Dear Future Self
How to become the person you desire to be
MASTER THESIS
MSc Business Administration: Leadership & Management By Anna-Lena Grimm
Student number: 11370963
Thesis supervisor: Thesis group: Date:
Hannah Berkers and Dr. Stefan Mol Work Identity
Statement of originality
This document is written by Anna-Lena Grimm, who declares to take full responsibility for the content of this document.
I declare that the text and the work presented in this document is original and that no sources other than those 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.
No one likes to be criticized, of course, but if the things we successfully strive for do not make our future selves happy, or if the things we unsuccessfully avoid do, then it seems reasonable (if somewhat ungracious) for them to cast a disparaging glance backward and wonder what the hell we were thinking.
Abstract
The current job market is increasingly characterized by change and uncertainty. These career changes can lead to changes in an individual’s self-definition – ‘work identity change’ (WIC) – as work plays an important role in defining one’s identity. Little research has been done yet on how imposed WIC influences who a person wants to be – the possible self. This study aimed to open up the black box between WIC and the possible self. Specifically, it was examined to what extent gender changes the way ‘need for social validation’ (NSV) and ‘impression management’ influence ‘confidence in attaining the desired possible self’ (CPS) in a situation of WIC. As the possible self is a cognitive construct that is difficult to measure, this study focused on confidence as antecedent of the possible self to draw conclusions on how to attain the desired possible self. Previous literature has emphasized the importance of NSV and impression management in the identification process, which is why the two concepts were explored to open the black box between WIC and CPS. As gender also influences the identification process, it was expected that the mediated relationship of NSV and impression management with CPS is moderated by gender. An experimental vignette study and an additional questionnaire were conducted among 253 accountants and teachers. A regression and process analysis with a between-subject design, controlling for WIC, revealed that WIC and CPS are indeed related, however only among teachers and not accountants. No support for a mediation effect of NSV and impression management, nor for these effects being conditional upon gender differences were found. The different findings between teachers and accountants give new insights into the importance of work context in constructing one’s (future) identity. It remains important to further investigate the antecedents of the possible self and open the black box between WIC and CPS. Findings might give new insights on how to realize one’s desired possible self and become the person one desires to be.
Key Words: Work identity change, confidence in attaining the desired possible self,
Table of Content
Introduction 6
Literature Review 10
Confidence in Attaining the Desired Possible Self (CPS) 10
Work Identity Change (WIC) 12
Need for social validation (NSV) 13
Impression Management 15 Gender 16 Method 17 Procedure 17 Sample 19 Measures 19 Analysis strategy 21 Results 23 Hypothesis testing 24 Hypothesis 1 24 Hypothesis 2 24 Hypothesis 3 24 Hypothesis 4 25 Discussion 27 Theoretical implications 31 Practical implications 32
Strength and Limitations 33
Future research 34
Conclusion 36
References 37
Appendix A - Measurements 45
Appendix B - Assumptions for regression analysis 48
Appendix C - Reliability analysis 51
Index of Figures and Tables
Index of Figures
Figure 1 Graphic overview of the research model
Index of Tables
Table 1 Demographics of the research sample
Table 2 Means, Standard deviations, and Correlations of the research variables Table 3a Findings process analysis – teachers
Table 3b Findings process analysis – accountants
Table 4 Kurtosis analysis for the dependent variable ‘Confidence’, mediator 1 ‘Need for Social Validation (NSV)’, and mediator 2 ‘Impression Management’
Table 5 Skewness analysis analysis for the dependent variable ‘Confidence’, mediator 1 ‘Need for Social Validation (NSV)’, and mediator 2 ‘Impression Management’ Table 6 Cronbach’s alpha for the dependent variable ‘Confidence’, mediator 1 ‘Need for
Social Validation (NSV)’, and mediator 2 ‘Impression Management’ Table 7 Principal Axis Factor Analysis for the variable ‘Confidence’
Introduction
Today, the job market is characterised by quick changes and steadily rising demands for high performance (e.g. Lyons, Schweitzer, & Ng, 2015; Storey, 2000). Employees consistently have to enhance their skills and abilities to stay competitive (Eby, Butts, & Lockwood, 2003). To achieve high job performance, employees need to identify with their job (Wayne, Randel, & Stevens) and develop themselves constantly (Kuncel & Hezlett, 2004). This drive for improvement might be based on a person’s image of who one would like to be, the possible self (Markus & Nurius, 1986). The possible self can be defined as the representation of one’s future identity. It is divided in a desired possible self and a feared possible self. The concept has been introduced as expansion of earlier identity concepts that only defined who we are right now (Markus & Nurius, 1986). Who we are and who we want to be is highly influenced by someone’s job, as people spend most of their lifetime at work (Lloyd, Roodt, & Odendaal, 2011). Work identity can be defined as the meaning added to the self, based on personal characteristics, group membership, and (social) roles (Miscenko & Day, 2015). If work changes, it brings uncertainty as well as new possibilities for the self (Ashforth, Schinoff, & Rogers, 2016). Work identity change (WIC) can be defined as changes in an individual’s self-definition based on career changes (Chreim, Williams, & Hinings, 2007; Miscenko & Day, 2015; Ibarra, 1999). WIC occurs more frequently in today’s dynamic work environment, which is why it is important to study the increasing possibilities for defining the self (Chreim et al., 2007).
The process of defining one’s identity is called identity construction (Ashforth & Schinoff, 2016). Identity construction often refers to identity work (Pratt, Rockmann & Kaufmann, 2006), so the two terms will be used synonymously in this research. Two identity construction processes are need for social validation (NSV) and impression management (e.g. Ashforth & Schinoff, 2016, Pratt et al., 2006), which might explain how to attain the desired possible self under a situation of change. NSV can be defined as the feedback one obtains to
get from the social environment (Ashforth & Schinoff, 2016). It is well known that NSV influences the current identity, however, it is unclear how NSV influences the desired possible self (Ashforth & Schinoff, 2016). Impression management can be defined as presenting oneself better than one actually is (Rioux & Penner, 2001). During identity change, people often run the risk of misleading others in order to become who they desire to be. Both identity construction processes, might occur in a situation of WIC in order to construct one’s desired self. The likelihood of one process occurring over the other might be determined by someone’s gender, as earlier research reveals significant gender differences in constructing one’s identity (e.g. Singh & Vinnicombe, 2001; Singh, Kumra & Vinnicombe, 2002; Kling, Hyde, Showers & Buswell, 1999). For example, it has been found that women consider NSV as more important, whereas men show more impression management. Thus, it is unclear, whether the relation of WIC and attaining the desired possible self, runs via NSV or impression management, and whether gender influences the usage of both or either of the two identity construction processes.
