Master thesis, MscBA, specialization Change Management University of Groningen, Faculty of Management and Organization
February, 2012
PAULIEN RENKEMA Student number: 1690469 Burgemeester Stramanweg 75 1191 CX Ouderkerk aan de Amstel
tel.: +31 (0)627335750 e-mail: psrenkema@hotmail.com Supervisor/ university Dr. J. Rupert Second evaluator Dr. C. Reezigt
Supervisor/ field of study M. Koen and L. Rood
Acknowledgment: helpful comments on earlier drafts of this thesis were given by Dr. J. Rupert. Moreover, I would like to thank the Organization for providing the opportunity to conduct this research. Also, I would like to thank all the respondents of the questionnaire for their
WHETHER CHANGE MESSAGE FACTORS AND QUALITY OF CHANGE COMMUNICATION INFLUENCE PROJECT MEMBERS’ WILLINGNESS
ABSTRACT
This research investigates the relationships between change message factors, quality of change communication and the willingness to support the change of project members. A questionnaire has been distributed among the members of a change project at a multinational organization in the Netherlands in order to obtain the necessary data. The added value of this research also lies in this specific target group, which is different from other investigations that focus on the people who are affected by the change.
The linear regression analysis showed that a significant positive relationship exists between project members’ perceived discrepancy, perceived management support, experienced self-efficacy, perceived organizational valence and the willingness of the project members, and between the perceived quality of change communication and the willingness of the project members. The findings of the multiple regression analysis showed that perceived discrepancy and experienced self-efficacy are relatively to the other independent variables more important when an organization would like to increase project members’ willingness to support. Overall, the control variables nationality and full-time/part-time did not turned out to have a significant influence on any of the previously described relationships, except during the multiple regression analysis the control variable nationality had a significant effect on project member’s willingness.
This means that when enlargement of project members’ willingness is needed, the previously mentioned factors can make a contribution to this enlargement, especially perceived discrepancy and experienced self-efficacy. Moreover, these findings indicate more investigation should be done on this specific target group.
TABLE OF CONTENT
1. INTRODUCTION……… 4
2. THEORY………...……… 9
2.1 Willingness to Support the Change……...……… 9
2.2 Change Message……….……..…….………..…...……… 11
2.2.1. Model of Armenakis………...……… 13
2.2.2 Perceived Discrepancy………..………...……… 15
2.2.3 Perceived Management Support………...………17
2.2.4 Experienced Change Efficacy.……… …….………...……… 18
2.2.5 Perceived Organizational Valence.…...………….………..……… 19
2.2.6 Perceived Quality of Change Communication. ………….………….……… 22
4.2 Multiple Regression Analysis.…….………...……… 47
5. DISCUSSION AND CONCLUSION………..………..……… 50
5.1 Summary of Findings………...………...……….……...… 50
5.2 Theoretical Implications and Future Research…………...……….…… 52
5.3 Practical Implications………...………...……….……...… 54
5.4 Limitations………...…………...…………...………….…… 58
5.5 Conclusion………...…………...…………...………….…… 59
6. REFERENCES………..………..……… 61
APPENDIX 1: Covering Letter Questionnaire………..………..……… 66
APPENDIX 2: Questionnaire…...………..………..………..………67
1. INTRODUCTION
These days the topic of change is commonplace in organizations. In order to be able to respond to new technologies, environmental factors and/or the changing labor forces and compete with other organizations on the global market, organizations reorganize themselves and/or develop and implement a change initiative (Bordia, Restubog, Jimmieson and Irmer, 2011; Allen, Jimmieson, Bordia and Irmer, 2007). The same counts for a multinational organization, the subject of this research, who wants to change some of her subsidiary organizations in order to be able to better compete with competitors in the near future. The organization is an international retail group based in the United States and Europe. In November 2009 the organization
announced a series of changes in its European and US businesses to create a strong platform for future growth. The main goal of the changes is to simplify and standardize processes and structures within the organization, so that in the near future the organization is able to integrate acquisition more easily (source: www.organization.com).
For the development and implementation of the changes, in order to drive the above described strategy, a project team has been created which exists of approximately 260 persons, mostly external employees. The largest part of the external employees is working for one of the external companies which have been contracted by the organization in order to support them, for example by providing the organization with systems and manpower, during the change initiative. Others are working for themselves or for a consultancy company and have been hired in order to strengthen the project team during the change project.
governance structure gives a clear overview of the different positions within the project, and provides the project members with information about where should I go to if I experience a certain problem. FIGURE 1. Project Governance Steering Committee Program Manager Management Team Program Office Business Implementation
Business Stream Functional Stream BPM Organizational Design UAT Learning & Application Training Green & Blue
The project can be divided into three phases: initiation phase, design phase and the delivery and implementation phase. The project initiation phase started in June 2010 and ended in October 2010. The initiation phase of the project had several objectives, including determining the scope of the project, developing an agreed way of working with all the parties within the project, and the development of an integrated plan and approach for the implementation phase of the project. The design phase of the project, which started in November 2010, included among other
objectives the development of the store-CENTER and HQ (IT-solution), the development of the cut over plan, and the performance of the end-to-end test. At this moment the project is moving into the delivery phase, with the initial go-live set at 4th November 2011. The go-live of the project will be in different phases including the roll out of headquarter, distribution centers, and the pilot stores.
The changes will be implemented at two subsidiary companies of the organization. The
implementation will be done in sequential, whereby one of the subsidiary companies will be the first target, and the other subsidiary company will be the second target of the project team. The lead of the project team is in the hands of a program manager. According to the program manager the support of the team members towards the change is not optimal. This impression is based on the observation of project members who are mostly occupied with realizing their own objectives, and not willing to support each other to reach their objectives. Whereas the program manager would like to see the project members working together to reach shared project
The research question that will be examined in this research is what is the relationship between aspects of the change message, the quality of change communication and project members’
willingness to support the project?
A questionnaire has been developed and sent out, meaning that this research will be quantitative, to collect the data necessary for providing an answer to the above described research question. Moreover, the following sub-questions have been developed to be able to give an answer to the previous question:
What is the relationship between project members’ perceived discrepancy and their
willingness to support the project?
