Employee reactions to technological identity change:
Investigating the moderating role of computer self-efficacy and work identity
rigidity.
Student name: Jens van de Meulenhof
Student number: 11110910
Thesis supervisors: H.A. Berkers MSc & dr. S.T. Mol Course: Master Thesis
Track: Leadership and Management
Master: Business Administration
University: University of Amsterdam
Date: 23 June 2017
Version: Final version
Statement of Originality
This document is written by Jens van de Meulenhof, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this docu-ment is original and that no sources other than those docu-mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
With the growing number of new technologies introduced at work, the way individuals have to do their job will change eventually. Since work is an essential part of a person’s identity, technological changes in the workplace could lead to changes in the identity, and these changes could result in both positive and negative reactions from employees. This paper ar-gues that technological identity changes will be positively related to resistance to change and negatively related to willingness to change. Since employees differ in their technological skills and identities, this study expects to find a weakening interaction effect of computer self-efficacy and a strengthening interaction effect of work identity rigidity in the relationship of technological identity change with resistance to change and willingness to change. An exper-imental vignette study among 253 Dutch accountants and teachers was conducted using an online survey. The results did not find that technological identity change was related to re-sistance to change and willingness to change, and no moderating effect for computer self-efficacy and work identity rigidity was found. The author recommends further investigation of the concept of technological identity change and the investigation of other professions in the identity-related research as well.
Table of Contents
1. Introduction p. 5
2. Literature Review p. 7
2.1. Technological Identity Change p. 7
2.2. Resistance to Change p. 9
2.3. Willingness to Change p. 10
2.4. Computer Self-efficacy p. 11
2.5. Work Identity Rigidity p. 13
2.6. Conceptual Model p. 14 3. Method p. 15 3.1. Research Design p. 15 3.2. Procedure p. 16 3.3. Sample p. 17 3.4. Materials p. 18 3.4.1. Independent variable p. 18 3.4.2. Dependent variables p. 19 3.4.3. Control variables p. 20 3.5. Data Analyses p. 21 4. Results p. 22 4.1. Descriptive Analysis p. 22 4.2. Regression Analyses p. 23 5. Discussion p. 24 5.1. Findings p. 24
5.2. Limitations and Recommendations for Future Research p. 26
5.3. Conclusion and Practical Implications p. 27
6. Reference List p. 28
7. Appendices p. 34
7.1. Appendix A: Scenarios p. 34
1. Introduction
Highly competitive markets and rapid technological innovations are pushing organizations to constantly adapt and change. In the ever-evolving business market, organizational change is necessary and inevitable for companies to remain competitive against other businesses (Sackmann, Eggenhofer-Rehart, & Friesel, 2009; Salerno & Brock, 2008). As a result of the growing number of new technologies introduced in workplaces, the way individuals have to do their job will eventually change as well. Work is an essential part of a person’s identity (Elsbach, 1999), and since work plays an increasingly central role in peoples’ lives (Hall & Chandler, 2005), this could mean that when the work changes, the identity attached to it – the work identity (Miscenko & Day, 2015) – will change too. When the change in identity is then related to new technologies or innovation, the change can thus be described as technological identity change: the extent to which the introductions of new technologies in the workplace affect the employees’ identity.
It is important to understand how employees adjust to change, since change can only be successful when the individuals confronted with the change show support and adapt to the change (Armenakis & Harris, 2009). However, successfully implementing change within an organization does not appear to be self-evident: around 70% of the change initiatives in or-ganizations fail (Burnes, 2011). While many reasons have been given to explain failures in change (Burnes & Jackson, 2011), one of the most frequently cited explanations is employee resistance (Dent & Goldberg, 1999; Erwin & Garman, 2010; Herold, Fedor & Caldwell, 2007; Oreg, 2006). Next to the resistant outcome of change, research has recently focused on the adaptive attitude of employees as well, such as the willingness to change (Oreg, Vakola & Armenakis, 2011). Although these studies have contributed to the knowledge of attitudinal reactions to change, Shoss, Witt and Vera (2011) argue that little research has been done re-garding the intentions and actual behavior of employees in situations of change. Since the intention is assumed to be the immediate predictor of behavior (Ajzen, 1991), more knowledge of employee intentions in relation to change will mean more knowledge about the actual behavior of employees. Therefore, this paper investigates both the intention to resist change and willingness to act towards change.
This paper argues that technological identity change will be positively related to re-sistance to change and negatively related to willingness to change. Changes can force individ-uals to reconstruct their identity (Snow & Anderson, 1987; Sveningsson & Alvesson, 2003). Reconstructing the identity can be viewed as a loss of status or feeling of uncertainty (Bordia, Hobman, Jones, Gallois & Callan, 2004; Dent & Goldberg, 1999), and therefore trigger
re-sistance (Piderit, 2000), and thus lower willingness to change. By investigating both outcome variables, the results make it possible to compare and examine which variable has the strong-est relation with technological identity change. In other words, the results can indicate wheth-er the employees’ resistance to change or the willingness to change will be strongwheth-er when technological identity change occurs.
This paper provides a second unique contribution to the scientific literature by intro-ducing two opposing moderator variables in the relationship of technological identity change with resistance and willingness to change. Not every individual has the same skills when it comes to technology. Therefore, the first moderator is computer self-efficacy, the extent to which one believes he or she can handle a computer related task (Compeau & Higgs, 1995). The conservation of resources theory (COR) suggests that people want to protect their re-sources and prevent themselves from negative outcomes (Hobfoll, 1989). Self-efficacy can be seen as a resource (Bandura, 1979). According to Hobfoll (2002), resources will generate oth-er resources. Individuals with high self-efficacy will gathoth-er more resources, which will make them better at coping with new technologies (Hobfoll, 2002). Therefore, this paper argues that high computer self-efficacy can be seen as a strengthening concept that will weaken the rela-tionship of technological identity change with resistance and willingness to change.
Since every individual has a unique identity (Brewer, 1991), it can also be argued that the relationship between technological identity change with resistance and willingness to change could differ, based on how rigid one’s identity is. Research suggests that the stronger an identity is, the harder it is for an individual to reconfigure or let go of that identity (Ash-forth, Harrison & Corley, 2008). It can therefore be suggested that individuals with a strong identity will find it harder to adapt to change and thus have a more rigid work identity (Car-dador & Caza, 2012). Since people with a rigid work identity are unable or unwilling to adjust their viewing to fit with the reality of their work situation (Briscoe & Hall, 1999), this auto-matically implies that for them it will be even harder to adjust their identity when change oc-curs. Therefore, this paper argues that high work identity rigidity can be seen as an obstruc-tive concept that will strengthen the relationship of technological identity change with re-sistance and willingness to change.