The possible self has gained increasing attention in identity research (Packard & Conway, 2006). However, research on how to realise the desired possible self is still missing as it is a cognitive, intangible concept that is difficult to measure (Packard & Conway, 2006). The most popular method in assessing the possible self are structured surveys that assess self-perceptions from participants such as confidence in achieving the possible self (Packard & Conway, 2006). Also other researchers reveal a high correlation between confidence in one’s abilities and the possible self (Buday, Stake, & Peterson, 2012). Therefore, this research focuses on confidence in attaining the desired possible self (CPS) as the outcome variable. Confidence is considered as a motivational force behind attaining the desired possible self in order to give more insight into realizing who one desires to be.
The current study contributes to extant literature in several ways. Current literature on WIC and the possible self is limited, as the possible self is a cognitive, intangible construct,
which is difficult to measure (Packard & Conway, 2006). This issue will be addressed by studying confidence as an underlying construct of the possible self. In this regard, conclusions about realizing the desired possible self can still be made. By focusing on underlying identity work concepts, NSV and impression management, it is possible to open up the black box and explain the process between WIC and CPS. In this regard, integral knowledge about antecedents of the possible self can be achieved. New insights can contribute to gain a better understanding how the possible self is realized under change. Further, gender differences between these two paths will be addressed, to deepen understanding of different outcomes of realizing the desired possible self for men and women. Finally, current researchers mostly focused on influences of macro changes, changing between jobs, instead of micro changes, changes within jobs (e.g. Ashforth, 2001). Studying the influence of micro changes on the possible self can show how smaller changes influence the self. This might be of high importance, as e.g. role transitions occur more often than macro transitions (Dukerich, n.d.).
Besides, individuals and organizations may stand to benefit from more insight into the realization of the desired possible self and its antecedents. It is important to give insights on how to realize the desired possible self under WIC, as change requires individuals to actively (re)construct their identity (Ashforth et al., 2016). This reconstructing process might be experienced as difficult, which makes it important to support individuals in reaching their desired self. Coherent knowledge about the role of NSV and impression management in the relation between CPS and WIC might allow individuals to cope with WIC and realize their desired possible self. Furthermore, consistent insights about attaining the desired possible self can ensure more confident and capable employees, who use the full range of their capabilities. As work identity influences employee behaviour, and consequently organizational outcomes (Amiot, de la Sablonnière, Terry, & Smith, 2000), it is also important that organizations support their employees in attaining their desired possible self. As micro work changes are increasingly occurring due to an era of digitization, which includes Big Data and new IT
systems (Bijl, 2009), organizations need more insight in how these changes influence their employees and finally organizational and financial outcomes (Amiot et al., 2000).
Based on previous argumentation, the research question of the current endeavour is: To what extent does gender change the way ‘need for social validation’ (NSV) or ‘impression management’ influence ‘confidence in attaining the desired possible self’ (CPS) in a situation of ‘work identity change’ (WIC)? Because of the two identity work perspectives, NSV and impression management, and concomitant gender differences, it is important to give a clear answer in studying the relationship between WIC and CPS. The current research is structured as follows. First of all, a theoretical background of all variables and the research hypotheses are given. Second, the research design is elaborated and the method and analyses are described. Third, hypotheses are tested and results are specified. Finally, all findings and limitations are discussed and suggestions for future research are given.
Literature Review
Confidence in Attaining the Desired Possible Self (CPS)
The concept of the possible self has been introduced as past concepts about identity ignored how the future is represented in the self-concept (Markus & Nurius, 1986). Identity can be defined as a person’s self-definition, which consists of the different roles an individual fulfils (Bothma, Lloyd, & Khapova, 2015). People are constantly trying to answer the question ’Who am I?’, and actively construct their identity (Pratt et al., 2006). The possible self can be defined as the representation of one’s future identity. It includes one’s desired self, who one wants to be, and one’s feared possible self, who one does not want to be. It contains hopes and dreams, as well as anxieties and fears (Ashforth & Schinoff, 2016). Current research increasingly focuses on identity, not as a one entity, but consisting of many possibilities. It includes one’s current behaviour as well as future dreams (Ibarra, 2004). Still, it is unclear what happens to the possible self under changing circumstances and how to turn a possible self into reality. It is important to expand the literature on the possible self with these findings, as change might enable the realization of one’s dreams and might help one become who one desires to be.
An important reason for the gaps within the current literature is the difficulty to measure the possible self, as it is a cognitive, intangible concept that relies on self-schemas and self-reports (Packard & Conway, 2006). Earlier research has attempted to gain more insights into one’s future identity by advancing the concept of the possible self and introducing the provisional self (Ibarra, 1999). The provisional self can be defined as a temporary self that functions as trial for the possible self. It helps to bridge the gap between one’s current and future representation of the self. To adopt a provisional self, individuals actively observe role models, experiment with provisional selves and evaluate their behaviour. The notion of provisional selves is however more a repertoire of possible selves. It
remains unclear when a provisional self is a successful trial of the possible self and allows turning the possible self into reality.
The current research suggests that the realization of the possible self might depend on other internal processes within an individual, such as confidence in one’s abilities. Confidence can be defined as the belief in one’s abilities (Perrewé, et al., 2004). It has been shown that the desired possible self is highly correlated with confidence in one’s abilities (Buday et al, 2012). Therefore, confidence is considered as underlying process of realizing the possible self, and the outcome variable in this research is confidence in attaining the desired
possible self (CPS). Confidence is considered to be a motivational force behind attaining the
desired possible self in order to gain more insights in realizing the desired possible self. Studying the possible self is highly important due to two main reasons. First, it is important to have extensive knowledge about the possible self as it influences people’s future behaviour (Markus & Nurius, 1986). Specifically, knowledge about the future self can act as an incentive to realize a possible self. It might result in people working towards their possible self, and showing proactive behaviour. More knowledge about the possible self can provide guidance in attaining the desired possible self and help in avoiding their feared possible self. Second, the possible self serves as reference for interpreting the current self. Thinking about the possible self can function as a benchmark. Knowing who one wants to become allows to compare the possible self to the current self. Thus, knowing one’s possible self can help to gain more knowledge about one’s current identity (Markus & Nurius, 1986), which is a prerequisite for employees to function well and to fit into the organization they work for (Ashforth, 2016). As the possible self can act as a benchmark for the current self, it shows the importance and interrelation between the current and future identity. Finally, in this study, it is assumed that more knowledge, and respectively confidence in who one is right now and who one wants to be, can help individuals to become the person they desire to be.
Work Identity Change (WIC)
Knowing one’s identity, who one is and who one wants to be, has become increasingly important, as our society and job market are characterized by dynamic and constant change (Lyons et al., 2015). The great influence of work on identity can be explained by the fact that work is often a central aspect in people’s life and determines to a great extent who a person defines to be (Berkman, 2014). Work identity is defined as the meaning added to the self, based on personal characteristics, group membership, and (social) roles (Miscenko & Day, 2015). For employees, a strong work identity can provide dignity, purpose to life and a feeling of caring for oneself and one’s family (Berkman, 2014).