What is the relationship between project members’ perceived management support and
their willingness to support the project?
What is the relationship between project members’ experienced self-efficacy and their
willingness to support the project?
What is the relationship between project members’ perceived organizational valence and
their willingness to support the project?
What is the relationship between project members’ perceived quality of change
communication and their willingness to support the project?
are drivers of change. They supported the project during the initiation, development and the implementation phases of the project, and are therefore a very important group for the successfulness of the project. The second reason for this specific target group is because
according to the program manager of the project, improvement on the field of willingness of this group to support the change is desirable.
2. THEORY
Research has shown that the percentage of failing to implement large-scale changes successfully is relatively high, and lies somewhere between 60 and 70 percent (Burnes, 2009). Failing to implement a change initiative successfully, has both financial consequences as it has also a most likely negative effect on the success of future change initiatives. According to Gardner (2009) a major reason for failure of big change projects is the resistance of employees. Unless there are imperative reasons for the change, in general employees are not motivated to support a change initiative because of the associated uncertain future (Bouckenooghe, Devos and Van den Broeck, 2009). If the readiness for change is low, it is necessary to increase the willingness to change before change interventions are designed and implemented (Cummings and Worley, 2008; Holt, Armenakis, Bernerth, Pitts and Walker, 2007).
Therefore, the focus of this research lies on investigating the factors that have an influence on the project members’ willingness to support the change initiative. So that, if a significant
relationship is found between the independent variable and the willingness to support the change, the organization can focus on those factors in order to enlarge the project members’ willingness to support the change.
2.1 Willingness to support the change
A positive model of resistance has been chosen which focuses on the constructive value of resistance within an organization, in the form of willingness, because resistance is often
Willingness to change has been defined by Metselaar (1997: 34) as ‘A positive behavioral
intention towards the implementation of modifications in an organization's structure, or work and administrative processes, resulting in efforts from the organi zation member's side to support or enhance the change process’.
When the word resistance falls, this is often associated with negative feelings like pain, tension and frustration (Maurer, 1996). Through this negative association, a logical reaction is to avoid or reduce resistance. Some people have the opinion that resisting change is wrong, and that those who are trying to overcome resistance to change are right. When managers perceive resistance as a deconstructive behavior towards the change, they can respond hereupon through competitive and/or defensive behavior. However, resistance to organizational change can also be seen as an unavoidable and desired behavior which is a necessary part for the learning process (Block, 2010). Resistance can provide you with important feedback about the change, which can be beneficial towards the change, and thus the organization (Waddell and Sohal, 1998; Ford and Ford, 2009). An example hereof is that resistance can be seen as a signal that employees are not enough involved during the change (Waddell and Sohal, 1998).
Mossholder, 1993). In other words, in this research the focus will lie on factors that enlarge the willingness to support a change initiative, instead of focusing on factors that reduce the
resistance to the change initiative.
According to Cummings and Worley (2008), availability of resources to support the change, dissatisfaction with the status quo and significant time spent by management on the change are some indicators for willingness to the change. Other drivers for willingness to support
organizational change are: understanding the nature of the change, the level of trust and confidence, and the level of satisfaction on communication (Bekker, 2008: 25). It should be noted that when these aspects are positive, this will have a positive influence on willingness to change. However, when these aspects are negative, for example there is no understanding of the nature of the change; this will lead to a decrease in the willingness to support the change and an increase in the resistance towards the change.
Overall, a positive model of resistance in the form of willingness has been chosen, which tries to enlarge employees’ willingness through communication, facilitation and participation instead of trying to reduce or eliminate resistance (Metselaar, 1997).
2.2 Change message
The model of Armenakis, Harris and Field (1999) has been used for the measurement of factors influencing the willingness to support the change initiative. Their model is based on the needed change message for a successful implementation of a change initiative.
The sentiments shaped by the content of the change message can result in positive behavior (willingness and support), but also negative behavior (resistance) towards the change initiative. In other words, the change message sent during a change initiative has a large impact on the feelings employees have towards the change. These feelings subsequently have an influence on the reaction of these employees towards the change (Armenakis and Harris, 2001). Through this relationship between the change message and employees’ reaction towards the change (see Figure 2), it is important to ensure that the change message sent during the change initiative encourages employees to embrace and support the change initiative (Armenakis, Bernerth, Pitts, and Walker, 2007).
FIGURE 2 Armenakis Model Basic
The model of Armenakis measures the beliefs employees have towards the change initiative, when measuring the willingness to support the change initiative. According to Armenakis et al. (2007: 483) ‘A belief is an opinion or a conviction about the truth of something that may not be readily obvious or subject to systematic verification’. It might be that a significant gap exists between the beliefs of the change leaders and the employees related to the change. This can be caused through a discrepancy between the information that is available to the change leaders, and
the information that it communicated towards the employees. If this would be the case, it is desirable to close this gap because otherwise resistance from the employees towards the change initiative can be expected (Holt et al., 2007).
Within a change initiative, there are some parties, like the change agents, who can influence the content of the change message which is sent to the employees. When evidence is found for the relationship between the change message send during the change initiative and the reaction of the project members towards the change (see Figure 2), this could be a reason for the people who can influence the change message sent to respond hereupon. This can be done by adjusting the change message in such a way that it will have a positive influence on the reaction of the employees towards the change initiative. This previous described relationship has also been acknowledge by Neves (2009), who describes that for the success of a change initiative it is important that managers create willingness to support the change, while making use of a change message which encourages employees to adopt behaviors that facilitate the change initiative. So instead of showing negative behavior like for example boycotting the change or turnover, employees understand the benefits of the change and they will support the change.
2.2.1 Model of Armenakis
message predictors, on the change readiness. Later research (Holt et al., 2007; Armenakis et al., 2007), added three other dimensions, namely personal valence, organizational valence and senior leadership support, to the change message factor, which leads to the model as can be seen in Figure 3.