The unique aspect of this research lies in the investigation of the chosen moderators. Computer self-efficacy and work identity rigidity can be seen as two opposing variables that both connect with the employee’s identity. Computer self-efficacy can be seen as a positive, strengthening concept that connects to the ‘technological’ aspect of technological identity change, while work identity rigidity can be seen as a negative, obstructive aspect that
con-nects with the ‘identity’ aspect of technological identity change. This research attempts to compare the outcomes of both variables towards the relationship of technological change identity with resistance to change and willingness to change. Organizations will benefit from this insight because the results can help organizations to pick the right tools to support their employees in times of technological change; tools for helping employees to operate with new technologies, or tools to lessen the rigidity of the employees’ identities. Furthermore, the find-ings of this study can contribute to a clear picture of how employees experience new technol-ogies in the workplace and what the psychological impact of technological identity changes are for individuals.
This study seeks to answer the question to what extent computer self-efficacy and work identity play a role in the relationship of technological identity change with resistance and willingness to change. Therefore, the research question of this study is: “How do
comput-er self-efficacy and work identity rigidity influence the relationship of technological identity change with resistance to change and willingness to change?”. This paper tries to answer the
research question by conducting an experimental vignette study that focuses on the introduc-tion of new software and Big Data tools within an organizaintroduc-tion. First, a literature review is given to examine what is known about the subject of this paper. Subsequently, the research design, procedure and materials of the experimental vignette study are described. Finally, the results of the experiment are examined and discussed, the research question is answered and a conclusion is given.
2. Literature Review 2.1. Technological Identity Change
This study focuses on technological identity change, the extent to which a technological change affects the employee's identity. Technological change is change that arises as the re-sult of innovations in technology, such as new production methods, new designs, and new techniques of organization (Mansfield, 1968). Identity is described as “a self-referential de-scription that provides contextually appropriate answers to the question ‘Who am I?” (Ashfort et al., 2008, p. 327). An identity can be seen as an individual’s unique way of being and act-ing. According to social identity theory and the closely related self-categorization theory (Tajfel & Turner, 1986; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), identity at the individual level can be divided into two parts. One part is the personal identity, which con-tains specific characteristics that differentiate an individual from others. The second part is the social identity, which contains categorizations of the self into social units - such as race,
gen-der, and organizational membership - that depersonalize the self-concept (Brewer, 1991). Since individuals want to be a part of a collective and want to be unique at the same time, a dynamic tension exists between the social identities and the personal identities (Brewer, 2012). From this dynamic perspective, it can be concluded that the individual’s identity is something cyclical and ongoing, and can therefore change (Pratt, 2001).
The dynamic relationship between the individual’s social identities and personal iden-tities can also be related to work. When an individual actively responds to the dynamic rela-tionship between the personal and social identities, it is called identity work, which Snow and Anderson (1987, p. 1348) define as the “range of activities that individuals engage in to cre-ate, present, and sustain personal identities that are congruent with and supportive of the self-concept”. Sveningsson and Alvesson (2003, p. 1165) elaborate further on the concept and state “identity work refers to people being engaged in forming, repairing, maintaining, strengthening or revising the constructions that are productive of a sense of coherence and distinctiveness”. Thus, when individuals engage in identity work, they try to compromise, balance and maintain the relationships and divisions between their personal identity and the social identities.
Work plays a role in the individual’s identity. Since work is a social identity (Brewer, 1991), organizations can be seen as crucial in shaping an individual’s identity (Elsbach, 1999). Indeed, research has shown that the types of identities that individuals work on influ-ence the way they make decisions in organizations (Alvesson & Willmott, 2002; Amiot, de la Sablonnière, Terry & Smith, 2007), and influence their careers (Ibarra & Barbulescu, 2010). The identity a person has at work is called the work identity: a collection of meanings at-tached to the self by the individual based on personal characteristics, group membership, and (social) roles (Miscenko & Day, 2015). Since identity can be subject to change (Pratt, 2001), this means that when the work changes, it is possible that the identity attached to it will change too. Furthermore, research has found that with large changes, there is less perceived continuity of identity when compared to smaller changes (Van Knippenberg, van Knippen-berg, Monden, & de Lima, 2002).
It can be argued that change can be identity-related or non-identity related, since re-search has found that contextual influences can change the role content for an employee and thus change the corresponding identity (Vähäsantanen & Eteläpelto, 2009). This division can be explained by an example in which an employee encounters changes in the work role. For example, when an employee encounters an organizational change in which the employee has to adjust his or her existing tasks, the work role changes, which means that the social identity
attached to it will have to be adjusted. According to identity work, the individual will then actively have to reorganize the identity because of the changed social contexts (Snow & An-derson, 1987; Sveningsson & Alvesson, 2003). Since the individual engages in identity work, the change is identity-related. On the other hand, employees can also encounter organizational changes that do not lead them to adjust their existing work tasks, which means there is no change in social context. The employee does not have to engage in identity work in this situa-tion and the change is therefore non-identity related. Thus, this paper divides identity change into two types: identity related changes and non-identity-related changes. When the changes are then related to technology, the change of identity can be called technological identity change.
2.2. Resistance to Change
Individuals can react to change in different ways. They can show either resistance to the change or approval to the change (Vähäsantanen & Eteläpelto, 2009). Furthermore, people differ in their general tendency and disposition towards change (Oreg, 2003; Oreg at al., 2008). In organizational literature, resistance to change is often regarded as problematic be-havior for organizations, since it is regarded as a critical determinant of failure of change ini-tiatives (Herold et al., 2007).
Resistance to change can be defined in several ways. Folger and Skarlicki (1999, p. 36) define resistance to change as “employee behavior that seeks to challenge, disrupt, or in-vert prevailing assumptions, discourses, and power relations”. However, the definition for resistance to change used in this paper is based on Ajzen’s (1991) theory of planned behavior. According to this theory, the individual’s attitude influences the behavior via the individual’s intention. The attitude toward the organizational change can be defined as an individual’s overall evaluation of change within the organization (Lines, 2005), and this evaluation can be either positive or negative. The intention indicates the readiness of an individual to perform a given behavior (Ajzen, 1991). When an individual has a strong, negative attitude towards change, he or she will more likely have the intention to resist and hinder change (Lines, 2005). Since intention is assumed to be the immediate predictor of behavior (Ajzen, 1991), this research focuses on researching the intention an individual has to resist change. Thus, in this paper, resistance to change relates to the extent to which an individual would resist to act towards a change when the change would occur.
The concept of resistance to change can be further explained by examining what indi-viduals resist exactly when change occurs. Dent and Goldberg (1999) argue that indiindi-viduals
do not resist change itself, but rather the anticipated expectations attached to the associated change, such as loss of status. Furthermore, processes in which change plays a role are often associated with uncertainty, and uncertainty is often quoted as a source of resistance to change (Bordia et al., 2004). These statements imply that uncertainty and the feeling of losing status or control over the work triggers resistance (Piderit, 2000). Employees might for exam-ple be afraid that the change requires them to perform more work, or that the change makes them less able to perform the job, leading to job loss. Resistance that relates to the individu-al’s belief that the change might have a negative impact on his or her experience of work is what Van Dijk and Van Dick (2009) call ‘person-oriented resistance to change’ (p. 144).