As work plays an important role in defining one’s identity, career and role changes are decisive factors in constructing one’s work identity (Pratt et al., 2006). It has been found that change at work often leads to new identity roles, which can change a person’s self-definition (Pratt et al., 2006). If an employee attains a different job role, his or her sub-identities are more or less affected depending on how central the role is for the individual. Thus, change at work leads to change in how one defines oneself, resulting in work identity change (WIC) (Pratt et al., 2006). Here, WIC is defined as changes in an individuals’ self-definition based on career changes (Chreim et al., 2007; Miscenko & Day, 2015; Ibarra, 1999). It can be described as personal adjustments emerging from job changes (Nicholson, 1984). WIC reflects that identity and change are constantly interacting (Ashforth & Saks, 1995). It also highlights the power of change at work, as work change disrupts a person’s self-image, which requires an individual to actively construct his or her identity (Pratt, 2006). As identity is made up of many possibilities (Ibarra, 2004), rearranging one’s identity after WIC might allow an individual to realize his or possible self.
The reconfiguring of one’s identity might be experienced as opportunity, as well as a threat (Ashforth et al., 2016). Indeed, it has been found that identification under change invokes helpful and harmful effects (Ashforth et al., 2016). Individuals who perceive change
as opportunity can increase their self-enhancement and are more willing to try and explore provisional selves (Ibarra, 1999). However, individuals who perceive change as threatening try to defend their identity and internalize other’s weaknesses (Ashforth et al., 2016). The findings can be explained by the influence of disrupted knowledge of the self. Career changes often lead to unfamiliar situations, where people identify with new role models, try out behaviour and construct possible identities (Ashforth & Saks, 1995). Thus, the findings show that change can have positive outcomes, and at the same time, they warn of possible negative effects on an individual’s identity. The reaction to change might be even stronger for the possible self than the current self, as the possible self is a cognitive construct, which is less dependent on social reality and highly dynamic (Markus & Nurius, 1986).
For identity construction and CPS, it is important to achieve a positive outcome of the self under change. The expectancy-value model by Carver and Scheiver (1999) states that progressing towards a goal can already increase someone’s confidence to succeed. Even more, it seems that only imagining a positive outcome can already increase confidence in success (Meevissen, Peters, & Alberts, 2011). The findings suggest that having the desired possible self in mind might be a first step in increasing CPS. Based on previous argumentation, the first hypothesis is:
H1: There is a positive relationship between WIC and CPS.
Need for social validation (NSV)
As change disrupts a person’s self-image, people have to conduct identity work to redefine their self-image (Pratt et al., 2006). In Ashforth and Schinoff’s (2016) identity construction circle, a model showing how individuals construct their identities, social validation is seen as important part in order to fortify one’s identity. Need for social validation is defined as the desire to get positive feedback from others (Ashforth, 2001). It
reveals the importance for individuals to get confirmation from their environment, about who they are, and that what they do is right. Social validation signals the acceptance of someone’s identity, which can strengthen one’s identity formation process. In organizations, employees that obtain social validation feel more accepted and identify more with the organization (Gioia & Chittipeddi, 1991). This entails important outcomes, as a committed workforce results in a lower turnover rate, and decreased employee costs (Smith et al., 2013).
Previous research states that NSV influences who one is right now, the real self (Asforth & Schinoff, 2016). For example, it has been shown that only identities that are positively evaluated by others turn into one’s self-definition (Stryker & Serpe, 1982). These findings reveal the importance of NSV as it strongly influences a person’s identity and behaviour. However, little research has been conducted on the relation between NSV and one’s future identity (Ashforth & Schinoff, 2016).
It might be that NSV influences the possible self via self-verification (Swann et al., 2003). Self-verification can be defined as a person’s drive to gain confirmation about one’s being (Swann, 1983). This need for confirmation about who one is often makes individuals uncertain when others think better of them then they actually are. Hence, NSV in terms of self-verification might reduce one’s confidence and diminish the realization of the possible self. This can be supported by the concept of social comparison, where uncertainty leads to comparing oneself with higher-status individuals (Brown, Ferris, Heller, & Keeping, 2007), which can lead to a negative self-view (Wheeler, 2000). A negative self-view however hinders individuals in imagining their desired possible self (Markus & Nurius, 1986). Thus, WIC accompanied by uncertainty might diminish CPS and hinder people in becoming who they desire to be. This results to the following hypothesis:
H2: The positive relationship between WIC and CPS is mediated by NSV (H2), whereby WIC and NSV are positively related (H2a), however NSV and CPS are negatively related (H2b).
Impression Management
Alongside to NSV, another identity construction processes called ‘impression management’ might explain how to attain one’s desired identity under WIC. Impression management can be defined as a person’s behaviour in influencing the impression of others about his/her identity (e.g. Meister, Jehn, & Thatcher, 2014; Rosenfeld, Giacalone, & Riordan, 1995). Individuals often use impression management to strengthen their self-image, reduce negative emotions, and/or construct and maintain one’s identity. High self-esteem individuals use favourable and acquisitive impression management, and low self-esteem individuals try to avoid failure and show protective impression management (Baumeister, Tice, & Hutton, 1989). In organizations, employees use impression management to be perceived as respectable and competent, thereby enhancing their status and their chances of getting a better job (Kim, 2008). For example, job seekers often use impression management to augment their self-confidence (Knouse, 1994).
Earlier research on impression management has shown how impression management is used in identity construction (e.g. Meister, et al., 2004). For example, Ibarra and Barbulescu (2010) looked at how rhetorical impression management, such as story telling, is used to construct one’s identity images (Ibarra & Barbulescu, 2010). People use narratives to communicate their provisional selves and manage others impressions to create the social environment needed for the desired possible self (Schlenker, 2003). The findings show the importance of rhetorical identity work for the provisional self to manage impressions. As the provisional self is a trial of the possible self (Schlenker, 2003), impression management might likewise help to attain one’s desired possible self. This results to the following hypothesis:
H3: The positive relationship between WIC and CPS is mediated by Impression Management (H3), whereby both WIC and Impression Management (H3a), and Impression Management and CPS (H3b) are positively related.
Gender
Whether NSV or impression management mediate the relation of WIC and CPS depends on gender. Earlier research shows that women seem to have less confidence in their overall abilities and deal differently with identification processes than men (Singh et al., 2002). For example, it seems that men have a more independent self-image, whereas women base their self-esteem on their relationships. This suggests that there might be differences in gender regarding the influence of others on one’s self image. Indeed, it has been found that women react more strongly than men to feedback due to lower self-esteem in their self-image (Kling et al., 1999). These findings show the impact that feedback, and accordingly NSV, can have on one’s identity - especially for women.
