly change agents compared to younger change agents. Therefore, the following set of hypotheses is developed:
H4: The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by age.
H5: The dependency of the perceived freedom of choice of a change approach on the degree of business complexity of the situation is moderated by age.
Secondly, Hambrick & Mason (1984) discuss the number of years inside the organization; the tenure of the managers in relation to the choices that they make. From a related variable, inside versus outside succession, the authors conclude that outsiders and therefore the managers with less tenure, tend to make more changes in structure, procedures and people. Hambrick & Mason (1984) mention amongst others a lesser commitment to the status quo as a behavioural reason for this tendency to change. In addition, it is argued that the longer that a manager stays with one organization, the less experience they will have from other organizations leading to a more restricted knowledge base.
Subsequently, again following this line of reasoning, it can be argued that the freedom of choice of a change approach, dependent upon the situational variables, will be less for change agents that work a long period in an organization compared to change agents that work a short period in an organization. Hence, the following set of hypotheses was developed:
H6: The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by tenure.
Conceptual model
This chapter theoretically explored approach on both business complexity and socio age and tenure on these dependencies
room for the change agents to exercise
hypotheses developed are graphically represented in In addition, the model conceptualizes that the the type of change approach chosen. This
that perceive a high degree of freedom of choice, approaches that they select. That is, it could
and the variety in the change approaches selected is choice of a change approach becomes
H8: The perceived freedom of choice of a change approach
By presenting the conceptual model, the
we proceed to the next chapter in which the methods used to test the hypotheses set will be discussed.
The aim of this second chapter is to explain been tested. Data were collected by condu
these two ways of data collection are
introduced. Finally, the techniques that have been
theoretically explored the dependency of the perceived freedom of choice of a change approach on both business complexity and socio-technical uncertainty including the moderat
on these dependencies. In addition, we reviewed a theory and an approach that seem to offer exercise influence on the situation as to increase their freedom of choice. The are graphically represented in the conceptual model underneath.
, the model conceptualizes that the perceived freedom of choice of a change approach the type of change approach chosen. This addition to the model is made to find out if a group of change agents that perceive a high degree of freedom of choice, also display a high degree of variety
That is, it could occur that simultaneously the perceived freedom of choice is high change approaches selected is low. Hence, in that case the degree of actual
becomes questionable. Therefore, the following hypothesis is
freedom of choice of a change approach is associated to the type of change approach
FIGURE 3: Conceptual model
By presenting the conceptual model, the necessary theoretical frameworks have been set. Henceforth, we proceed to the next chapter in which the methods used to test the hypotheses set will be discussed.
METHODS
chapter is to explain how the hypotheses introduced in the previous chapter collected by conducting an online experiment followed by a series of interviews. Firstly,
are described. Secondly, the measures that have been used in this process are that have been used to analyze the data are presented.
the dependency of the perceived freedom of choice of a change technical uncertainty including the moderating affect of both and an approach that seem to offer the situation as to increase their freedom of choice. The
freedom of choice of a change approach affects out if a group of change agents variety in the actual change simultaneously the perceived freedom of choice is high the degree of actual freedom of , the following hypothesis is developed:
the type of change approach selected
ecessary theoretical frameworks have been set. Henceforth, we proceed to the next chapter in which the methods used to test the hypotheses set will be discussed.
Data collection
Experiment. Experiments are: “studies involving intervention by the researcher beyond that required for measurement” (Cooper & Schindler, 2009: 274). In the experiment (see appendix A) the respondents were presented with cases, also called vignettes. A vignette is: “a hypothetical scenario” (Schoenberg & Ravdal, 2000). Each of these cases presented a different situation to the respondents that a change agent might face. According to Barter & Renold (2000), literature clearly demonstrates the ability of vignettes to capture how meanings, beliefs, judgments and actions are situational positioned. Hence, this makes it a perfect technique for testing the conceptual model (figure 3, p.15) presented in the previous chapter.
However, as will be elaborated upon in the measures section, not all the hypotheses from the conceptual model can be tested by means of an experiment. In table 2 underneath it can be seen which hypotheses were tested by means of the experiment.