FIGURE 3 Model Armenakis
The item ‘personal valence’, refers to the extent someone feels that he or she will benefit from the implementation of the change (Holt et al., 2007), and includes questions about the situation at the organization after the change implementation. The target group of this thesis includes the members of a large project, which has a temporarily character, with a relatively high amount of temporarily hired employees. Most of the people who are working for the project will leave the
organization as soon as the project is successfully implemented. Because most of the project members will leave the organization after the implementation and therefore will not experience the situation after the implementation, the item ‘personal valence’ has been excluded in this thesis. Instead, organizational valence, which measures whether project members are confident that the change will have long-term benefits to the organization, will be investigated. The reason for including organizational valence is that the willingness to support the change of the internal project members, who are working for the organization, could be determined through whether the project has organizational valence for the organization or not. If the project has a detrimental effect on the organization, why would the internal project members support the project?
It is expected that the external project members are motivated to support the change because most of the external project members fly every week back and forth to the Netherlands in order to be able to work on the project, which has a relative large impact on their personal lives. Therefore, the expectation is that they are not willing to work on a project which has relatively large impact on their own lives, but has no valence to the organization they are working for (the multinational organization). Below, the four items of the Armenakis model, that will be included in this thesis, will be further explained.
2.2.2 Perceived Discrepancy
Discrepancy is concerned with the perceived difference between the actual state and the desired state that a company would like to achieve. This determines whether a change is needed or not (Katz and Kahn, 1978). When an organization experiences a discrepancy between the actual and desired state, as specific change is needed (Armenakis et al., 2007). The change will only
unique organizational situations and therefore it should correspond with the cause of the discrepancy.
As described by Kotter (1995) it is important that people are motivated to support a change, otherwise people are not willing to help and the change initiative probably fails. Starting a change initiative while the employees are not willing to support the change, is one of the most common errors why change efforts fail. Therefore, in order to create willingness to support the change employees need to see the urgency and necessity of the change for the organization, which will be seen through the experience of a discrepancy between the actual state of the organization and the desired state. Implementing a change initiative when members of an organization do not experience a discrepancy between the actual and desired state of the organization can be an indication for the organization wanting to be in fashion (Armenakis, Harris and Field, 1999). With such an implementation the organization tries to create a proactive impression for the organization. However, the long-term commitment for such implementations is missing by both the organization as also by its employees and therefore does not have a high probability of success. This stresses the importance of having a discrepancy which the change corresponds to.
following hypothesis has been developed which describes the positive relationship which is expected between discrepancy and willingness to support the change.
Hypothesis 1: a stronger belief that the organization has a need for change will have a positive influence on the project members’ willingness to support the change
2.2.3 Perceived Management Support
‘Senior leadership support refers to the extent to which one feels that the organization’s
leadership and management are committed to and support the implementation of the prospective change’ (Holt et al., 2007: 239). Kotter (1995) describes in his article that in order for the change to be successful the head of the organization should be actively supporting the change. The reason for this is that employees, who will be influenced by the change initiative, are not willing to support the change unless organizational leaders have shown clearly that they support the change initiative (Neves, 2009). The management support condition, in order for employees to be willing to support the change, could be a result of employees who have seen many change initiative fail due to a lack of management support. Besides these failure experiences, also program-of-the-month changes and quick fix have contributed to the skeptical attitude of employees against the commitment that is shown by the management team towards the change initiative (Armenakis and Harris, 2001).
talk’ (Kotter, 1995), otherwise their behavior will be inconsistent with their words and this will undermine the supportive behavior of employees towards the change. The previous has also been described by Kotter (1995) as one of the eight most common failures that prevent a change from successful implementation.
Through the usage of nonverbal and verbal communication, an employee can sense whether the manager supports a change initiative or not. If an employee believes a manager does not support the change, this will have an influence on their willingness to support the change (Armenakis et al., 2007). Based hereupon, a positive relationship is expected between perceived management support and willingness to support the change as described in the following hypothesis.
Hypothesis 2: a stronger belief among project members that the leaders are committed to the proposed change has a positive influence on the project members’ willingness to support the change
2.2.4 Experienced Change Efficacy
Normally, the feeling of self-efficacy during an organizational change increases when employees are able to perform more complex skills. This will have a negative influence on the level of stress employees can experience when a change initiative forces them to abandon old behavior (Neves, 2009). However, employees who believe that their skills and capabilities are insufficient for the activities they are asked to perform will try to avoid these activities and often experience stress (Bandura, 1986). Employees, who think that they are not capable of dealing with the new situation and the thereby coming tasks, will put their focus on the difficulties the new situation brings, which makes it more difficult for them to change their behaviors. In other words, a low self-efficacy can result in a negative cyclical relationship (Berneth, 2004).
So, when skills are required during a change initiative from which the employee feels he or she does not possess these skills sufficiently, the possibility exists that he or she tries to avoid these activities. This results in not supporting the change initiative in an optimal way by the employee. Based on the above, the following hypothesis describes the positive relationship between change efficacy and willingness to support the change.
Hypothesis 3: a stronger belief among project members that they are capable of implementing a proposed change has a positive influence on the project members’ willingness to support the change
2.2.5 Perceived Organizational valence
with the ever changing environment (Armenakis, Harris and Field, 1999). Employees who are working for the organization also experience these change implementation attempts of the organization, which most of the time end in a failure. According to Burnes (2009) the failure rate of change initiative is relatively high, and lies somewhere between 60 and 70 percent. Through this relatively high failure rate, a skeptical attitude by employees is caused, who see their organization fail when implementing a change initiative (Armenakis and Harris, 2001). This attitude can result in a lack of follow-through for the change initiative and will influence the successfulness of the initiative (Armenakis, Harris and Field, 1999).