In impression management research, it has been found that women and men use impression management actively to signal high commitment (Singh et al., 2002). Both are aware of potential benefits of using impression management at work, such as an elevated perception of their competences. However, it seems that men use more work-related impression management strategies than women (Singh & Vinnicombe, 2001). Women prefer to focus on commitment to show their abilities, whereas men focus on self-promotion. The reason provided for this is that female employees feel more uncomfortable than men about aggrandizing their self-image. This can be explained by theories on counter-stereotypical impression management, which posit that, in the case of women, impression management comes at the cost of social attraction (Rudman, 1998). Women have to deal with negative consequences of both impression management and social attraction. However, social attraction might be of more importance, as women depend more on their social surrounding to create their identity (Blichfeldt & Gram, 2017). Research on women’s identity work states that for women even single social events have meaning for their identity work. These findings show that there are gender differences in identification and organizational processes, which
gives reason to study the influence of gender under the current endeavour. Based on previous argumentation the fourth hypothesis is:
H4: Gender moderates the relationship of NSV and Impression Management with CPS (H4) so that the negative relationship of NSV with CPS is stronger for women than for men (H4a) and the positive relationship of Impression Management with CPS is stronger for men than for women (H4b).
The prior established four sets of hypotheses are illustrated in the research model in Figure 1.
Figure 1: Graphic overview of the research model. WIC is the independent variable; CPS is the dependent
variable; NSV and impression management are the mediating variables; gender is the moderator.
Method
Procedure
To study CPS, a quantitative research was conducted using a cross-sectional survey design, including an experimental vignette study and a questionnaire. The experimental vignette asked participants to imagine a described situation and respond respectively, to
H1(+)
H3b(+) H2a(+)
Work identity change (WIC)
Confidence in attaining the desired possible self
(CPS)
Impression management Need for social validation (NSV) Gender H2 H3a(+) H4b(+) H4a(-) H4 H2b(-)
manipulate the independent variable WIC. By combining a traditional survey with a vignette experiment, WIC was measured actively by the controlled manipulation and internal validity could be increased (Atzemüller & Steiner, 2010). Furthermore, the scenarios described situations that fitted to the work context of the sample to ensure contextual realism and increase external validity (Raaijmakers, Vermeulen, Meeus, & Zietsma, 2015). In this regard, experimental vignettes allowed for the measurement of the influence of the independent variable, WIC, and for drawing conclusions about the hypothesized causal relations with the other research variables (Aguinis & Bradley, 2014). As WIC was stimulated hypothetically instead of changing someone’s identity in real life, the scenarios also took ethical considerations into account (Hughes & Huby, 2004). The questionnaire was used for manipulation checks and additional measures of the different variables.
A non-probability convenience and snowball sample was used due to the difficulty to reach participants outside personal contacts. A pilot study (N = 80) with a 3x2 between-subject design with five scenarios was carried out. The scenarios described a work situation under change, which influenced the core of the person’s job. Conditions of the five scenarios were the kind of change (no change, non-identity related / identity related) and the way the change was framed (threatening / non-threatening). As the no change condition related strongly with the non-identity change condition, the final design was adjusted to a 2x2 between-subject design with four scenarios (non-identity related / identity related x threatening / non-threatening). All items were originally in English however as a Dutch sample was used, a back-forward translation method has been performed to ensure better understanding and increase internal validity. One person carefully translated items into Dutch, while another person back-translated the items into English. By comparing the re-translation to the original item, it was ensured that the Dutch translation kept its original meaning. Finally, two surveys were conducted, to fit the scenarios to the sample of two professions (teachers and accountants), and distributed on Qualtrics.com. Within participant’s profession,
people were automatically and randomly assigned to one of the scenarios. Instructions were provided, declaring that participation was anonymous and voluntary with the chance of winning a 100€ voucher. In the end, participants were asked to fill in their demographics.
Sample
The sample consisted of both accountants and teachers. These two professions were chosen as first, accountants and teachers often experience a strong identification with their work (e.g. Bejaard, Meijer, & Verloop, 2004; Hamilton, 2013). To measure WIC, a preceding strong work identity was important so that work change had severe influences on someone’s work identity. Second, both professions are exposed to many internal work changes, e.g. Big Data and new IT systems (e.g. Chiu, 2017; Fineberg, 2012). Internal work changes are important for this study, as it is expected that WIC is caused by work changes, influencing someone’s identity (Miscenko & Day, 2015; Ibarra, 1999). Originally, 253 teachers and accountants participated in this research altogether, however, as not all participants filled in the item ‘gender’ (N = 9), which was a moderator in this research, these participants were taken out of further analysis resulting in data of 160 teachers and 84 accountants. Demographics of the two samples are presented in Table 1.
Table 1
Demographics of the research sample
Teachers Accountants
N 160 84
Gender Male = 57, female = 103 Male = 56, female = 28
Age M = 43.45, sd = 13.30 M = 33.85, sd = 11.79
Tenure M = 17.27, sd = 12.98 M = 10.93, sd = 9.65
Hours per week M = 29.91, sd = 10.39 M = 39.41, sd = 5.59
N = 244 (teachers and accountants); M = mean; sd = standard deviation
Measures
Independent variable: Work identity change (WIC). An experimental vignette study with four scenarios was used, including two scenarios that indicated identity-related change
and two scenarios that indicated non-identity related change. In the identity-related change condition, a work situation was described that asked employees to acquire a new role in advising and collaborating with IT developers about software or Big data tools. It was stated that the described change also alters the core of one’s job. By asking participants to empathize with the situation, controlled manipulation of WIC was achieved. In the non-identity related condition, it was stated that only the current work method is changing, while the work itself remaines the same. Hereby change was characterized by the usage of new Big Data Tools in the office or more software in classrooms. The scenarios were designed by Hannah Berkers, PhD student at the University of Amsterdam, and validated by professional accountants and teachers.
Dependent variable: Confidence in attaining the desired possible self (CPS). Based on Perrewé, et al. (2003), originally 9-item 5-point Likert scale (1=strongly disagree to 5 = strongly agree) with Cronbach’s alpha α = .72, five items were selected to shorten the overall questionnaire and avoid the risk of a lower response rate. The selection of items was based on suitability of the context of the current research. Item two, three and five are reversed items. An example question was “I have confidence in my ability to do my job.“. One reversed item was for example “There are some tasks required in my job that I can not do well.“
Mediator: Need for social validation (NSV). Based on Smith et al.’s (2013) 6-item 7-point Likert scale (1 = strongly disagree to 7 = strongly agree), with Cronbach’s alpha α = .97, five items were selected to shorten the questionnaire. Again, items were selected based on their suitability for use of the items in the context of the current research. One question was for example, “My leader makes me feel that my opinions about the correct way to do my job are valid.“.