TABLE 2: Hypotheses tested in the experiment
H1 The perceived freedom of choice of a change approach is dependent upon the degree of socio-technical uncertainty of the situation.
H2 The perceived freedom of choice of a change approach is dependent upon the degree of business complexity of the situation.
H4 The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by age.
H5 The dependency of the perceived freedom of choice of a change approach on the degree of business-complexity of the situation is moderated by age.
H6 The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by tenure.
H7 The dependency of the perceived freedom of choice of a change approach on the degree of business complexity of the situation is moderated by tenure.
H8 The perceived freedom of choice of a change approach is associated to the type of change approach selected.
By manipulating the situational variables (independent variables; IV) in the vignettes presented to the respondents and by asking questions, the effect on the perceived freedom of choice of a change approach (dependent variable; DV) could be measured. In addition, questions were asked concerning the respondent’s age and tenure. In this way, the hypothesized moderating effect of these factors upon the dependency of the DV on the IV’s could be tested. The experiment was designed as a two-group post-test (Cooper & Schindler, 2009).
The respondents were presented with four cases, two concerning the variable socio-technical uncertainty and two concerning the variable business complexity. Each of the two pairs consisted of one case in which the variable was designed to score high and of one case in which the variable was designed to score low. After the cases were read by the participants, the respondents were asked which change approach they would select. In addition, the respondents were presented with a number of statements concerning the freedom of choice that they experienced and their rating of the situational variable in question. The responses to these statements could be scored on a 7-point Likert scale ranging from 1; completely disagree up to 7; strongly agree. At the end of the experiment a number of evaluative questions were asked. The experiment was conducted by means of an online tool and was presented to 43 employees of various companies, all having a management function. Participation in the experiment was voluntary and confidentiality was guaranteed.
change managers. However, for this study it is assumed that working as a manager gives the respondents enough background and experience to realistically think of themselves as being in a situation that that he or she would face in a change process.
Another limitation of this experimental design is the distance between the vignette and social reality. What people believe they would do in a given situation is not necessarily how they would behave in actuality (Barter & Renold, 2000). Up to this point, scientists do not agree on the degree to which vignettes can mirror reality. On the one hand there are studies indicating that responses to cases mirror how individuals act in reality, on the other hand there are studies that conclude that this is doubtful. To overcome this issue, as described in the next section, interviews were conducted after the experiment. In addition, to ensure that the cases in the experiment were as realistic as possible and to test if (only) the stimulus aimed for was created and to check if the respondents interpreted the cases as intended, a pilot study was conducted preceding the experiment.
The pilot group consisted of 6 respondents, which were asked to complete the entire experiment individually, while sharing their thoughts out loud. The respondents were not allowed to ask questions at the conductor of the experiment. Afterwards, the participants of the pilot were asked to share their thoughts about the experiment. In addition, they were asked on what they considered as possible improvements. The outcomes of these sessions have been used to improve the cases and the questions. As a consequence one measure of business complexity was excluded, because respondents reported that they did not understand the accompanying statement. Moreover three statements were rephrased because they were wrongly interpreted. In addition, some suggested changes have been made to make the cases more realistically. Finally, words and parts of sentences in the cases and in the statements have been rephrased because they led to misinterpretations.
Interviews. To enhance the validity of the findings from the experiment and to test the remaining part of the conceptual model triangulation was applied. According to Olsen (2004: 3), in social science triangulation can be defined as: “the mixing of data or methods so that diverse viewpoints or standpoints cast light upon a topic”. Olsen (2004) states that mixing survey data with interviews is a profound form of triangulation. Moreover, she argues that besides increasing the validity of the claims made, triangulation also helps in deepening and widening the understanding of the results.
Therefore, semi-structured in-depth interviews (see appendix B) were conducted with thirteen respondents that also participated in the experiment. The respondents were selected based on their willingness to participate and their availability. In the table 3 underneath it can be seen which hypothesis was tested solely by means of the interviews. Note that the other hypotheses also have been tested in the interviews, but with the aim of getting additional validation and support to the results obtained in the experiment.