In order to convince the employees that the organization is implementing a change initiative which has significant value for the organization, the choice for a specific change initiative needs to be communicated towards the employees of the organization. In general, an augmentation based on the feelings that a specific (senior) manager has about the value this change initiative will bring to the organization will not be sufficient for the development of willingness by
employees. It is better when the organization can explain the specific choice made while making use of more objectively augmentation. An example hereof is an analysis or diagnose of the organizations’ current and desired performance (e.g. position in the market, financial
During the communication process it is important that the information about the organizational valence of a specific change initiative reaches all the employees who will be involved during the change initiative, otherwise an information asymmetry could be the result. The employees who have been involved during the selection process of the change initiative may be well known with the organizational value of that specific change initiative. However, those employees who were not involved during the selection process have less knowledge about the needs of the organization and the particular valence of a specific change initiative for the
organization (Armenakis and Harris, 2001). So, in order to prevent the occurrence of information asymmetry, the communication of the final choice and its valence for the organization should be communicated to all the employees involved during the change initiative.
Hypothesis 4: a stronger belief among project members that the organization will benefit from the implementation of the change initiative has a positive influence on the project members’ willingness to support the change
2.2.6 Perceived Quality of Change Communication
Communication has been the subject of many research studies, but still communication is a concept which is difficult to define (LittleJohn and Foss, 2008). Communication can be defined as: ‘The process or act of transmitting a message from a sender to a receiver, through a channel and with the interference of noise’ (DeVito, 1986: 61). According to Bordia et al. (2004: 347) the quality of change communication can be described while making use of the following three components: ‘timely, accurate and trustworthy communication’. Miller, Johnson and Grau (1994), however, use other determinants for the quality of change communication, namely clarity, frequency, and openness of the communication.
Within this thesis, the two previously described definitions of the quality of change
communication will be combined, so that the most important determinants of the quality of change communication will be included, which are the perceived timeliness, the perceived clarity and the perceived frequency. First, the perceived timeliness of communication is described as informing people as quickly as possible in order for the information to still be valid (Cummings & Worley, 2008). This is in line with the definition of timeliness according to Maltz (2000: 115): ‘Refers to whether information is transmitted quickly enough to be utilized. Therefore, timeliness can be associated with the keywords current and actionable. Secondly, ‘The degree of
receiver of the message understands the message, but also whether it knows what is meant by the message send. Finally, perceived communication frequency can been described while making use of the amount and/or the duration of contact between organizational members (Mohr and Nevin, 1990). If the amount and/or the duration of contact between employees are higher, the perceived communication frequency also will be higher. The opposite is also true.
In situations in which an organization goes through a change, employees experience a high level of uncertainty. If an organizations fails to communicate towards its employees about the change initiative, this failure brings about even more uncertainties about the change initiative for the employees. This insecurity often causes more uncertainty than the change initiative itself (Schweiger and DeNisi, 1991). According to Kramer (1999), an increase in the quality level, which can be measured through the perceived timeliness, the perceived clarity and the perceived frequency, of communication can lead to a reduction of the uncertainty that is experienced by employees and caused by the change initiative. The influence of communication on the reduction of uncertainty has also been acknowledged by Burnes (2009: 460), ‘The establishment of a regular and effective communication process can significantly reduce people’s level of uncertainty’. As a result, effective communication can reduce one of the major reasons why people are not willing to get involved during the change process.
more information about the change available than is communicated, which will lead to distrust between the people who experience that feeling and those people who are responsible for the communication. Based on the above, a positive relationship is expected between the quality of communication during the change initiative and the willingness to support the change.
Hypothesis 5: a by the project members higher perceived quality (in the form of high perceived timeliness, high perceived clarity and/or high perceived frequency) of the change related
communication will have a positive influence on the project members’ willingness to support the change.
2.3 Conceptual Model
The previous described relationships between the five independent variables and the dependent variable are translated into a conceptual model which can be seen in Figure 4.
FIGURE 4. Conceptual Model
Discrepancy Self-efficacy
Management support Willingness to
3. METHODS
3.1 Sample
This research has been done at the headquarters of a multinational company in the Netherlands. The intention is to investigate both the influence of the four change message items:
organizational valence, discrepancy, management support and change efficacy, and the quality of change communication on the project members’ willingness to support the project, while making use of a questionnaire. The questionnaire has been developed in such a way that they will
measure the concepts of the previously mentioned dependent and independent variables. The members of the project who develop and support the implementation of the change initiative received the questionnaire, and are therefore the respondents for this investigation.
In Augustus 2011 the questionnaire has been sent to 120 people who are actively working on the project. A covering letter was included with the questionnaire (see Appendix 1), which provided the project members with some background information about the research subject. The
questionnaire included questions about the willingness to support the change, the perceived discrepancy, the perceived management support, the experienced self-efficacy, the perceived organizational valence and the perceived quality of change communication.
After one week, a reminder has been sent to the project members by email because only 23 project members had filled in the questionnaire. The reminder has yielded after five days 14 additional respondents. A second remember, also by email, has been sent after the second week of the opening of the online questionnaire. After the third week a total of 46 project members had filled in the questionnaire. Based upon these numbers the decision has been made to visit the company after the third week in order to remember and encourage the project members to fill in the questionnaire. During the visit several project members filled in the questionnaire. Moreover, two project members emailed their colleagues another reminder by email to fill in the
questionnaire. Finally, the program members has highlighted the importance of filling in the questionnaire during the stand-up meeting that day, which is attended by a large number of team members. Several days after my visit an additional 42 project members filled in the
deviation of 8.9. In total 56.8% of the respondents has a Dutch nationality, 2.7% English, 33.8% Portuguese and 6.7% has a different nationality. From the 74 respondents, 60 employees are working full-time on the project whereas 14 employees work part-time on the project. Because the respondents have different nationalities and a difference in the scope of their employment (full-time or part-time) which could have an influence on the relationship between an independent variable (perceived discrepancy, perceived management support, experienced self-efficacy, perceived organizational valence and perceived quality of change communication) and the dependent variable willingness to support, the control variables nationality and full-time/part-time are included in the analysis performed in this research.