Mediator: Impression management. Based on Bolino and Turnley’s (1999), originally 22-item 5-point Likert scale (1 = totally disagree to 5 = totally agree) including five subscales, the subscale “Exemplification” with 5-items was chosen as it was considered to fit best to the
context of this study (Cronbach’s alpha α = .75). One item was taken out, as it did not fit to the work field of teachers. One question was, for example, ” Try to appear like a hard-working, dedicated employee.“.
Moderator: Gender. Based on a three-item scale, participants were asked to indicate whether they consider themselves to be male, female or ‘other’.
For a detailed description of all research measurements, please see Appendix A.
Analysis strategy
All analyses were conducted by using SPSS version 21. The study variables were recoded, data was cleaned and the assumptions required for regression analysis were examined (normality check, linearity and homoscedasticity) (Appendix B). The data did not include any outliers as the value of the standardised scores was z > ⎮3⎮. Participants who did not fill in the item “gender” (N = 9) were taken out of the analysis, as gender is a moderator in this research. For participants (max. N = 6), who did not complete other items of the questionnaires a single imputation method was applied by calculating mean scores across participants. As only few participants had missing values, there is a minor risk that this method artificially decreased the variation of scores, which might have changed the value of correlation (Field, 2013). Further, computing mean scores for missing data allowed for using the whole data set and produced internally consistent sets of results (Field, 2013). Additionally, a reliability analysis has been done for all scales with a threshold of an acceptable Cronbach’s alpha criterion of α >.7 (Field, 2013). Results of the reliability analysis and a manipulation check can be found in Appendix (C) and (D). For further analysis, an overall score was calculated for each variable by computing scale means.
Age was chosen as control variable, as research shows that there is a relation between age and identity in that individuals are categorizing their identity to categories such as age (Ashforth & Schinoff, 2016). Moreover, research revealed that age leads to more self-esteem
and confidence (Lovatt, 2011). By controlling for participants age, the aim was to highlight if any further statistical findings are potentially subject to a participant’s age (Field, 2013). To test the first hypothesis, a regression analysis was conducted. For further hypothesis, a moderated mediation analysis with Process Macro was conducted, including a Bootstrap analysis (Hayes, 2013). The indirect effect of WIC (X) on CPS (Y) through NSV (M1) and impression management (M2) with gender (V), as second stage moderator, was analysed by using “Model 14”. To make outcomes more objective, analysis was run with and without age as control variable. Without age, all results were not significant. The following results include the controlled relationship with age.
Results
Table 2 presents the descriptive statistics, correlations and alpha coefficients of all variables.
Table 2
Means, Standard deviations, Correlations, and reliabilities of the research variables
Variables M SD 1 2 3 4 5 6
1. Work identity change .50 .50 (.)
2. Confidence 3.72 .74 -.12* (.76)
3. Need for social validation 3.07 .87 .03 -.22** (.89)
4. Impression management 2.21 .66 -.04 .04 .26** (.67)
5. Gender 1.56 .50 .09 -.18** .04 .01 (.)
6. Age 40.13 13.59 -.02 -.20** -.20** -.36** .01 (.)
N = 244 (teachers and accountants). Numbers in brackets are alpha coefficients. *Correlation is significant at the .05 level (one-tailed)
** Correlation is significant at the .01 level (one-tailed)
As shown in Table 2, only the correlations between confidence and WIC (r = -.12, p <
.05), confidence and NSV (r = -.22, p < .01), NSV and impression management (r = .26, p < .01), and confidence and gender (r = -.18, p < .01), was significant, whereas the relationships
among the other variables was not. The non-correlation indicates that there was no linearity between these variables and these relationships should be considered as weak throughout further analysis. As linearity is one assumption of regression analysis, the following results have to be taken with caution. The variables confidence, NSV, impression management and gender seem to be unrelated to the independent variable WIC, by which prediction of these variables based on their relationship with WIC is not possible. The control variable age was only related to confidence (r = -.20, p < .01), NSV (r = -.20, p < .01), and impression management (r = -.35, p < .01).
Hypothesis testing Hypothesis 1
Regarding the first hypothesis, a positive relationship between WIC and CPS was expected. For teachers, analysis revealed that WIC and CPS are significantly related (F(1,156) = 5.37, p = .02). However, for accountants, WIC and CPS were not significantly related (F(1,81) = .02, p = .89). Based on these outcomes, Hypothesis 1 is supported among teachers, however not among accountants.
Hypothesis 2
Further, in Hypothesis 2, it was expected that the positive relationship between WIC
and CPS is mediated by NSV (H2), whereby WIC and NSV are positively related (H2a), and NSV and CPS are negatively related (H2b). The analysis revealed that WIC was not indirectly
related to CPS via NSV neither for teachers nor for accountants. This is shown by the bootstrapped 95% Confidence Interval around the indirect effect, which ranged from -.10 to .02 for teachers, and -.17 to .08 for accountants. As both included zero, the mediating effect of NSV was not supported by the data (Hayes, 2013). Also the direct relation of WIC and NSV (H2a) was not significant for teachers (F(2,157) = .47, R2 = .01, b = .07, t(157) = .47, p = .64), neither for accountants (F(2,81) = 4.99, R2 = .11, b = -.02, t(81) = -.12, p = .90).
Further, the direct effect of NSV and CPS (H2b) was neither significant for teachers (b = -.03,
t(152) = -.13, p = .90) nor for accountants (b = -.47, t(76) = -1.59, p = .11). Based on these
outcomes, Hypothesis 2 is not supported for neither teachers nor accountants. The results are presented in Table 3a (teachers) and Table 3b (accountants).
Hypothesis 3
In Hypothesis 3, it was expected that the positive relationship between WIC and CPS
Management (H3a), and Impression Management and CPS (H3b) are positively related. The
analysis revealed that WIC was not indirectly related to CPS via impression management in neither teachers nor accountants. This is shown by the bootstrapped 95% Confidence Interval around the indirect effect, which ranged from -.13 to .02 for teachers, and from -.19 to .05 for accountants. As both included zero, no significant mediating effect of impression management could be found (Hayes, 2013). Also the direct relation of WIC and impression management (H3a) was not significant for teachers (F(2,157) = 10.73, R2 = .12, b = -.12, t(157) = -1.16, p = .24) nor for accountants (F(2,81) = 2.76, R2 = .06, b = .05, t(81) = .42, p
= .67). Further, the direct relation of impression management and CPS (H3b) was not
significant for teachers (b = .18, t(152) = .53, p = .59), nor accountants (b = .39, t(76) = -1.16, p = .25). Based on these outcomes, Hypothesis 3 is not supported for neither teachers nor accountants. The results are presented in Table 3a (teachers) and Table 3b (accountants).