An association between technological change and resistance to change can be made as well. Oreg (2003) found that higher levels of resistance to change were associated with re-sistance to innovation. Technological change is change that arises as the result of innovation (Mansfield, 1968). It can therefore be argued that resistance to change could also be associat-ed with technological change. This paper states that technological changes can be split up in non-identity related change and identity-related change based on whether an individual has to engage in identity work or not. According to Burns and Stalker (1961), innovation within a company requires a significant amount of change from the employees, such as adjusting and redefining the individual tasks through symbolic negotiation with others (Ibarra, 1999). How-ever, for most people it is difficult to engage in this level of change (Burns & Stalker, 1961). With identity-related change, individuals have to reconstruct their identity because of changed work contexts (Snow & Anderson, 1987; Sveningsson & Alvesson, 2003). When the change affects the work role, this could be viewed as a loss of status or feel as uncertainty (Bordia et al., 2004; Dent & Goldberg, 1999), and therefore trigger more resistance (Piderit, 2000). Indi-viduals that face non-identity related changes will not have to reconstruct their identity and will not encounter uncertainty. Thus, the first hypothesis states that technological identity change will be positively related to resistance to change.
H1: Technological identity change is positively related to resistance to change.
2.3. Willingness to Change
The opposite of resistance to change is willingness to change. This concept can be described on the basis of the theory of planned behavior (Ajzen, 1991) and employees’ attitudes towards organizational change (Elias, 2009; Lines, 2005). With resistance to change, individuals have a negative attitude towards change, which will lead to the intention of resistance to act
to-wards the change when the change is encountered (Ajzen, 1991). Individuals with a strong, positive attitude towards change will be more likely to behave in an effortful way to support and facilitate the change initiative (Lines, 2005). Acceptance and support from employees towards changes are regarded as important determinants for successful changes (Herold et al., 2007). Thus, in this study, willingness to change relates to the extent to which an individual would be willing to act towards a change when the change would occur.
Willingness to change can further be explained by the concept of change-supportive employee behavior. According to Kim, Hornung and Rousseau (2011), individuals with pro-active career behavior and focus on the work value of others have higher levels of engage-ment when facing change, while individuals that show passive career behavior and self-centered work values have lower levels of involvement during change. Research also found that when employees have the anticipation of a positive outcome of change, they show greater acceptance of the change and higher willingness to participate in it (van Dam, 2005). Fur-thermore, the extent to which an organization provides support also appears to determine whether an employee is willing to change. Among individuals who tend to be dogmatic and close-minded, support from the work environment will make individuals more willing to par-ticipate in an organizational change (Wanberg & Banas, 2000).
This paper argues that technological identity change will be negatively related to will-ingness to change and uses the same argumentation as for resistance to change. When identi-ty-related change occurs, individuals have to reconstruct their identity because of changed work contexts (Snow & Anderson, 1987; Sveningsson & Alvesson, 2003). The reconstruction of the identity could be viewed as a loss of status or feel as uncertainty (Bordia et al., 2004; Dent & Goldberg, 1999), and therefore trigger more resistance (Piderit, 2000). Since ness is the opposite of resistance, the reconstruction of the identity will lead to less willing-ness. Individuals that face non-identity related changes will not have to reconstruct their iden-tity and will not encounter uncertainty. Thus, the second hypothesis states that technological identity change will be negatively related to willingness to change.
H2: Technological identity change is negatively related to willingness to change.
2.4. Computer Self-efficacy
Since very person is unique, technological skills will differ per person. It can therefore be argued that the relationship between technological identity change with resistance and will-ingness to change will differ for individuals, based on their technological skills. According to
Bandura’s social cognitive theory (1977), self-efficacy is the degree to which an individual believes he or she is capable of fulfilling a given task. The self-efficacy of an individual plays an important role in the way a person approaches goals, challenges, and tasks. An individual with high self-efficacy is more likely to believe he or she can cope with challenging problems and can swiftly recover from disappointments and setbacks. Individuals with low self-efficacy do not believe they are able to perform well, which leads them to avert tasks that are challeng-ing (Bandura, 1977). Self-efficacy can also be focused on technology. This type of self-efficacy can be defined as computer self-self-efficacy - the extent to which an employee believes he or she is capable of handling a computer oriented task (Compeau & Higgs, 1995; Gallivan, Spitler & Koufaris, 2005).
The differences in individual’s computer self-efficacy can be explained on the basis of a person’s resources. Hobfoll’s (1989) conservation of resources theory (COR) states that people strive to gain, maintain, foster and protect resources. These resources can be object resources, such as a house, condition resources, such as employment, energy resources, such as money, or personal resources, such as self-esteem and self-efficacy, When an individual looses a resource, or there is a threat of loss, individuals will experience stress (Hobfoll, 1989). Thus, the theory suggests that people want to protect their resources and prevent them-selves from negative outcomes. Self-efficacy can be seen as a resource that can help to cope with stress (Bandura, 1979). From a COR perspective, self-efficacy can therefore be seen as a resource to prevent resource loss and cope with threats to resources.
In addition, individuals do not only want to protect their resources, but also want to accumulate them. Hobfoll (2002) argues that resources will generate other resources, and this may result in positive outcomes, such as better coping with the encountered situation or better adaption to new technologies. This is demonstrated in the paper of Chen, Westman, and Eden (2009). For a newly implemented IT system, the researchers developed an intervention in the form of a workshop that facilitated resources that could prevent or alleviate stress. A part of the participants followed a workshop that gave them information about the means, equipment, and materials needed to perform the job effectively, while the other part of the participants did not follow the workshop. The results revealed that the participants that had followed the workshop had increased efficacy about the new system. The workshop strengthened the par-ticipants’ resources, which enabled them to cope with the difficulties of the change.
Based on the individual’s technological skills, the relationship between technological identity change with resistance to change and willingness to change may differ for individu-als. The first hypothesis of this paper stated a positive relationship of technological identity
change with resistance to change, while the second hypothesis stated a negative relationship of technological identity with willingness to change. This paper argues that a high computer self-efficacy can be seen as a strengthening concept, because an individual with high comput-er self-efficacy believes he or she is capable of fulfilling the task (Bandura, 1977). Furthcomput-er- Further-more, individuals with high self-efficacy will gather more resources, which make them even better at coping with new technologies (Hobfoll, 2002). From this perspective, two new hy-potheses can be stated. Firstly, a high self-efficacy will weaken the positive relationship of technological identity change with resistance to change. Secondly, a high self-efficacy will also weaken the negative relationship of technological identity change with willingness to change.
H3a: Computer self-efficacy influences the relationship of technological identity
change with resistance to change, such that when computer self-efficacy is high, the positiverelationship of technological identity change with resistance to change will be weaker.