3.2 Measures
The questionnaire used for this research, see Appendix 2, contains both existing statements from other scientific research and statements that have been developed for the purpose of this
research. The respondents have been asked to rate the statements while making use of a 7-point scale, whereby 1 corresponds with strongly disagree and 7 with strongly agree.
organization. Such that, ‘I intent to make time to implement the change’, turned into ‘I intent to make time to implement the project’. The Cronbach’s alpha for the scale is .75, which means that these four questions measure willingness in an appropriate manner.
Change recipients’ belief scale. The independent variables will be measured while making use of the questions developed by Holt et al. (2007). In their article, Holt et al. (2007) describe five factors that influence the variable willingness to support the change: change self-efficacy, discrepancy, personal valence, organizational valence, and senior leadership support. The factor personal valence will not be included in the questionnaire, because the items related to this factor are mostly based on the personal valence after the change implementation. Most of the project members will leave the organization after the implementation, which makes measuring the personal valence after the implementation of the project superfluous.
project’, ‘The time we are spending on the project should be spent on something else’, and ‘I think we are implementing the project just because we can’. A possible explanation of the insignificance of the previous statements is that all the four statements are reversed coded statements. All the other statements, which turned out to be significant, are non-reversed coded statements. With the coefficient alpha of .82, the internal consistency of the four included statements about the discrepancy factor is sufficient.
Perceived management support. Ten statements have been included in the questionnaire in order to measure the perceived management support. The coefficient alpha estimate of
reliability of all the ten statements was .76, when we excluded statement 5, 6, 9 and 10 the coefficient increased till .88. The statements that have been excluded from the analysis are: ‘I think we are spending a lot of time on the project and her changes when the program managers of the project do not even want it implemented’, ‘The program managers of the project has sent a clear signal: through the project the organization is going to change’, ‘The program managers of the project have not been personally involved with the implementation of the project’, and ‘I am sure that the program managers of the project will change their mind before we actually
and the limited time spent by a project member together with a project leader, which could be caused because project members are working for a different organization in a foreign country, are two possible explanations for a project member not being able to judge about the personal involvement.
With a coefficient alpha estimate of reliability of .88 we can use these six statements to measure the construct of management support appropriately. Also for this factor it is important to clarify which management the statement is about, because otherwise people who are working for a different company may interpret the statements about management support dissimilar. Therefore the statement ‘Our organization’s top decision makers have put all their support behind this change effort’ (Holt et al., 2007: 239) has turned into ‘The program managers of the project have put all their support behind the project’.
Experienced change self-efficacy. Eight statements have been included in the
questionnaire for the factor change efficacy, which measures the extent to which employees have the feeling that they are capable of executing tasks associated with the implementation of the change initiative. For this research change self-efficacy also covers the feeling of being capable to develop the change initiative, because the respondents are also responsible for the
development of the change initiative. The eight statements of self-efficacy have a coefficient alpha estimate of reliability of .66. When we deleted the first, second, fifth and seventh statement the coefficient increased until .72. The deleted statements include: ‘I do not anticipate any
out to be unreliable is missing. Only the four reliable statements of the original eight statements have been used during the data analysis. The coefficient alpha estimate of reliability of these four statements is .72, which means that the four questions used are appropriate to measure the
construct of self-efficacy. Also here, the questions have been customized whereby the words ‘the change’ has been replaced by’ the project. Some of the customized items which have been included in the data analysis are ‘I have the skills that are needed to make the project work’ (Holt et al., 2007: 238), and ‘When we implement the project, I feel I can handle it with ease’.
Perceived organizational valence. Measuring the extent to which the project members feel that the organization will benefit from the project or not, will be done while making use of seven statements. The organizational valence factor within the questionnaire contains eight statements, but with the deletion of the fifth factor ‘The organization will lose some valuable assets when we implement the project’ the reliability raised from .91 to .94. A possible explanation of the unreliability of this statement could be that it is difficult for the project members to determine what are the valuable assets within the organization and whether these valuable assets will be lost because of the project. Moreover, a significant part of the project members is working for a different company than the multinational organization, which makes the previously described determination even harder.
priorities of our organization’ has been turned into ‘The project matches the priorities of the organization’. The coefficient alpha of the factor organizational valence is .94 which indicates a sufficient internal consistency.
Perceived Quality of Change Communication. Twelve statements about the quality of change communication have been incorporated within the questionnaire. Of these twelve statements, five are based on the items developed by Bouckenooghe et al. (2009: 565). These five statements have in the questionnaire the following numbers: 1, 2, 3, 5 and 8. Also here, the statements have been modified in order to fit the target group. For example the statements ‘There is good communication between project leaders and staff members about the organization’s policy toward changes’ has been changed into ‘There is good communication between project leaders and project members concerning the project’.
Besides the questions of Bouckenoogh et al. (2009), eight other statements have been included in the questionnaire, which have been developed for this research. Most of the questions are related towards perceptions of quality of change communication like timely, clear and formal and informal communication. The self-developed questions have the numbers 4, 6, 7, 9, 10, 11 and 12 in the questionnaire. Two examples of these self-developed questions are: ‘Informal
decisions, I rely more on formal communication channels than on informal communication channels’. A possible explanation of the unreliability of the first statement could be that it has more to do with involvement of the project members than with the communication during the project. The second and third unreliable statements are related to the formal and informal communication channels within the organization. Within the questionnaire two other statements are included which are related to the formal and informal communication channels which turn out to be reliable. However, these questions are related to quality indicators of these channels like timely and clear. Probably that is the reason why these two statements, including a quality aspect, are reliable and the other two, which do not include a quality indication, are not. So, three of the twelve statements have been deleted and together these nine statements have a coefficient alpha estimate of reliability of .94, which indicates that these nine statements measure the quality of change communication in an appropriate way.
3.3 Factor Analysis
The deletion of the items that have been described previously, has been done while making use of a combination of factor analysis and reliability analysis. The original factor analysis, without the deletion of any of the items, can be found in Appendix 3. The Varimax rotation method has been used and a total of six factors have been extracted from the data. These six factors
During the deletion process both the results of the factor analysis and the reliability analysis has been used. By for example the variable management support, the factor analysis did not improve significantly with the deletion of the sixth statement. However, the reliability of the variable management support increased from .73 till .88 with the deletion of the sixth statement. Based on the reliability analysis the sixth statement of the variable management support has been deleted. The opposite is true for the variable self-efficacy, whereby the reliability analysis did not show a significant improvement in the reliability with the deletion of the second, fifth and seventh factor. However, the factor analysis improved significantly with the deletion of these items. Before the deletion, the statements of self-efficacy had loadings on four different factors, and some statements did not have a significant loading at all. After the deletion, the statements all had significant loadings on the same factor.