Hypothesis 4
Finally, it was expected that Gender moderates the relationship of NSV and
Impression Management with CPS (H4) so that the relationship of NSV with CPS is stronger for women than for men (H4a) and the relationship of Impression Management with CPS is stronger for men than for women (H4b). The analysis revealed that the interaction term of
NSV and gender (H4a) was not significant for teachers (b = -.12, t(152) = -.85, p = .39) nor for accountants (b = .27, t(76) = 1.21, p = .23). There was no indirect effect of NSV and men as the bootstrapped 95% Confidence Interval included zero for both teachers (-.1038, .0270) and accountants (-.05, .09). There was also no indirect effect of NSV and women as the bootstrapped 95% Confidence Interval included zero for both teachers (-.10, .05) and accountants (-.12, .08). Further, the interaction term of impression management and gender was also not significant (H4b) for teachers (b = .13, t(152) = .70, p = .49), nor accountants (b
men as the bootstrapped 95% Confidence Interval included zero for both teachers (-.03, .09) and accountants (-.01, .06). There was also no indirect effect of impression management and women as the bootstrapped 95% Confidence Interval included zero for both teachers (-.07, .01) and accountants (-.18, .05). Based on these outcomes, Hypothesis 4 could is not supported for neither teachers nor accountants. The results are presented in Table 3a (teachers) and Table 3b (accountants).
Table 3a
Findings process analysis among teachers
Bootstrapping analysis df1 df2 R2 B t p 95% CI Lower - Upper WIC – NSV - CPS (H2) -.10 .02 WIC - NSV (H2a) 2 157 .01 .07 .47 .64 NSV - CPS (H2b) -.03 -.13 .90 WIC – IM - CPS (H3) -.13 .02 WIC - IM (H3a) 2 157 .12 -.12 -1.16 .24 IM - CPS (H3b) -.18 -.53 .59 NSV*Gender - CPS (H4a) -.12 -.85 .39 NSV-men -10 .02 NSV-women -.10 .05 IM*Gender - CPS (H4b) .13 .70 .48 IM-men -.03 .09 IM-women -.07 .01
N = 164; WIC: Work identity change; CPS: Confidence in attaining the desired possible self; NSV: Need for social validation; IM: impression management.
* p < .05. ** p < .01.
Table 3b
Findings process analysis among accountants
Bootstrapping analysis df1 df2 R2 B t p 95% CI Lower - Upper WIC – NSV - CPS (H2) -.17 .08 WIC - NSV (H2a) 2 81 .11 .02 .12 .90 NSV - CPS (H2b) -.47 -1.59 .11 WIC – IM - CPS (H3) -.19 .05 WIC - IM (H3a) 2 81 .06 .05 .42 .67 IM - CPS (H3b) .39 1.16 .25 NSV*Gender - CPS (H4a) .27 1.21 .23 NSV-men -.05 .09 NSV-women -.12 .08 IM*Gender - CPS (H4b) -.26 -1.06 .29 IM-men -.01 .06 IM-women -.18 .05
N = 84; WIC: Work identity change; CPS: Confidence in attaining the desired possible self; NSV: Need for social validation; IM: impression management.
Discussion
The purpose of this research was to obtain more insight into attaining the desired possible self under WIC. The following question was examined: To what extent does gender change the way NSV or impression management influence CPS under WIC? The analysis revealed that Hypothesis 1 could be confirmed, however, Hypothesis 1 could only be confirmed for teachers and not for accountants. None of the other hypotheses could be confirmed for both professions.
Hypothesis 1
It was hypothesised that there is a positive relationship between WIC and CPS. The analysis revealed significant results for Hypothesis 1 among teachers, however not among accountants. The significant results among teachers show that WIC is indeed related to CPS. As this was not the case for accountants, findings reveal a difference in CPS under a situation of change among the two professions. The different results between teachers and accountants might be explained on the following grounds.
First, earlier research found that individuals might experience a changing situation as either harmful or as opportunity (Ashforth et al., 2016). Perceiving change as opportunity can increase self-enhancement and can lead to more exploration of provisional selves (Ibarra, 1999). On the contrary, perceiving change as threatening leads to defending one’s identity and the internalization of other’s weaknesses (Ashforth et al., 2016). The different outcomes of CPS among teachers and accountants might be explained by a difference in experience of a changing situation. It might be that teachers see change as an opportunity for their identity, whereas accountants perceive change as threat to their identity. This might be explained by differences in characteristics of the two professions. A teacher’s job is often steadier, where individuals often stay at one school for a long duration of their career (Brown, 2015). On the other side, research shows that the accountancy profession is characterised by many more
changes, such as globalization, restructures, mergers and acquisitions (Sweeney & Quirin, 2009). Most accountants stay only for a certain amount of time at one company and change positions more frequently and not always voluntarily. This job uncertainty raises existential concerns (Berkman, 2014) rather than confidence, which might explain non-significant results for the relation between WIC and CPS among accountants.
In addition, teachers and accountants might differ in how much they focus on the possible self. For example, the earlier mentioned expectancy-value model by Carver and Scheiver (1999) emphasizes that having a goal in mind can already function as trigger to achieve that certain goal. However, research shows that especially accountants often have to work under considerable stress (Ozkan & Ozdevecioglu, 2013). They have to achieve certain results to stay competitive. The enormous workload for accountants requires accountants to strongly focus on their current job performance and likewise might hinder to focus on the construction of a possible self. Therefore, realizing the possible self might not be the most important and present goal in mind for accountants, which might explain the difference between teachers and accountants in CPS under a situation of change.
The different mind-set between accountants and teachers might also be related to the “Regulatory focus theory” by Higgins (1997). Regulatory focus theory describes how people self-regulate needs during goal pursuit by either a promotion focus or a prevention focus. Promotion focus can be defined as a focus to strive for advancement and accomplishment to gain desired end-stages. Prevention focus on the other side is a focus where people strive for responsible behaviour and avoid undesired end-states. It might be that accountants have a stronger prevention focus than teachers, as accountants are increasingly exposed to job insecurity and vulnerable to job loss due to for example economic shocks (Sweeney & Quirin, 2009). The different characteristics of the two professions, along with a different mind-set, might explain the differences in outcomes between teachers and accountants of this research.
Hypothesis 2
Hypothesis 2 was not confirmed for both teachers and accountants in that NSV did not mediate the relation between WIC and CPS. The findings are in contrast with previous research, which suggests that NSV strengthens the identification process (Ashforth, 2001). For example, Ashforth and Schinoff (2016) claim that in the identity construction circle, social validation helps individuals to construct their identity. This study was focused on the need for social validation however, what has not been studied is actually receiving social validation. To cope with the uncertainty of WIC and increasing one’s confidence in becoming who one desires to be, it might be necessary to actually receive confirmation from others (Smith, et al., 2013). Individuals who are unsure about their identity and need social validation might actually not be able to increase their level of confidence when they do not receive it. This might explain why NSV did not mediate the relation of WIC and CPS in this study.
