Applying the Identity Status Paradigm to Managers’ Moral Identity

We investigated the applicability of the identity status paradigm in identifying different stages of moral identity maturity among managers, focusing on how they solve moral conflicts in the context of work. Researchers conducted two theory-driven studies. Study 1 was based on focus group discussions among 16 managers, while Study 2 was based on open-ended questionnaire items from 180 managers. Both studies supported the hypothesized identity statuses. The status named diffusion included a lack of commitment to moral values and associated with avoiding moral questions at work. In foreclosure, extrinsic (e.g., organizational) values were adopted and applied to personal decision-making. Managers in moratorium showed an active exploration of personal values, while those with achieved moral identity showed a commitment to self-chosen moral values. Study 1 showed that the participants applied different ethical principles when describing how they solved moral conflicts. Foreclosed managers tended to use utilitarianism, that is, to prioritize the benefits to their organization, when making ethical decisions. Managers in moratorium used ethics of care more often than others, and achieved managers relied on virtue ethics more than managers with other statuses. Our study makes a significant contribution to our understanding of the variation in moral identity maturity in adulthood. Understanding how value exploration and commitment relate to moral identity in the work domain can support more mature forms of moral identity among managers.


Introduction
Moral identity includes one's personal sense of morality and the degree to which being a moral person is important to one's identity (Aquino and Reed, 2002;Hardy and Carlo, 2011;Shao, Aquino and Freeman, 2008). A recent meta-analysis (Hertz and Krettenauer, 2016) found that moral identity positively predicts moral behavior, and it suggests that moral identity is the foundation of moral agency (Weaver, 2006). Therefore moral identity offers an important angle from which to try to understand and support managers' moral behavior at work.
However, there are two significant limitations in the field. First, even though it has been suggested that moral identity development extends into adult years (Bergman, 2002), thus far only one theoretical paper (Krettenauer and Hertz, 2015) and one empirical study (Krettenauer, Murua and Jia, 2016) have addressed this issue. We therefore have very limited understanding of the different developmental processes that relate to moral identity acquisition in adulthood. The second limitation is that the majority of previous studies have used student samples or scenario studies. The few studies that have been carried out using employee samples suggest that organizational factors have a significant influence on the employee's moral self (Jennings, Mitchell and Hannah, 2015), therefore more research needs to be conducted among adults actually working in organizations in order to better understand how moral identities might develop in the work context.
We contribute to moral identity research by investigating how managers experience and reflect their personal moral values when facing a range of moral questions at work. We focused on managers because of the paradoxical nature of their work: managers have positional power to influence their organization (e.g., through their personal moral decisions), but at the same time they have numerous commitments to organizational roles, with potentially conflicting expectations from upper managers, subordinates, and other interest groups which can limit their ability to act according to their moral identities (Weaver, 2006). We address the developmental processes of moral identity by testing whether the identity status paradigm (Marcia, 1966) can be applied to moral values. This approach takes into consideration that identity issues are negotiated through two main processes: the presence or absence of selfexploration (questioning and considering various identity alternatives) and of commitment (choices made in areas relevant to identity). Using these two dimensions, four identity statuses can be distinguished (Marcia, 1966): achievement (commitment present after a period of self-exploration), moratorium (commitment absent, selfexploration ongoing), foreclosure (commitment present with little or no personal self-exploration), and diffusion (neither commitment nor self-exploration are currently present). We used a multiple-study, mixed method design based on focus group discussions and questionnaires.
Although our two samples were both cross-sectional and we cannot therefore come to any conclusions about long-term development in moral identity, we contribute to understanding how the different maturity levels in moral identity are manifested among working adults. If we can find that individuals differ from each other in their moral identity maturity, we can assume that there is (developmental) variation between them. We can then assume that there are also differences in the underlying mechanisms that result in this variation in maturity. Value exploration and commitment are central processes in identity development, and therefore our findings will be the first step to understanding if and how these processes relate to the manifestation of moral identity in working adults in organizations (Jennings et al., 2015), an issue that has been given surprisingly little research attention recently.

Moral identity: Conceptual viewpoints
Moral identity is the degree to which moral ideals and actions, and being a moral person, are central to the individual (Hardy and Carlo, 2011). A person who feels that moral values such as honesty, fairness, compassion, and generosity are a central part of their self-concept has a strong moral identity. Individuals vary in how important these moral qualities are to them, and the degree to which these values are actualized in their choices and behavior (Jennings, Mitchell and Hannah, 2015). When an individual has an identity that is very much centered on morality, they are highly motivated to act morally. This is due to our tendency to self-consistency (Blasi, 1983), which makes us feel the desire to live according to our sense of self.
When we turn to the context of work, the concept of moral identity, and how to measure it, the teachings of Aquino and Reed (2002; the Self-Importance of Moral Identity Questionnaire, SMI-Q) has dominated the field of behavioral ethics (approximately 70% of the empirical research; Jennings, Mitchell and Hannah, 2015). According to this conceptualization, moral identity includes two dimensions: the first is internalization, which refers to the centrality of moral values, and the second is symbolization, which refers to the degree to which moral values are expressed publicly through individual actions. Research has shown that people with strongly moral identities are less likely to behave unethically at work (May, Chang and Shao, 2015) and less likely to adopt moral disengagement strategies (Detert, Treviño and Sweitzer, 2008). However, Aquino and Reed's (2002) model does not take into account how differences in internalization and symbolization are developed or whether and how a person could progress toward a stronger moral identity. This is a significant shortcoming, because development is at the core of the concept of identity (see, e.g., Bosma and Kunnen, 2001). Our study takes a step towards understanding the different levels of maturity in moral identity by applying the identity status paradigm to moral values at work.