Component 1 2 3 4 5 6 ComQuality1 ,647 ,226 ,069 ,167 -,001 ,315 ComQuality2 .836 -.034 .316 -.115 -.096 .030 ComQuality3 ,877 -,019 ,253 ,029 -,030 ,077 ComQuality4 ,892 ,051 ,173 -,010 ,076 ,060 ComQuality5 ,874 ,113 ,040 ,205 -,067 ,066 ComQuality6 ,686 ,308 ,040 -,170 -,014 ,164 ComQuality8 ,848 ,003 ,272 -,021 -,127 ,038 FormalInformal2 ,873 -,104 ,058 ,108 ,119 -,089 FormalInformal3 ,812 ,129 ,057 ,269 ,047 -,018
Based on table 1, the following labels have been given to the six components (see table 2).
TABLE 2. Factor Labels Component Label
1 Perceived Quality of Change Communication 2 Perceived Organizational Valence
3 Perceived Management Support 4 Perceived Discrepancy
5 Experienced Self-Efficiacy
6 Willingness to Support the Change
3.4 Data analysis
Several steps have been taken prior to the regression analysis. First the incomplete questionnaires have been removed from the data set. Next, the reversed statements have been recoded so that for example a high score on a negatively charged statement turns into a low score on that same statement but than in the positive form. Recoding was necessary in order to be able to combine the separate reliable statements into one construct. Also the categorical variables, like age, have been recoded, which reduces the amount of categories within one variable.
4. RESULTS
Table 3 presents the means, standard deviations, and correlations for the dependent, independent and significant control variables which have been investigated in this research. From the table we can conclude that a significant correlation exists between the following independent variables: discrepancy and management support, discrepancy and self-efficacy, discrepancy and
organizational valence, discrepancy and the quality of change communication, management support and self-efficacy, management support and organizational valence, management support and quality of change communication and finally between self-efficacy and organizational valence. Moreover, we can see that all the independent variables have a significant positive influence on the dependent variable willingness to support the change.
The two significant control variables within this research are the nationality of the project members, and whether the project members are working part-time of full-time on the project. Nationality has a significant correlation with the perceived discrepancy, experienced self-efficacy and the perceived organizational valence, whereas the amount of time spent on the project by project members has a significant correlation with the self-efficacy experienced by the project members.
Besides nationality, the control variable full-time/part-time, whereby 0 correspond with working full-time on the project and 1 with working part-time on the project, also has a significant correlation with the experienced self-efficacy of the project members (see table 3). The
significant negative correlation found indicates that members who are working full-time on the project have a higher perceived self-efficacy than the members who are working part-time on the project.
TABLE 3.
Means, Standard Deviations, and Correlations
Variable Mean S.D. 1 2 3 4 5 6 7 1. Nationality .63 .48 2 Full-time/Part-time .47 .50 .28* 3. Discrepancy 6.13 .74 .30** .06 4. Management Support 5.50 .96 .11 -.08 .42** 5. Self-efficacy 5.64 .76 .23* -.32** .41** .23* 6. Organizational valence 5.84 .95 .42** -.16 .37** .32** .30** 7. Quality of Change Communication 4.78 1.18 .01 .03 .23* .43** .01 .17 8. Willingness to support 5.96 .82 .03 -.07 .55** .45** .48** .33** .29* *. Correlation is significant at the .05 level (2-tailed)
According to Tsui, Ashford, ST. Clair and Xin (1995), a correlation should exceed .75 in order to cause a multicollinearity problem. As can be seen in table 2, there is no correlation higher than .75, which confirms that there is no multicollinearity problem with the data.
4.1 Test of Hypotheses
In order to be able to test the hypothesis developed in this research a linear regression analysis has been performed. Whereby during the first step the control variables nationality and full-time/part time are inserted, by the second step in the analysis one of the independent variables (perceived discrepancy, perceived management support, experienced self-efficacy, perceived organizational valence or perceived quality of change communication) is inserted. The dependent variable stays the same during the analysis and is project members’ willingness to support the project. There has been chosen to include the control variables nationality and full-time/part-time based on their correlation with the independent variables as can be seen in table 3. Even though the control variables do not have a significant correlation with all the independent variables they will be included in all the linear regression analysis that will be performed. The influence of the control variables on the relationship between the independent and dependent variable can be examined when taking these control variables into account during the linear regression analysis. Because the project members originate from different countries, controlling for nationality can be meaningful.
and full-time/part-time (step 1), the independent variable is perceived discrepancy (step 2) and the dependent variable is the project members’ willingness.
TABLE 4.
Willingness as a Function of Perceived Discrepancy Willingness Step Variable 1 2 1 Nationality .021 -.200 Full-time/Part-time -.067 -.167 2 Perceived Discrepancy .627*** R2 .006 .353 ΔR2 .006 .347 F-test .202 12.709*** * p <.05 ** p <.01 *** p <.001
As can be seen in table 4, a significant positive relationship has been found between perceived discrepancy and willingness (β = .627, p < .001). In other words, the first hypothesis is accepted. So the more project members perceive a discrepancy between the current and desired state of the organization, the more they are willing to support the change initiative.
The second hypothesis expects a higher willingness to support the project when the perceived management support of project members is higher. Indeed, a significant positive relationship has been found between perceived management support and the willingness to support the project (β = .449, p < .001), which can be seen in table 4. In total 20.4% of the variance in the willingness to support variable can be explained by perceived management support. Therefore the second hypothesis is accepted, which means that if project members will be more willing to support the project when they experience support of the management of the project. Also here, the control variables turned out to be insignificant between the relationship of perceived management support on willingness (Nationality: β = -.024, p > .05; Full-time/Part-time: β = -.040, p > .05).