H3b: Computer self-efficacy influences the relationship of technological identity
change with willingness to change, such that when computer self-efficacy is high, the negative relationship of technological identity change with willingness to change will be weaker.
2.5. Work Identity Rigidity
Every individual has a unique identity (Brewer, 1999), and it can therefore be argued that the relationship between technological identity change with resistance and willingness to change could also differ, based on the individual’s identity. Research suggests that the stronger an identity is, the harder it is for an individual to reconfigure or let go of that identity (Ashforth et al., 2008). Furthermore, people differ in the degree of flexibility they attach to their work identity (Cardador & Caza, 2012). This suggests that individuals with a strong identity will find it harder to adapt to change and thus have a more rigid work identity. According to Car-dador and Caza (2012), work identity rigidity is described as the degree of unwillingness and reluctance to change one’s work identity even if change is necessary. Individuals with a rigid work identity hold on to a fixed idea of who they are at work, which limits their opportunities to adapt to change. When individuals encounter stressors that challenge their current identity or face changes that require identity change, the person’s identity becomes even more rigid (Block, 1961). Furthermore, research has found that individuals with a rigid work identity
have a one-dimensional image of who they are at work and this makes them unable or unwill-ing to adjust their viewunwill-ing to match the reality of their work situation (Briscoe & Hall, 1999).
As with computer self-efficacy, the phenomenon of work identity rigidity can also be explained on the basis of Hobfoll’s (1989) COR model. The theory suggests that people want to protect their resources and prevent themselves from negative outcomes (Hobfoll, 1989). According to COR, individuals strive to obtain, maintain, protect and accumulate their re-sources (Hobfoll, 2002). Individuals with a rigid work identity are less willing to adjust their identity (Briscoe & Hall, 1999). When an individual with a rigid work identity has to deal with identity changes, this dealing might be perceived as a loss, which is considered stressful (Hobfoll, 1989). The individual will try to maintain his or her resources and thus resist adjust-ing the identity, which gives him or her a rigid work identity. This implies that people with a rigid work identity will naturally try to resist any change to their identity, and that rigid work identity can be seen as an obstructive concept. From this perspective, two new hypotheses can be stated: high work identity rigidity will strengthen the positive relationship of technological identity change with resistance to change, and strengthen the negative relationship of techno-logical identity change with willingness to change.
H4a: Work identity rigidity influences the relationship of technological identity
change with resistance to change, such that when work identity rigidity is high, the positive relationship of technological identity change with resistance to change will be stronger.
H4b: Work identity rigidity influences the relationship of technological identity
change with willingness to change, such that when work identity rigidity is high, the negative relationship of technological identity change with willingness to change will be stronger.
2.6. Conceptual Model
The conceptual model of this paper is shown in Figure 1. As the research model illustrates, hypothesis 1 states that technological identity change will be positively related to resistance to change. Hypothesis 2 states that technological identity change will be negatively related to willingness to change. The first moderator of this study, computer self-efficacy, is investigat-ed in hypothesis 3a and 3b. Computer-self-efficacy is expectinvestigat-ed to have a weakening effect on the relationship of technological identity change with both resistance to change and with will-ingness to change. The second moderator of this study, work identity rigidity, is investigated
in hypothesis 4a and 4b. Work identity rigidity is expected to have a strengthening effect on the relationship of technological identity change with both resistance to change and willing-ness to change.
Computer Self-Efficacy (H3a - & H3b -)
Technological Resistance to Change (H1+)
Identity Change
Willingness to Change (H2 -)
Work Identity Rigidity (H4a + & H4b +)
Figure 1. Conceptual Model.
3. Method 3.1. Research Design
The overall design of this study is a quantitative research design, aimed at Dutch accountants and teachers (elementary education and secondary education). For this study, an online exper-imental vignette design was used (i.e., 2 x 2) in which change (i.e., non-identity related, or identity related) and the way the change was framed (i.e., threatening vs. non-threatening) was manipulated. Although this paper will only focus on the type of change, the framing of change is also mentioned here, since this experimental vignette study has been used for the individual theses of 6 other MSc students too.
The choice for an experimental vignette study was made because this paper investi-gates work behaviors that are not easily observable, and experimental vignette studies can provide valuable insights into these behaviors (Aguinis & Bradley, 2014). Furthermore, the design of experimental vignette studies can enhance the experimental realism (Aguinis & Bradley, 2014). The possibility to manipulate relevant variables while retaining realism bene-fits the internal validity (Finch, 1987). This paper investigates technological changes in the workplace and realism is therefore necessary. In the experimental vignette, a hypothetical situation (vignette) was shown to respondents to trigger a particular evaluation from the re-spondent. The respondents’ evaluation was then measured on the basis of questions about the vignette through a questionnaire in an online survey (Atzmüller & Steiner, 2010).
A pilot study (N = 80) among BSc students was done to ensure maximum variation in the scenarios used. No difference was found between vignettes of teachers and accountants and no gender differences were found. However, the control condition was problematic. Initially, six scenarios were made for this study: next to the ‘non-related identity condition change’ and ‘identity related change condition’, there was a ‘no change’ condition. However, the pilot study revealed there was not enough variance between the ‘no change’ condition and the ‘non-identity related change’ condition. Therefore, the ‘no change’ condition was removed. This had no further consequences for the study.
Every respondent was randomly assigned to one of the four conditions, which means he or she received one scenario and responded to a number of statements and questions relat-ed to that scenario. Table 1 shows the four conditions, including the number of respondents per condition. Since this research focused on both teachers and accountants, the content for the scenarios was different for both professions. This was done to make sure the scenarios would be as realistic and applicable as possible. The full scenarios in Dutch can be found in Appendix A.
Table 1. 2 x 2 between subjects factorial design.
Type of change Framing of change
As opportunity Scenario 1: Non-threatening non-identity related change; N=61
Scenario 3: Non-threatening identi-ty related change; N= 63
As threat Scenario 2: Threatening non-identity related change; N=66
Scenario 4: Threatening identity related change; N=63
3.2. Procedure
Separate questionnaires with separate links were created for the accountants and the teachers. The content of the scenarios differed for the professions, but the measurement scales used were exactly the same. The online survey started with an introduction in which the purpose of the participating studies was explained. On the second page, the consent form was presented. Here, respondents were assured that the answers given would be treated anonymously and that all data gathered would be treated confidentially. Furthermore, it was explained that re-spondents could stop with the survey whenever they wanted, and contact information was provided so that respondents could reach the researchers for questions. In addition, respond-ents were informed that the results of the research could be acquired. To start the survey,
re-spondents had to agree with the consent form. Survey administration started on March 11th 2017 and ended on April 18th 2017.