TABLE 3: Hypothesis tested solely in the interviews
H3 Change agents frequently exercise influence on the situation in change processes as to increase their perceived freedom of choice.
Measures
Age and Tenure. It was hypothesized that age and tenure both have a moderating effect upon the dependency of the perceived freedom of choice of a change approach on both business complexity and socio-technical uncertainty. To test this, before the cases were presented, the respondents were asked to fill out their age and their tenure in their current organization.
Freedom of choice. Carter (2004: 69) defines freedom of choice as: “lacking constraints on the reasoned selection and performance of one or more of the items on an action-menu”. In addition, in the article of Erlander (2003: 511) the proposition is stated that: “freedom of choice can be measured by the number of alternatives in the opportunity set.” Subsequently, it can be argued that the degree of freedom of choice in selecting a change approach can be measured by counting the number of alternatives that are left when taking into account the constraints apparent.
However, there is an objection to this measure that can be found in an article of Erlander (2003). Imagine a situation in which the constraints leave four options open that can be selected. Now suppose that one of these options is clearly unacceptable to a rational decision maker. Hence, although being an option it is not a realistic alternative. Therefore, the number of alternatives is not a viable measure of freedom of choice and Erlander (2003) argues that the variety of options selected is a more suitable alternative.
Based on the latter discussion, it was decided that the degree of freedom of choice of a change approach had to be measured by: the variety of alternative change approaches selected by the respondents; the standard deviation of all the selected alternatives. Nevertheless, now it could still occur that the variety of change approaches selected is low, while the respondents hold a perception of being free in the selection process. To resolve this issue, the degree of freedom of choice of a change approach was also measured by the responses to the statements that can be found in table 4 underneath. Because literature did not provide any questions and/or scales for measuring the freedom of choice of a change approach, the statements were self-developed by the researcher. In addition, the degree of freedom of choice of a change approach and the freedom of choice in the change process in total were also measured by questioning the respondents in the interviews.
TABLE 4: Items freedom of choice of a change approach Nr
1. I do not feel constrained in selecting a change approach in this situation.
2. I experience a great degree of freedom of choice of a change approach in this situation. 3. I find that approaches 1, 2 and 3 can be applied in this situation.
TABLE 5: Items socio-technical uncertainty Nr
1. There are many ways known to achieve the goal set.
2. This situation cannot be resolved by following an easy to understand sequenced steps method. 3. There is no identifiable set of established procedures and practices known to revolve this situation.
Business complexity. Previously, business complexity was defined as: “referring to the intricacy of the system, in essence, the number of components and the extent of differentiation in the organization in which it is to be implemented” (Buono & Kerber, 2010: 8). In addition, factors as: “size, geographical dispersion, the number of products and services and the number of critical stakeholders”, were mentioned as the indicators for business complexity. Therefore, again one case was developed in which the score on these indicators was intended to be high and one case was developed in which the score on these indicators was intended to be low. Again, to check if the intended scores were achieved, the items from table 6 were presented to the respondents.
TABLE 6: Items business complexity Nr
1. The organization in this vignette is large.
2. This organization in this vignette operates in many different countries. 3. This organization in this vignette offers many products and services. 4. The amount of critical stakeholders is high in this situation.
Type of change approach. In the previous chapter, change approach was defined as: “an overall perspective on the phenomenon of change and how to bring it about, including implicit and explicit beliefs concerning both the content and the process of good change” (Werr, Stjernberg & Docherty, 1997: 288). Moreover, the three change approaches of Buono & Kerber (2010) were discussed: directed, planned and guided. In the experiment the type of change approach is defined as being one of these three approaches. Based upon the information on the dimensions accompanying the change approaches presented in table 1 (p.10) and based on the text in the article, the following descriptions were developed to be presented to the respondents:
Ø Directed: you explain to the employees what has to be done and why this has to be done. The change is immediately enacted and is carried out at a high speed, without consulting the employees. The change process is tightly directed and the goals set are unchangeable.
Ø Planned: you devise a plan. The change is carried out at a high speed in cooperation with the employees, but is carefully planned beforehand. The change process is participative and the goals set are flexible.