TABLE 4.
Willingness as a Function of Perceived Management Support Willingness
Step Variable 1 2
1 Nationality .021 -.024
Full-time/Part-time -.067 -.040 2 Perceived Management Support .449***
A higher level of project members’ self-efficacy will have a positive influence on their willingness to support the project, is the expectation of the third hypothesis. From the data provided by table 5, it can be concluded that a significant positive relationship exists between experienced self-efficacy and willingness (β = .520, p < .001). In other words, project members will be more willing to support the change initiative when they experience self-efficacy. So, also the third hypothesis is accepted. The experienced self-efficacy variable explains 24.2% of the variance in the willingness to support variable. Both the control variables did not influence the previously described relationship (Nationality: β = -.060, p > 0.05; Full-time/Part-time: β = .077, p > .05).
TABLE 5.
The fourth hypothesis describes that a positive relationship is expected between project members’ perceived organizational valence and their willingness to support the project.
The previous described relationship can be support by the data provided by table 6, (β = .382, p < .01). So also this hypothesis can be accepted based on the outcomes of the regression analysis found in table 6. The more project members perceive the valence of the project for the
organization the more they are willing to support the project. Of all the variance in the
willingness variable, 12.5% can be explained by the perceived organizational valence variable. Also here, both control variables turned out to be insignificant for this specific relationship (Nationality: β = -.137, p > 0.05; Full-time/Part-time: β =.-.049, p > .05).
TABLE 6.
Willingness as a Function of Perceived Organizational Valence Willingness
Step Variable 1 2
1 Nationality .021 -.137
Full-time/Part-time -.067 -.049 2 Perceived Organizational Valence .382**
The last hypothesis of this research, investigates the influence of the project members’ perceived quality of change communication on the willingness to support the project. According to the fifth hypothesis a positive relationship is expected between the perceived quality of change communication and the willingness to support the project. Also the last hypothesis is accepted by the data provided in table 7. A positive significant relationship has been found between the independent and dependent variable (β = .295, p < .05). This means that the higher project members perceive the quality of the change communication the more willing they are to support the project. The perceived quality of change communication explains 9.2% of the variance of the willingness to support the project variable. Also for the fifth hypothesis the control variables are insignificant (Nationality: β = .012, p > 0.05; Full-time/Part-time: β =.-.079, p > .05).
TABLE 7.
Overall, all the five hypothesis described in the theory section are accepted with the data described in the diverse tables. So the perceived discrepancy, the experienced self-efficacy, the perceived management support, the perceived organizational valence, and the perceived quality of change communication all have a significant relationship with the dependent variable
willingness to support the change initiative. Moreover, a significant positive correlation exists between the control variable nationality and the perceived discrepancy, perceived self-efficacy, and perceived organizational valence. A significant negative correlation exists between the control variable full-time/part-time and the experienced self-efficacy. During the linear regression analysis both control variables turned out to be insignificant for each individual relationship between an independent variable and the dependent variable.
4.2 Multiple Regression Analysis
TABLE 8.
Multiple Regression Analysis
Willingness
Step Variable 1 2
1 Nationality .021 -.222*
Full-time/Part-time -.067 -.033
2 Perceived Discrepancy .363**
Perceived Management Support .156 Experienced Self-Efficacy .296** Perceived Organizational Valence .123
Perceived Quality of Change Communication .120 R2 .006 .483 ΔR2 .006 .478 F-test .202 8.818*** * p <.05 ** p <.01 *** p <.001
In the multiple regression analysis, the control variable nationality turns out to be significant (β = -.222, p < .05), whereas the full-time/part-time control variable remains insignificant (β = -.033, p > .05). During the linear regression analysis the independent variables perceived management support (β = .449, p < .001), perceived organizational valence (β = .382, p < .01) and perceived quality of change communication (β = .295, p < .05) were significant. However, when looking at table 8, these independent variables, perceived management support (β = .156, p > .05),
communication (β = .120, p > .05), appear insignificant for the dependent variables willingness to support the change initiative during the multiple regression analysis. Finally, according to table 8 the perceived discrepancy (β = .363, p < .01) and the experienced self-efficacy (β = .296 p < .01) turn out to be important independent variables for the willingness to support factor, whereby perceived discrepancy is the most important independent variable. So in order to increase the willingness to support the change initiative the perceived discrepancy and experienced self-efficacy are according to the multiple regression analysis the two most influential independent variables within this research.
A possible explanation for the significance of these two independent variables could be that they are the minimal conditions that are needed in order to gain the support of the employees. The organization should have a need for the change (discrepancy) and the employees themselves should be able to support the change (self-efficacy), in terms of skills and ability. When these two conditions are met a project member will be willing to support the change. The other three variables, management support, organizational valence and quality of change communication are also important and required during the change process, but they are not a minimal condition for creating willingness to support the change. For example, when an employee does perceive for example good management support for the change initiative, but it does not perceive the
5. DISCUSSION AND CONCLUSION
The willingness to support a change initiative has been the central theme within this research. Unlike other researchers (Herscovitch and Meyer, 2002; Bordia, Hunt, Paulsen, Tourish, and DiFonzo, 2004; Allen et al., 2007; Salem, 2008), the target of this research has not been the employees which are affected by the change. Instead, the employees who are developing and implementing the change initiative have been the research object of this thesis.
The individual influence of the change message factors (discrepancy, management support, self-efficacy and organizational valence) and the quality of change communication on the willingness to support the change initiative have been investigated.
5.1 Summary of Findings
Within this research the main research question has been what is the effect of the change
message factors and the quality of change communication on project members’ willingness to support the project? This research question has been divided into five sub research questions
which will be individually discussed below.
Also a positive significant relationship has been found between the perceived management support and the willingness to support the change initiative from a project member. This proves that when project members see that the management of the project is willing to put energy, time and resource in the project, they are themselves also more willing to support the project
(Armenakis and Harris, 2001).