The questionnaire started with a participation check, in which questions were asked about the respondent's work, such as tenure. After this, questions about the respondents’ work attitude and work identity were asked. Hereafter, one of the four scenarios was randomly as-signed to the respondent. After the scenario was read, respondents had to answer questions about the scenario shown, such as how the respondent felt about the scenario and how he or she would respond to this scenario when it would occur in real life. Finally, questions con-cerning gender and age were asked. Respondents could indicate whether they wanted to be informed about the results, wanted to be contacted for future research, or wanted to win a Bol.com voucher. If they checked one of these options, they had to provide their e-mail ad-dress.
3.3. Sample
A total of 344 respondents started the survey, but only those who completed the entire survey were included in the dataset. The final sample contained 253 respondents and consisted of 89 Dutch accountants and 164 Dutch teachers (73.5% response rate). This paper focused on teachers and accountants for multiple reasons. First of all, new innovations and technologies are often introduced and applied in the workplace of teachers (Zhang & Nunamaker, 2003) and accountants (Taipaleenmäki & Ikäheimo, 2013). Since respondents with these professions often encounter technological changes in real life, the technological changes shown in the scenarios might reflect their personal experiences and feel as recognizable and real, which will increase the realism of the vignette study and thus benefit the external validity (Finch, 1987). Furthermore, teachers and accountants are often related to strong work identities (e.g. Beijaard, Meijer & Verloop, 2004; Brouard, Bujaki, Durocher, & Neilson, 2017), and there-fore, these professions are well suited for research that focuses on identity changes and identi-ty rigidness. The questionnaire in the online survey contained scales for the individual theses of 7 MSc students, and the data was collected collectively as a group with each student being responsible for the collection of data from 25 to 35 teachers or accountants.
The sampling technique used in this study was a non-probability convenience sample: student’s recruited respondents via their own network. Respondents were contacted through e-mails and social network websites such as Facebook and LinkedIn. In addition, different schools and accountancy companies were contacted to ask if they could spread the survey among employees. To maximize the number of respondents, respondents were able to win a
Bol.com voucher worth €100,00 when they completed the entire survey (one voucher for the teachers and one for the accountants). The researchers used an online document to list com-panies, teachers and accountants from their own network. In this document, notes were made with information about the persons and companies, when and how these persons were ap-proached to fill in the survey, and whether reminders to fill in the survey had been sent to these persons.
The final sample contained 253 respondents and consisted of 89 Dutch accountants and 164 Dutch teachers. The age of teachers ranged from 21 to 67 years, with an average of 43.45 years (SD = 13.33). For accountants, the age ranged from 21 to 63 years, with an aver-age of 34.22 years (SD = 11.93). 35.6% of the teachers were men, and 64.4% were female (M = .64, SD = .48), while 66.7% of the accountants were men, and 33.3% were female (M = .33,
SD = .47). The amount of time a person had been teacher or accountant was measured as
ten-ure. For teachers, tenure ranged from 9 months to 44 years and 8 months (M = 17.01, SD = 12.59). For accountants, tenure ranged from 1 month to 37 years and 8 months (M = 11.20,
SD = 9.75). In the total dataset (N = 253), the average age of respondents was 40.21 years (SD
= 13.56). 46.3% of the respondents were men and 53.7% were female (M = .54, SD = .50). Tenure ranged from 1 month to 44 years and 8 months (M = 14.96, SD = 11.97).
3.4. Materials
Apart from the questions concerning age, gender, working hours and tenure, all constructs of this research were measured with existing scales. Since the survey was administered in Dutch, the scales had to be translated from English to Dutch. This was done with independent back-translations, which means items were translated from the original English version of the scales into Dutch, and back to English by a second translator to check if the re-translations matched the original English scales (Brislin, 1980). In addition to the translation of the scales, the context of the scales was adjusted to match the scenarios. The original versions and trans-lations of the scales can be found in appendix B.
3.4.1. Independent variable
Type of change. The independent variable of this study is the type of change. The
ma-nipulation of the type of change was incorporated into the four scenarios. Each scenario start-ed with a brief text including background information. This text was the same for all scenarios of the teachers and all the scenarios of the accountants. For teachers, the background infor-mation stated that the rapid technological advances of recent years made the use of
technolo-gy in everyday life more important. Using technolotechnolo-gy would become crucial for the future of students. The scenarios for accountants started with background information about the rapid technological advances of recent years that have made it possible for organizations to obtain more data and make better analyses for their decision-making. As a result, organizations are using more and more Big Data to make their decision-making evidence-based and data-driven.
In the actual manipulation of the type of change, the change could either be identity related or non-identity related. In the scenarios with the non-identity related type of change, the scenarios stated that new software in class was introduced for teachers and new Big Data tools were introduced for accountants. The essence of their work would not change, but the existing tasks could be performed using these new tools. This means that the essence of their work role would not change. In the scenarios with the identity related type of change, teachers would have intensive cooperation with IT companies to develop new software solutions and accountants would have intensive cooperation with IT companies to develop new Big Data tools. Their new responsibilities would become a part of their existing work role, meaning the essence of their work would change drastically.
3.4.2. Dependent variables
Resistance to change. Three items measured resistance to change (Berkers, Mol &
Den Hartog, 2017). Items included questions such as ‘I would look for another employer in this scenario’, and were rated on a five-point scale from 1 (completely disagree) to 5 (com-pletely agree). Cronbach’s alpha for this scale was .67 (M = 1.91, SD = .72), which can be interpreted as acceptable (Loewenthal, 2004). The corrected item-total correlations indicated that all 3 items had a good correlation with the total score of the scale (all above .30). Fur-thermore, none of the items would substantially improve the reliability if they were to be de-leted.
Willingness to change. Three items measured willingness to change (Berkers et al.,
2017). Items included questions such as ‘I would be willing to help my colleagues in this sce-nario’, and were rated on a five-point scale from 1 (completely disagree) to 5 (completely agree). Cronbach’s alpha for this scale was .63 (M = 3.90, SD = .66), which is acceptable (Loewenthal, 2004). In the corrected item-total correlations, all 3 items scored above .30, in-dicating a good correlation with the total score of the scale. Deleting one of the items would not substantially improve the reliability.
Work identity rigidity. Eleven items measured work identity rigidity (Berkers et al.,
2017). Items included questions such as ‘I am who I am at work, even when the situation re-quires change’. Items were rated on a seven-point scale from 1 (completely disagree) to 7 (completely agree). First, items 7 to 10 were reverse coded. A reliability analysis showed that Cronbach’s Alpha for the work identity rigidity scale would be substantially improved if items 1, 7, 8, 9 and 10 were deleted (see Appendix B for all 11 items). After deleting these items, Cronbach’s Alpha increased from .79 to a high reliability of .86 (M = 4.13, SD = 1.19).
A principal axis factoring analysis (PAF) was conducted on the scale to see if the scale included only 1 factor. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for analysis, KMO = .88. Bartlett’s test of sphericity χ2 (15) = 616.31, p = .000, indicated that correlations between items were sufficiently large for PAF. Next, an initial analysis was run to obtain eigenvalues for each component in the data. Only one component had an eigenvalue over Kaiser’s criterion of 1, and explained 51.51% of the variance, meaning that the work identity rigidity scale had only 1 factor.