Second, earlier research explained the influence of NSV on the self by comparing the self with others (Brown, Ferris, Heller, & Keeping, 2007). Thereby, upward comparison often diminishes confidence in one’s abilities. It might be that teachers as well as accountants do not compare themselves to others as much as other professions. Teachers are mostly working by themselves, instead of in a team (Desurmont, Forsthuber & Oberheidt, 2008). Also for accountants job autonomy and freedom is an important characteristic to perform well (Chu & Lai, 2011). The strong focus on independence in both professions might lead to fewer opportunities for accountants and teachers to compare their work identity to those of others. Again, this might explain why NSV did not mediate the relation between WIC and CPS among teachers and accountants.
Hypothesis 3
Also Hypothesis 3 has not been confirmed for neither teachers nor accountants in that impression management did not mediate the relation between WIC and CPS. Again, this is contrary to what earlier research has suggested. Schlenker (2003) emphasized that impression management does often take the form of narratives, where people communicate their future identity and manage other’s impressions. Thereby, individuals can use narratives to construct their identity (Ibarra & Barbulescu, 2010). However impression management can take many different strategies next to narratives, such as integration, where people show flattery behaviour (Bolino & Turnley, 1999). To identify how WIC relates to CPS, it might be important to study a broader range of impression management strategies.
Second, research has shown that there is a difference in using impression management between high self-esteem individuals and low self-esteem individuals (Baumeister et al., 1989). It seems that high self-esteem individuals use favourable and acquisitive impression management, and low self-esteem individuals try to avoid failure and show protective impression management. To increase one’s confidence level in order to attain the desired possible self, it might be more important to show favourable, acquisitive impression management techniques to be able to proactively work on one’s confidence and attain one’s desired possible self. Acquisitive impression management might be highly important in times of WIC, as research on job crafting shows that proactive behaviour is important in order to conform to the desired possible self (Kira & Balkin 2014, Wrzesniewski & Dutton 2001). As impression management also did not mediate the relation of WIC and CPS, the black box between these variables remains.
Hypothesis 4
The last hypothesis has been that Gender moderates the relationship of NSV and
stronger for women than for men (H4a) and the positive relationship of Impression Management with CPS is stronger for men than for women (H4b). Also this hypothesis could
not be confirmed in that gender did not moderate the relationship of NSV and impression management with CPS. This is contrary to earlier research, where women react more strongly to feedback compared to men. This can be explained by the relatively low self-esteem in womens’ self images (Kling et al., 1999). However, as this study focused on the need for social validation instead of actually receiving social validation, it might be difficult to draw conclusions about differences in reaction to feedback between men and women and therefore on the moderating effect of gender on NSV with CPS. Regrading impression management, earlier findings by Singh and Vinnicombe (2001) claimed that men use more impression management than women in the workplace. This is contradictory to this study as there have been non-significant findings of a moderating effect of gender on impression management with CPS. The non-significant results might be explained by the specific type of impression management, measured in this study. As there are many different impression management strategies (Bolino & Turnley, 1999), it is important to further investigate possible gender differences in these strategies. Furthermore, Singh et al. (2002) claim that women in their thirties start to use impression management equally as men. As average age in this study was M = 43.45 for teachers and M = 33.85 for accountants, it might be that women already adjusted their usage of impression management to that of men.
Theoretical implications
The present findings have important theoretical implications for the literature on work identity and the possible self. Existing research of the possible self mostly focused on creating and testing possible selves (Ibarra, 2004). This research builds on previous findings and aimed to expand the literature by gaining more insight in how the possible self can actually be realized. It has been found that WIC leads to more CPS among teachers, while not among
accountants. With these findings, it has been shown how important surrounding and context are in the identity construction process and in realizing the desired possible self. This is in line with earlier research emphasising the influence of the environment on the self (e.g. Chreim, et al., 2007; Miscenko & Day, 2015; Ibarra, 1999). The environment is seen as constraint to individual agency. It supports the perspective of a more dynamic identity concept that adapts and changes with the context (Alvesson, 2010).
Further, the current research focused on the effects of micro changes, changes within jobs, whereas earlier research mostly focused on influences of macro changes, changes between jobs (e.g. Ashforth, 2001). As it has been shown that WIC influences the possible self among teachers, this indicates that small changes already influence the possible self. These findings are highly important as smaller changes occur more frequently, than macro changes (Dukerich, n.d.) and might therefore be of more influence on the identity formation process.
Still, the black box between WIC and CPS remains as no mediation effect of NVS and impression management with CPS, even as no effect of gender has been found. It might be that the focus on micro change gives an explanation for the fact that so few hypotheses have been supported. Although it has been shown that micro changes occur more frequently than macro changes (Dukerich, n.d.), macro changes have a stronger influence on one’s identity (Ashforth, 2001). With this research on WIC under micro changes, it has been shown how CPS is influenced among teachers. It might be possible that more severe macro changes are needed to show the relation of WIC and CPS also among other professions and to finally open up the black box between these variables.
Practical implications
The study results imply important practical implications for managers in organizations. First, organizations should be aware of the influence of a changing work
situation on the identity construction of employees. It has been found that WIC leads to CPS among teachers, which shows how WIC does not only influence the current identity (Ashforth, 2001), but also the future identity. Research has revealed that one’s future identity can function as motivational force to improve and develop oneself into a better self (Markus & Nurius, 1986). Therefore, consistent insights about attaining the desired possible self can help managers to build a high performing workforce.
Thereby, it is important that managers frame a changing situation as chance instead of threat to foster the realization of the desired possible self (Chreim, et al. 2007). Managers should support employees in coping with change and concomitant uncertainty to enable building the desired possible self. Consequently, organizations can facilitate employees in getting the most out of their potential. Supporting employees in their development contributes to organizational success, as work identity influences employee behaviour, and consequently organizational and financial outcomes (Amiot et al., 2000). The findings are important for organizations as change takes place frequently in the current working environment (Storey, 2000).
Strength and Limitations
An important strength of this study is the differentiation between teachers and accountants, resulting in relevant outcomes for both professions. Differences can be analysed, which shows the importance of context on identity construction (Chreim, et al., 2007; Miscenko & Day, 2015; Ibarra, 1999). However, the study also has some limitations. As only accountants and teachers have been researched, sampling bias might have occurred. It might be that the jobs under investigation are already saturated with change, as earlier research shows how frequently teachers and accountants experience change (Bejaard, Meijer, & Verloop, 2004; Hamilton, 2013). Therefore, general conclusions about other professions have to be made with caution. Still, by choosing more than one profession, it is expected that the
impact of this limitation will be reduced. Teachers and accountants are two different professions so that it can be differentiated and compared between distinct jobs.