Ø Guided: you point the way, instruct and watch over the employees. The change is immediately enacted in collaboration with the employees. The change process is improvised and no specific goals are set.
Influence exercised. In the literature chapter, it was hypothesized that change agents frequently exercise influence on the situation in change processes as to increase their perceived freedom of choice. In this study, influence exercised is seen as a synonym of manipulation, defined by Burnes (1996) as “influencing situational variables to align them with the organization’s preferred way of working” (Burnes, 1996).
in the cases presented to them. It is straightforward that the respondents as a consequence will experience and report a higher degree of freedom of choice. Imagine someone that wins a price resulting in a choice between two gifts. No suppose that a few minutes later this individual is told that he or she can also choose to receive the price in money instead of one of these two gifts. Hence, everyone would experience an increase in the freedom of choice. Therefore this hypothesis has been investigated solely in the interviews. The interviewees were asked if they knew ways to increase the freedom of choice in the cases earlier presented to them, in their own work and for a change agent in a change process. Subsequently, the interviewees were asked to elaborate upon their answers and to explain how, where and when they believe this increases the freedom of choice.
Reliability. Chronbach’s alpha was calculated to check the internal consistency and reliability of the measures used. The value 0.70 was used as a cut-off score (Malhotra, 2007). If Chronbach’s alphas scored equal as or higher than 0.7 the statements were regarded internally consistent and reliable. The results can be found in table 7 underneath.
Subsequently, it could be accepted that the items in the cases measured the same construct. For the variable freedom of choice, in all the four cases the third item was deleted to obtain acceptable or improved Chronbach's alpha scores.
TABLE 7: Chronbach’s alpha score
Validity. For each case, an exploratory factor analysis was conducted to test the validity of the constructs used (see Appendix C). A factor analysis can be used to research which item load on which variable in order to adequately minimize the number of items by combining them into the only relevant distinct variables (Malhotra, 2007). Hence, it can be tested if the items indeed measure the constructs as intended.
The analysis confirmed the theoretical constructs for all the four cases; each case revealed two factors based on eigenvalues of 1 or higher. So, in each case indeed the distinction between on the one hand freedom of choice and on the other hand either business complexity or socio-technical uncertainty was apparent. Again, the third item measuring freedom of choice was deleted because of its deviant nature; as to obtain acceptable item loads.
Data analysis
Experiment. IBM SPSS 13 was used to analyze the data obtained from the experiment. Three types of tests have been used to analyze the data.
Firstly, paired-sample t-tests have been conducted to investigate if there was a positive difference in the perceived freedom of choice of a change approach between case 1 and 2 and between case 3 and 4. Hence, the aim of these tests is to compare related samples. However, before the paired-sample t-test could be conducted, it had to be checked if the following assumptions were met: 1) the data type has to be interval, 2) the data has to be normally distributed, 3) the data has to have homogeneity of variance, 4) the data has to be independent (Field,
Case 1 Case 2 Case 3 Case 4
Freedom of choice 0,94 0,89 0,89 0,83
Business complexity 0,87 0,71 - -
2002). The data obtained with the Likert-scales is ordinal and not interval. Nevertheless, it is commonly accepted that data from Likert-scales is treated as being interval (Cooper & Schindler, 2000). Secondly, to test for the normality assumption and for the homogeneity of variance, a Kolmogorov-Smirnov test was conducted. This test compared the set of scores in the sample to a normally distributed set of scores with the same mean and standard deviation (Huizingh, 2006). If the test is non-significant (p>0,05) the distribution is not significantly different from a normal distribution. Hence, the data can be assumed to be normally distributed and as having homogeneity of variance. The results of the Kolmogorov-Smirnov test are displayed in table 8 underneath. From the table it can be seen that all three the variables are normally distributed across the cases in which they were measured. Finally, the independence of the data was assumed because the respondents were selected independently. All-in all, the use of the paired-sample t-test on the variables from the experiment could be considered as legitimate.