Computer self-efficacy. Four items measured computer self-efficacy (Gallivan et al.,
2005). Respondents were asked to rate their level of agreement with these four items using a five-point scale (1 = strongly disagree; 5 = strongly agree). Cronbach’s alpha for this scale was .58, which means reliability was questionable. By deleting the item ‘I could complete the job using the software if there was no one around to tell me what to do as I go along’, Cronbach’s Alpha improved to a reliable .76 (M = 3.73, SD = .77).
3.4.3. Control variables
Gender, age and tenure were added to this study as control variables. This was done to control for the possibility of a relationship between these variables and the independent variable, or an effect of the control variables on the dependent variables.
Gender. According to research, there might be differences for gender (male = 0,
fe-male = 1) in relation to resistance to change. Adams and Funk (2012) argue that for men, so-cial status, prestige and power are more important than for women. When change occurs, this usually goes together with changes in distribution of power, affecting an individual’s social status. Therefore, it can be suggested that men might have more resistance towards change than women.
Age. It can be argued that age (measured in years) could have a negative relationship
with change. Younger people tend to have a more flexible attitude towards technological change than older people (Aubert, Caroli, & Roger, 2006; Cordery, Sevastos, Mueller &
Par-ker; 1993). In addition, older employees often have more fear that changes in their job will result in a loss of privileges and status (Yeatts, Folts, & Knapp, 2000). The findings suggest that younger people will adapt more easily to change and show less resistance.
Tenure. Research has found that tenure (measured in years and months) might impact
the level of adaption to change. Fossum, Arvey, Paradise, and Robbins (1986) found that for employees with long tenure, it is more difficult to adapt to change, because they have to devi-ate from their successful established routines and need to conform to the new ways of work-ing. This finding supports the assumption that there could be a positive relationship between job tenure and resistance to change.
3.5. Data Analyses
The analysis of the data was done with IBM’s predictive analytics-software, SPSS Statistics
22. First, the dataset was inspected for missing values by checking the frequencies and
de-scriptives of the variables used in this study. Eight respondents did not give their age and six respondents did not give their gender. The missing values for these variables were not changed, since they had no further influence on the research. For tenure, one respondent an-swered that his tenure was more than 59 years, while he was only 58 years old. Therefore, the data for tenure of this respondent was removed. Four respondents did not answer the items for computer self-efficacy. These respondents did fill in the rest of the questions. Because this research attempted to maintain as much data as possible, these respondents have not been removed from the dataset. Thus, no respondents were removed from the dataset on the basis of missing values.
Secondly, counter-indicative items for the work identity rigidity scale were recoded, and scale reliabilities where measured for all variables. After that, descriptive statistics, skew-ness, kurtosis and normality checks were done. None of the variables were normally distribut-ed. The majority of the respondents reported relatively low levels of resistance to change, and relatively high levels of willingness to change, computer self-efficacy and work identity rigid-ity. Since the research model of this study contains two dependent variables and two interac-tion variables, multivariate analyses of variance (MANOVA) and multivariate analyses of covariance (MANCOVA) were conducted, because these tests allow you to test multiple de-pendent variables at once. Both MANOVA’s and MANCOVA’s were conducted to see if the effects of the independent variable and interaction variables with the dependent variables would differ if control variables were added to the analyses.
4. Results 4.1. Descriptive Analysis
The means, standard deviations, reliabilities, and correlations between variables are reported in Table 2. As the table indicates, willingness to change shows a moderate negative correla-tion with resistance to change (r = -.46, p < .001), which is not surprising because resistance to change is the opposite of willingness to change. For work identity rigidity, there is a small positive correlation with resistance to change (r = .13, p < .05), and a small negative tion with willingness to change (r = -.18, p < .01). The table shows a small positive correla-tion between computer self-efficacy with work identity rigidity (r = .20, p < .01). Gender has a small positive correlation with work identity rigidity (r = .17, p < .01), and computer self-efficacy (r = .13, p < .01). There is a high positive correlation between tenure and age (r = .82, p < .01), because the older someone is, the longer his or her work career is likely to be. Apart from the correlation between tenure and age, all correlations are smaller than .50, indi-cating moderate or small correlations (Cohen, 1992). A variance inflation factor (VIF) analy-sis was carried out to measure if multicollinearity was present between type of change, work identity rigidity, and computer self-efficacy. All obtained VIF’s were smaller than 1.10, which means none of the predictors were correlated with each other and that multi-collinearity does not exist here (Mansfield & Helms, 1982).
Table 2. Correlations between type of change, resistance to change, willingness to change,
work identity, computer self-efficacy, and control variables gender, age and tenure.
Correlations
N M SD 1 2 3 4 5 6 7 8
1 Type of change 253 .5 .5 -
2 Resistance to change 253 2.74 .73 .01 (.67) 3 Willingness to change 253 3.90 .66 .00 -.46** (.63) 4 Work identity rigidity 253 4.13 1.19 .02 .13* -.18** (.86) 5 Computer self-efficacy 249 3.73 .77 .08 .05 .07 .20** (.76)
6 Gender 244 .54 .50 .10 .06 -.03 .17** .13* -
7 Age 245 40.21 13.57 -.03 .06 -.11 .05 .01 .00 -
8 Tenure 252 14.96 11.97 .02 .07 -.11 .03 .01 .09 .82** -
Note: Cronbach’s Alpha’s are reported along the diagonal for each scale. * Correlation is significant at the .05 level (two-tailed).
4.2. Regression Analyses
This study investigates the relationship of technological identity change with resistance to change and willingness to change, and the influence of computer self-efficacy and work iden-tity rigidity as moderators in this relationship. The first hypothesis stated that technological identity change was positively related to resistance to change and the second hypothesis stated that technological identity change was negatively related tot willingness to change. To test both hypotheses, a MANOVA was conducted with type of change as independent variable, and resistance to change and willingness to change as dependent variables. The MANOVA revealed no statistically significant main effect for the type of change, Wilks’ λ = 1, F (2, 250) = .04, p = .966, partial η² = .000, which means that no differences were found for resistance to change and willingness to change on the basis of the type of change. An additional MAN-COVA was conducted with gender, age and tenure as covariates, but did not reveal a signifi-cant main effect either, Wilks’ λ = 1, F (2, 236) = .03, p = .967, partial η² = .000. Hypothesis 1 and hypothesis 2 are rejected.