Further, the method used in this research might be limited. By using an experimental vignette study, participants imagined a certain situation instead of experiencing it in reality. This might reduce external validity (Raaijmakers et al., 2015). Further, it might be possible that the manipulation did not succeed, as the scenarios did not reflect a real life situation as apposed to be. Still, an experimental vignette study has been conducted, as it is otherwise amenable to manipulate people’s identity ethically (Hughes & Huby, 2004). Furthermore, the use of self-reported questionnaires may lead to biases and social desirability in responses. In order to decrease this risk, it has been stressed that all answers are handled anonymously.
Future research
Future research is needed on how WIC relates to CPS as significant results for teachers, but not for accountants have been found. Further, as no mediation effect of NSV or impression management has been found, future research needs to further attempt to open up the black box between the WIC and CPS. As earlier research highlights the importance of social validation during the identity construction process (Ashforth & Schinoff, 2016), it is worthwhile to research social validation and its effect on CPS in greater depths. Likewise, it is important to gain more insights into the different strategies of impression management. Earlier research revealed the influence of narrative impression management strategies (Ibarra & Barbulescu, 2010). However, no mediating effect of impression management between WIC and CPS has been found in this research. Therefore, it is important to analyse these contradictory findings and differentiate between impression management strategies. Building upon knowledge about NSV and impression management can help to find out how WIC and CPS are related to each other.
macro changes. This research has focused on micro changes as not a lot of research has been undertaken on changes within a job (Ashforth, 2001). However, as most research hypotheses are not supported in this study, and macro changes are of stronger influence than micro changes (Ashforth, 2001), future research should investigate CPS and its relation with WIC also under macro changes.
In addition, earlier research described gender differences in the identification process by a relationship- vs. a more self-focused orientation (Singh et al., 2002). This differentiation relates to the difference between the two cultural dimensions collectivism and individualism by Hofstede (2001). Collectivism can be defined as highly integrated individuals that form a tight social network. Individualism can be defined as lose social structure, where people emphasize “I” over “We”. The definitions of collectivism and individualism highly coincide with the definition of a relationship – vs. a self-focused identity orientation. Therefore, future research might consider cultural influences on the relation between WIC and CPS. Cultural influences might be of particular importance as many changes in organizations occur due to globalization and an international workforce as well as trade relationships (Lyons et al., 2015).
Conclusion
The present study contributes to the literature on work identity and the possible self. It has been found that WIC leads to more CPS among teachers. Therefore, changing work situations accompanied by new opportunities are important in defining the self and realizing the possible self. No support for a mediation effect of NSV and impression management, nor for these effects being conditional upon gender differences were found. Therefore, the black box between WIC and CPS remains. As the possible self offers to become who one desires to be, and WIC is likely to occur more often in the work environment of today, which is increasingly characterized by uncertainty and change, it remains important to study the process of WIC and becoming who one wants to be – the desired possible self.
References
Agunius, H., & Bradley, K. J. (2014). Best practice recommendations for designing and implementing experimental vignette methodology studies. Organizational Research
Methods, 17(4), 351-371.
Alvesson, M. (2010). Self-doubters, strugglers, storytellers, surfers and others: Images of selfidentities in organization studies. Human Relations, 63(2), 193–217.
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, 364–388. Doi: 10.1177/10888 683073 04091 .
Ashforth, B. E. (2001). Role transitions in organizational life: an identity-based perspective. Mahwah, NJ: Lawrence Erlbaum Associates.
Ashforth, B. E., & Saks, A. M. (1995). Work-role transitions: A longitudinal examination of the Nicholson model. Journal of Occupational and Organizational Psychology, 68, 157- 175.
Ashforth, B. E., & Schinoff, B., S. (2016) Identity under construction: How individuals come to define themselves in organizations. The Annual Review of Organizational
Psychology and Organizational Behavior, 3(7), 1-27.
Ashforth, B. A., Schinoff, B. S., & Rogers, K. M. (2016). “I identify with her”, “I identify with him”: Unpacking the Dynamics of Personal Identification in
Organizations. Academy of Management Review, 41(1), 28-60.
Atzmüller, C. & Steiner, P. M. (2010). Experimental vignette studies in survey research.
Methodology, 6(3), 128-138. Doi: 10.1027/1614-2241/a000014
Baumeister, R. F., Tice, D. M., & Hutton, D. G. (1989). Self-presentational motivations and personality differences in self-esteem. Journal of Personality, 57, 547–579.
P.G.W. Jansen, G. Roodt (eds.): Conceptualising and Measuring Work Identity. Dordrecht: Springer Science + Business Media. Doi: 10.1007/978-94-017-9242-4_2 Beijaard, D., P. Meijer, P., N. Verloop, N. (2004). Reconsidering research on teachers’
professional identity. Teaching and Teacher Education, 20(2), 107–128.
Berkman, L. (2014). Commentary: The hidden and not so hidden benefits of work: identity, income and interaction. International Journal of Epidemiology, 43(5), 1517-1519. Doi: 10.1093/ije/dyu110
Bijl, D. (2009). Aan de slag met het nieuwe werken. Zeewolde: Par CC.
Blichfeldt, B. S., & Gram, M. (2017). Domestic hospitality, gender, and impression management among Danish women. Food and Foodways, 1-21. Doi:
10.1080/07409710.2017.1272294
Bolino, M. C., & Turnley, W. H. (1999). Measuring impression management in
organizations: A scale development based on the Jones and Pittman taxonomy.
Organizational Research Methods, 2, 187–206. Doi: 10.1177/109442819922005
Brown, E. (2015). Study: Far fewer new teachers are leaving the profession than previously thought. The Washington Post. Retrieved from
https://www.washingtonpost.com/news/local/wp/2015/04/30/study-new-teacher-attrition-is-lower-than-previously-thought/?utm_term=.213d552cf892
Brown, D. J., Ferris, D. L., Heller, D., & Keeping, L. M. (2007). Antecedents and
consequences of the frequency of upward and downward social comparisons at work.
Organizational Behavior and Human Decision Processes, 102, 59-75.
Buday, S. K., Stake, J. E. & Peterson, Z. D. (2011). Gender and the choice of science career: The impact of social support and possible selves. Sex Roles, 66, 197-209. Doi: 10.1007/s11199-011-0015-4
Carver, C. S., & Scheier, M. F. (1999). Optimism. In C. R. Snyder (Ed.), Coping: The psychology of what works (pp. 182e204). New York: Oxford University Press.