The relationship between experienced self-efficacy and willingness to support the change initiative also turned out significantly positive. The relationship also turned out to be significant during the multiple regression analysis. This means that when employees have the feeling they are capable of doing their tasks well and they possess the right skills in order to execute their job within the project they will be more supportive towards the project.
The fourth sub question describes a positive relationship between perceived organizational valence and willingness to support the change initiative turned out to be significantly positive. So when employees have the feeling that the organization will benefit from the change initiative, they will be more willing to support the initiative.
Finally, the significant positive relationship between perceived quality of change communication and willingness to support the change initiative also has been proven by the data. So when the perceived quality of the change communication grows, this will have a positive influence on the willingness to support the initiative.
significant relationship with willingness, are the two most important independent variables. In other words, discrepancy and self-efficacy are more important independent variables in order to create willingness to support than the independent variables management support, organizational valence and quality of change communication are.
5.2 Theoretical implications and future research
The focus of this research has been different than that of most researchers who investigated the willingness of employees, because those researchers investigated the willingness of employees affected by the change (Herscovitch and Meyer, 2002; Bordia, Hunt, Paulsen, Tourish, and DiFonzo, 2004; Allen et al., 2007; Salem, 2008), whereas this research investigates the willingness of the project members who are designing and implementing the change.
Within this research we incorporated the factors that have been considered in research whereby the target group has been the employees who are affected by the change. In other words, we intended to research whether the factors that influence the willingness of employees affected by the change also had an influence on the willingness of the employees working on the change, in this specific case the project members.
The independent variables which have been used in this research have also been the subject of other investigations. In the article of Armenakis et al. (1999), the positive significant relationship between the independent variables discrepancy, management support, self-efficacy and
organizational valence and the dependent variable of willingness has been found. However, the target group of this investigation has been the employees who are affected by the change. The same counts for the quality of change communication which turned out to have a positive
target group as the article of Armenakis et al. (1999). The significant relationships found in this research suggest that the same factors that influence the willingness of employees who are affected by the change (Armenakis et al., 1999; Bockenooghe et al., 2009) also influence the willingness of the project members who are working on the change initiative. This research suggests that this is true for the change message and quality of change communication factors. It could well be that factors, which have not been taken into account in this research, also have an influence on the employees who are working on the initiative. Factors which have not been investigated in this research could be the subject of future research on the willingness of
employees who are working on the change initiative. An example of such a factor is the personal valence of the change initiative for the employee, which has not been taken into account in this research (Armenakis et al., 1999). Also investigating the type of communication channels (formal or informal) could be interesting. Kotter (1995) describes in his article that in the more successful transformation efforts the organization made use of all existing communication channels the organization possesses. Therefore, it can be expected that a higher success rate will be obtained within an organization when it makes use of both informal and formal
communication channels during a change process. Besides, the individual influence of communication channels, like e-mail, face-to-face etcetera, can be taken into account. When project members are not willing to support a change initiative and behave
the willingness to support the change initiative of employees who are affected by the change will be relevant. This is also a reason to encourage future research to investigate the target group of employees who are working on the change initiative.
5.3 Practical implications
The previous described findings do have several practical implications for the organization from which the data is obtained. First of all, perceived discrepancy has a significant positive effect on the willingness to support the change initiative. This means, that in order for an organization to get support from its employees who are working on the change initiative, it is important that the employees see the need of the organization for the change initiative. A perceived discrepancy can arise by project members after for example a meeting in which the need for the change has been communicated towards the project members. Sometimes, in order to convince project members of the significant discrepancy between the actual and desired state, some exaggeration can be helpful (Kotter, 1995). An example hereof is to describe the current state of the
organization as a threat to the performance of the organization, even though this is only partial true. However, exaggeration must be done with caution, such that the employees do not lose their trust in the organization, its management and/or in the change initiative. The first practical
The second significant positive relationship exists between perceived management support and willingness to support the change initiative. In general, the management of the project does not have regularly contact with every project members. However, as is showed by the positive relationship it implies that the willingness of the project members is dependent on the support showed by the management towards the project. Therefore, it is important that the way that the management of the project shows its support to the project reaches all project members. This can be done in several ways, for example organize a meeting in which you invite all the project members. During this meeting, the project’s management gets the opportunity to explicitly show its support to the project to all the project members. A helpful tool could be a presentation in which they show the regular dedication to the project in terms of effort, time and money but also how they will be involved personally (tasks, responsibility, effort, and etcetera). In general, there will be a lot of changes during a change initiative, for example scope, timescale and budget. When these changes have a significant impact on the project and its members, it could be that the members lose their trust in the project. Therefore, when a significant change occurs it could be wise for the management to once again explicitly show their support to the initiative even after the changes. This can be done by for example an email which states that the management still believes in the success of the project even after a certain change to the project. Of course there are other ways to explicitly show the support of the management to the project. What is important is that the message reaches all the project members, that the message is clear and honest and finally that the management also behaves like said: ‘walk the talk’.
necessary skills and are able to perform the tasks which are linked to the change initiative. First, the organization should try to figure out whether employees have the feeling they can handle the tasks and activities. It could well be that the project member itself starts about this subject, and approaches his or her superior. However, not every project member that experiences some difficulties during the execution of its task is willing to address this to his or her superior. They might be afraid that it will highlight a weakness of them, which could for example influence their appraisal negatively. The appointment of a trustee, which the project members can consult when it experience difficulties, could help these people. When the organization has an indication of the difficulties experienced by employees they can offer the employees tools in order to try to reduce and/or eliminate these difficulties. Two examples are education and training. If desired a
development plan can be designed which describes review points to see whether the training and/or education helped the employee and if needed what should be changed in order to support the employee during its development. Sometimes, the employee has the feeling he or she has difficulties during the execution of its tasks whereas in practice this is not the case. These employees should get mental support in order to eliminate this incorrect feeling. Overall, the organization should sent the message towards its employees that having difficulties during the execution of their tasks is nothing to be ashamed of and that the organization is more than willing to help the employee to increase their self-efficacy.