Hypothesis 3a stated that high computer self-efficacy would weaken the positive rela-tionship of technological identity change with resistance to change. Hypothesis 3b stated that high computer self-efficacy would weaken the negative relationship of technological identity change with willingness to change. To test both hypotheses, a MANOVA was conducted with type of change as independent variable, computer self-efficacy as interaction variable, and resistance to change and willingness to change as dependent variables. There was no signifi-cant main effect for type of change, Wilks' λ = .99, F (2, 223) = 1.66, p = .193, partial η² = .015, nor a significant main effect for computer self-efficacy, Wilks' λ = .89, F (24, 446) = 1.13, p = .306, partial η² = .057. No significant interaction between type of change and com-puter self-efficacy was found either, Wilks' λ = .90, F (22, 446) = 1.08, p = .360, partial η² = .051. An additional MANCOVA was conducted with gender, age and tenure as covariates, but did not show significant changes in multivariate main effects. However, inspection of the univariate main effects revealed that type of change had a marginal significant relation with resistance to change when age, gender and tenure were controlled in the analysis, F (1, 242) = 3.43, p = .065 partial η² = .016. Nevertheless, from the results pertaining to hypothesis 3a and 3b, it cannot be concluded that high-self efficacy weakens the positive relationship of techno-logical identity change with resistance and weakens the negative relationship of technotechno-logical identity change with willingness to change. Thus, hypotheses 3a and 3b are rejected.
Hypothesis 4a stated that when work identity rigidity is high, the positive relationship of technological identity change with resistance to change would be stronger. Hypothesis 4b
stated that when work identity rigidity is high, the negative relationship of technological iden-tity change with willingness to change would also be stronger. To test both hypotheses, a MANOVA was conducted with type of change as independent variable, work identity rigidity as interaction variable, and resistance to change and willingness to change as dependent vari-ables. There was no significant main effect for type of change, Wilks' λ = 1, F (2, 195) = 1.66,
p = .843, partial η² = .002, and no main effect for computer self-efficacy, Wilks' λ = .72, F
(60, 390) = 1.18, p = .186, partial η² = .153. The multivariate analysis did not reveal an inter-action between type of change and computer self-efficacy either, Wilks' λ = .78, F (50, 390) = 1.05, p = .390, partial η² = .119). A MANCOVA with gender, age and tenure as covariates did not show significant changes of the multivariate main effects. From the results pertaining to hypothesis 4a and 4b, it cannot be concluded that work identity rigidity strengthens the positive relationship of technological identity change with resistance and strengthens the neg-ative relationship of technological identity change with willingness to change. Therefore, hy-potheses 4a and 4b are rejected.
5. Discussion 5.1. Findings
This study investigated how technological identity change was related to employees’ re-sistance to the change and willingness to change, and investigated the moderating effect of computer self-efficacy and work identity rigidity in the relationship of technological identity change with resistance and willingness to change. To investigate these relations, the following research question was formulated: “How do computer self-efficacy and work identity rigidity
influence the relationship of technological identity change with resistance to change and will-ingness to change?”. An experimental vignette study concerning the introduction of
techno-logical changes in the workplace was conducted among 253 Dutch accountants and teachers. Results did not reveal a relation of technological identity change with resistance to change and willingness to change. Furthermore, no moderating effect of computer self-efficacy and work identity rigidity was found in the relation of technological identity change with re-sistance to change and willingness to change.
Opposing to the expectations of this paper, the results did not reveal that employees that encountered identity related changes had more resistance to change or less willingness to change than employees that encountered non-identity related changes. A possible explanation for this can be given based on the argumentation of the chosen outcome variables of this study. This research focused on resistance to change and willingness to change as intentional
outcome variables, while the argumentation in the theoretical framework was based mainly on attitudinal outcome variables. Although, according to Ajzen (1991), the attitude predicts the intention, it is not the only predictor. The theory of reasoned action (Fishbein & Ajzen, 2010) explains that the formation of intention is also determined by the perceived norm, which is the social pressure an individual may feel to perform the behavior, and the perceived behavioral control, which is the belief that an individual is able to perform the behavior. It is therefore possible that the relationship between technological identity change and resistance to change offers different results than when intentional variables are used as outcome variables in this relationship.
Furthermore, this research did not find that high computer self-efficacy weakens the relationships of technological identity change with resistance to change and willingness to change. On the basis of Hobfoll’s (1989) conservation of resources theory, it was expected that for individuals with high computer self-efficacy, technological changes would be easier to cope with and cause less stress, because individuals with high computer self-efficacy have more resources (Chen et al., 2009; Hobfoll, 1989). Although self-efficacy can be seen as a strengthening resource (Bandura, 1979), this paper did not find that computer self-efficacy acted as an interaction effect. Since this research did not find a relationship of technological identity change with resistance to change and willingness to change in the first place, it is difficult to make assumptions about the possible interaction effect of computer self-efficacy. Nevertheless, a remark can be made regarding the operationalization of computer self-efficacy in this study (Gallivan, Spitler & Koufaris, 2005). In the scale used, respondents were asked to rate their level of agreement with questions about working with software when they had seen others use it before them or when they had enough time to use the software. Howev-er, the scenarios in the experimental vignette study did not explain or show what kind of software respondents would use; they only explained whether the software would supplement the work role of the employee or that the software would fully be integrated within the exist-ing work role. Since the scenarios and measurement scale for computer self-efficacy did not completely match, it can be argued that it was difficult for respondents to indicate how easy or hard it would be to use the software.
Lastly, this research did not find that rigid work identity strengthens the relationships of technological identity change with resistance to change and willingness to change. As with computer self-efficacy, the expected moderating role of work identity rigidity was explained on the basis of Hobfoll’s (1989) COR model. Individuals with a rigid work identity would see changes as a loss of resources, and would therefore try to maintain their resources by not
ad-justing their identity (Briscoe & Hall, 1999). However, work identity rigidity did not reveal to have a strengthening moderating role in this study. Since this research did not find a relation-ship of technological identity change with resistance and willingness to change in the first place, it is difficult to make assumptions about a possible interaction effect of work identity rigidity. Although no assumptions can be made about work identity rigidity as a moderator, something interesting can be noted about the relationship between the two moderators used in this study. Inspection of the normal distributions of these variables revealed that the majority of respondents reported both high levels of computer self-efficacy and high levels of work identity rigidity. Furthermore, a positive correlation was found between the two variables. This finding is interesting, because this paper initially argued that the variables would be op-posites: high computer self-efficacy suggests that people are open to adaptations and have the belief they can cope with technology (Gallivan, Spitler & Koufaris, 2005), while work identi-ty rigidiidenti-ty, on the other hand, states that individuals are unwilling to adjust their identities (Briscoe & Hall, 1999; Cardador & Caza, 2012). This finding implies that computer self-efficacy may have more similarities with work identity rigidity than was initially assumed.
5.2. Limitations and Recommendations for Future Research
Although this research was carefully prepared, a number of acknowledgements must be made. This study used an experimental vignette study, which makes it possible to investigate mental processes in a controlled environment and can provide valuable insight into work behaviors that are not easily observable (Aguinis & Bradley, 2014). However, a disadvantage of this method is that respondents are reading a scenario and do not really experience a real life situa-tion. Respondents are making decisions on the basis of hypothetical situations, and they might not make the same decisions in real life. A recommendation for future research would be to work with real life cases in a real work environment. However, this method is expensive and time consuming.