TABLE 8: Results of the Kolmogorov-Smirnov test
Case 1 Case 2 Case 3 Case 4
Statistic Sig. Statistic Sig. Statisic Sig. Statistic Sig. Freedom of choice 0,807 0,533 1,148 0,144 1,1013 0,257 0,724 0,671 Business complexity 1,153 0,140 1,178 0,125 - - - - Socio-technical uncertainty - - - - 1,265 0,082 1,038 0,232
Secondly, regression analyses have been performed to test the hypotheses listed in table 9 underneath. The aim is to explain variables; to find out if business complexity and socio-technical uncertainty cause the freedom of choice of a change approach. However, before it was clear that regression analyses could be performed it had to be checked if the necessary assumptions were met. A consultation of Field (2002) leans us that the assumptions that have to be met to conduct a regression analysis are similar to the assumptions that the paired-sample t-test requires. Hence, these necessary assumptions have been met and the use of regression analyses is legitimized.
TABLE 9: Hypotheses analyzed with regression analyses
H1 The perceived freedom of choice of a change approach is dependent upon the degree of socio-technical uncertainty of the situation.
H2 The perceived freedom of choice of a change approach is dependent upon the degree of business complexity of the situation.
H4 The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by age.
H5 The dependency of the perceived freedom of choice of a change approach on the degree of business-complexity of the situation is moderated by age.
H6 The dependency of the perceived freedom of choice of a change approach on the degree of socio-technical uncertainty of the situation is moderated by tenure.
H7 The dependency of the perceived freedom of choice of a change approach on the degree of business complexity of the situation is moderated by tenure.
Thirdly, Fisher’s exact test was conducted to analyze the results obtained with regards to the last hypothesis that was tested in the experiment presented in table 10 underneath. The aim is to determine association between variables; to find out if there is association between the freedom of choice of a change approach and the type of change approach selected. The first assumption, that the data have to be nominal is met (change type). Hence, so far the use of the chi-squared test of interdependence is legitimized. Nevertheless, before this test can be performed, also the assumption that: not more than 20% of the expected values may be less than 5, has to be met (Keller, 2005). After the preliminary analysis was done, from the note underneath the output tables it could be seen that this assumption had not been met. Hence, the numbers in the contingency table were too small. As a consequence, the chi-squared test of interdependence could not be performed.
TABLE 10: Hypotheses analyzed with Fisher’s exact test
H8 The perceived freedom of choice of a change approach is associated to the type of change approach selected.
Consulting the web resulted in the appropriate solution: Fisher’s exact test1. Again, before it was clear that Fisher’s exact test could be performed it had to be checked if the necessary assumptions were met: 1) the data are mutual exclusive 2) the observations are independent (Field, 2002). The first assumption was tested by calculating the sum of all cell frequencies in the table. This sum appeared equal as the number of subjects in the experiment. Hence, this assumption was accounted for. The second assumption is accounted for as well, because as explained before the respondents were selected independently. As a consequence, it was clear that the use of Fishers’s exact test was legitimate.
Interviews. The outcomes of the interviews have been prepared and analyzed with help of the techniques described in the book of Baarda & de Goede (2005). Preparing the data was done by transforming the recorded interviews into transcripts, without applying any form of interpretation yet. Next, the redundant information in the transcripts was deleted. The transcripts without the redundant information can be found in appendix E.
Subsequently, fragments were chosen as the unit of analysis, because a detailed image of the different aspects of the subjects involved is required and we want to learn about the relationships amongst the different aspects (Baarda & de Goede, 2005). Therefore, the transcripts were divided into fragments per variable. For this cause, both the variables ‘freedom of choice (of a change approach)’ and ‘influence exercised’, were divided into the freedom of choice of a change approach/ influence exercised in the cases, the freedom of choice /influence exercised in my daily work and the freedom of choice of/influenced exercised by a change agent.
Following, the fragments were assigned labels that were relevant to the hypotheses from table 9 (p.21) & 10. Labelling is defined as: naming, describing or else way coding text fragments (Baarda & de Goede, 2005, p.318). The relationships between the ordered labels were analyzed and interpreted, by means of a network analysis. Network analysis implies that the focus lies on discovering potential relationships between combinations of the labels (Baarda & de Goede, 2005). A summary table consisting of the most relevant labels and fragments can be found in appendix D.
1