Second, this paper used an online survey to conduct the experimental vignette study. An online survey offers convenience for respondents because it allows them to answer the questions at their own pace and it might make respondents more willing to share personal information since they are not facing another person directly. A disadvantage of online sur-veys is that respondents might be afraid that the answers given in the questionnaire will be made public, which could lead them to filling in socially desirable answers. However, this paper attempted to avoid socially desirable answers by explaining in the consent form that the
respondent’s answers would be treated anonymously and confidentially. Also, the necessity that respondents filled out the survey truthfully was communicated here.
A third limitation is related to the sample used in this study. This study chose teachers and accountants as professions to investigate, since these professions are often introduced to new innovations and technologies (Zhang & Nunamaker, 2003; Taipaleenmäki & Ikäheimo, 2013). However, it can be argued that the introduction of new technologies takes place not only in the workplace of teachers and accountants, but also in many other professions. By investigating multiple other professions in future research a broader understanding can be achieved of how technological identity changes at work are related to employee reactions in different professions. This will also make it possible to compare results between professions.
5.3. Conclusion and Practical Implications
This paper examined the concept of technological identity change, the extant to which a tech-nological change affects the employee’s identity. This paper is the first study to investigate the concept of technological identity change. Although this research did not find a relation of technological identity change with resistance to change and willingness to change, and no moderating effect of computer self-efficacy and work identity rigidity, it is still important to further investigate the concept of technological identity change. Highly competitive markets and rapid technological innovations are pushing organizations to constantly adapt and change. Organizations make changes to improve their performance. However, everything starts at the level of the individual employee, and if the employee cannot or does not want to handle the change, the change does not make sense. Since work is an essential part of the employee’s identity, changes in the work mean changes in the identity. The interplay of work and ties leads employees to reorganize and restructure their personal identities and social identi-ties. To date, little is known about the reactions this interplay could cause. Therefore, more knowledge of technological identity change and its outcomes is necessary to understand how employees experience new technologies in the workplace and what the psychological impact of technological identity changes is for individuals.
6. Reference List
Adams, R.B., & Funk, P. (2012). Beyond the glass ceiling: Does gender matter? Management
Science, 58(2), 219-235. doi:10.2307/41406385
Aguinis, H., & Bradley, K.J. (2014). Best practice recommendations for designing and im-plementing experimental vignette methodology studies. Organizational Research
Methods, 17(4), 351-371. doi:10.1177/1094428114547952.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Deci-sion Processes, 50(2), 179-211.
Alvesson, M., & Willmott, H. (2002). Identity regulation as organizational control: Producing the appropriate individual. Journal of Management Studies, 39(5), 619–644. doi:10.1111/1467-6486.00305
Amiot, C.E., de la Sablonnière, R., Terry, D.J., & Smith, J.R. (2007). Integration of social identities in the self: Toward a cognitive-developmental model. Personality and Social
Psychology Review, 11(4), 364-388. doi:10.1177/1088868307304091
Armenakis, A.A., & Harris, S.G. (2009). Reflections: Our journey in organizational change research. Journal of Change Management, 9(2), 127-142.
doi:10.1080/14697010902879079
Ashforth, B.E., Harrison, S.H., & Corley, K.G. (2008). Identification in organizations: An examination of four fundamental questions. Journal of Management, 34(3), 325-374. doi:10.1177/0149206308316059
Atzmüller, C., & Steiner, P.M. (2010). Experimental vignette studies in survey research. Methodology. European Journal of Research Methods for the Behavioral and Social
Sciences, 6(3), 128-138. doi:10.1027/1614-2241/a000014
Aubert, P., Caroli, E., & Roger, M. (2006). New technologies, organisation and age: Firm-level evidence. The Economic Journal, 116(509), F73-F93. doi:10.1111/j.1468-0297.2006.01065.x
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychogi-cal Review, 84(2), 191-215. doi:10.1037/0033-295X.84.2.191
Beijaard, D., Meijer, P.C., & Verloop, N. (2004). Reconsidering research on teachers’ profes-sional identity. Teaching and Teacher Education, 20(2), 107–128. doi:10.1016/j.tate.2003.07.001
Berkers, H.A., Mol, S.T., & Den Hartog, D.N. (2017). When changing who you are at work
feels unbearable: Development and validation of the work identity rigidity scale.
Block, J. (1961). Ego identity, role variability, and adjustment. Journal of Consulting
Psy-chology, 25(5), 392–397. doi:10.1037/h0042979
Bordia, P., Hobman, E., Jones, E., Gallois, C., & Callan, V.J. (2004). Uncertainty during or-ganizational change: Types, consequences, and management strategies. Journal of
Business and Psychology, 18(4), 507-532. doi:10.1023/B:JOBU.0000028449.99127.f7
Brewer, M.B. (1991). The social self: On being the same and different at the same time.
Per-sonality and Social Psychology Bulletin, 17(5), 475-482.
doi:10.1177/0146167291175001
Brewer, M.B. (2012). Optimal distinctiveness theory: Its history and development. In P.A.A. Van Lange, A.W. Kruglanski, & E.T. Higgins (Eds.), The Handbook of Theories of
Social Psychology, Vol. 2 (pp. 81-98). London: Sage.
Briscoe, J.P., & Hall, D.T. (1999). An alternative approach and new guidelines for practice.
Organizational Dynamics, 28(2), 37–51. doi:10.1016/S0090-2616(00)80015-7
Brislin, R.W. (1980) Translation and content analysis of oral and written material. In H.C. Triandis, & J.W. Berry (Eds.), Handbook of Cross-Cultural Psychology (pp. 389-444). Boston: Allyn & Bacon.
Brouard, F., Bujaki, M., Durocher, S., & Neilson, L. (2017). Professional accountants' identi-ty formation: An integrative framework. Journal of Business Ethics, 142(2), 225-238. doi:10.1007/s10551-016-3157-z
Brown, A.D. (2015). Identities and identity work in organizations. International Journal of
Management Reviews, 17(1), 20-40. doi:10.5465/amr.2013.0103
Burnes, B. (2011). Introduction: Why does change fail, and what can we do about it? Journal
of Change Management, 11(4), 445-450. doi:10.1080/14697017.2011.630507.
Burnes, B., & Jackson, P. (2011). Success and failure in organizational change: An explora-tion of the role of values. Journal of Change Management, 11(2), 133-162. doi:10.1080/14697017.2010.524655
Cardador, M.T., & Caza, B.B. (2012). Relational and identity perspectives on healthy versus unhealthy pursuit of callings. Journal of Career Assessment, 20(3), 338-353. doi:10.1177/1069072711436162
Chen, S., Westman, M., & Eden, D. (2009). Impact of enhanced resources on anticipatory stress and adjustment to new information technology: A field-experimental test of conservation of resources theory. Journal of Occupational Health Psychology, 14(3), 219-230. doi:10.1037/a0015282.