The identity status paradigm
We used the ego-identity model (Marcia, 1966(Marcia, , 2007Kroger and Marcia, 2011) to capture the different levels of moral identity maturity among working adults. The model includes four different statuses that describe how individuals make personal choices in different life domains through the processes of exploration and commitment. Identity diffusion is the status in which one has neither explored important areas of life nor made commitments to them. Individuals in this status may or may not have experienced an identity crisis, they may have little interest in matters of identity, or they may have repeatedly experienced indecision regarding their personal values and roles. Foreclosure refers to the status of commitment without having explored the alternatives: values and beliefs are accepted and adopted from the outside (e.g., from parents) without being questioned. Moratorium is an on-going process of exploration in which the individual is in the midst of a crisis and their commitments are either negligible or only vaguely defined as they weigh up the different alternatives; this can elicit feelings of anxiety because the different identity-defining possibilities are left open and uncertain. Identity achievement is the status in which an individual has gone through an identity crisis (e.g., explored different value options), and has made a commitment to a certain role or value that he or she has chosen for him-or herself (Marcia, 1966).
As Lapsley and Hardy (2017) have recently noted, the developmental tracks of both morality and identity are ideally conjoined in an adult personality. However, the identity status paradigm has not previously been used to study the moral dimensions of identity development in adulthood (Lapsley and Hardy, 2017). Our aim is to investigate the applicability of these statuses among adults, especially in the work domain.

Moral identity statuses in the organizational context
According to the socio-cognitive model (Aquino et al., 2009;Narvaez et al., 2006), moral identity can include some context specificity because the accessibility of moral schemas can vary across social contexts. Particular areas of life (e.g., workplace or family) may have different demands that activate moral schemas in different ways (Krettenauer, Murua and Jia, 2016). In the moral decision-making that happens in organizations, several contextual factors might affect both behavior and the moral identity processes themselves. For example, Aquino and Freeman (2012) proposed that financial rewards at the workplace could trigger a business frame of mind, which may weaken the power of a strong moral identity and instead make salient a material identity. Weaver (2006) proposed that besides individual selfimportance and the salience of moral identity, there are several other organizational influences that take place at work, in interaction with others. For example, the social expectations and behavioral models that are available in the organizational context provide the individual with modeling and learning opportunities. It is therefore important to consider what kind of moral examples the organizational norms, rules, executives, supervisors, and colleagues provide. The organization as a context may not only promote certain moral schemas, but also influence the content of individual moral identity.
At the same time, there are some moral demands that individuals face across a range of situations and contexts, such as being honest, fair, and trustworthy to both family members and co-workers (Krettenauer, Murua and Jia, 2016), so moral identities are also likely to include crosscontext stability. However, even those who would like to apply the same moral values consistently, regardless of the area of life, they might find additional challenges in applying their moral values in ethical decision-making in the workplace, as these dilemmas can differ significantly from everyday situations (Crane and Matten, 2007). At work, too, different stakeholders can have several conflicting expectations, demands, and values that the individual must try to accommodate and reconcile (Donaldson and Dunfee, 1999). External demands such as time pressure can further intensify the challenging nature of these dilemmas, and individuals can find it very difficult to choose the right way to act. In order to get their work done, individuals might yield to external pressure and even compromise their personal values.
When applying the identity status paradigm (Marcia, 1966) to moral identity in the work context, it is important to consider how value exploration and commitment take place within this complex environment. We propose that in moral identity diffusion (which is characterized by the absence of both exploration and commitment) the individual lacks a coherent set of committed moral values, because they have not considered whether they experience moral values as a central part of their identity. It is possible that they have not experienced conflicts or other situations in which they have been led to reflect on their moral values in the work context. This can relate to low moral awareness (not recognizing the moral nature of situations; Butterfield, Treviño and Weaver, 2000;Reynolds, 2006), which has enabled them to make decisions at work without considering their personal moral values. As a consequence, they might avoid moral questions by playing down the ethical aspect of the situation (such as ethical fading, see Tenbrunsel and Messick, 2004), by trying not to confront moral dilemmas, or by resorting to moral disengagement (Moore et al., 2012).
In moral identity foreclosure (commitment present with little or no personal exploration), the individual has committed to extrinsic moral values that they have adopted without personal exploration or critical evaluation. When we focus especially on individuals who are moral actors in the organizational context, the values they have adopted can come from the organization, managers, colleagues, etc. This is in line with Weaver's (2006) notion about the models that the organizational context provides for moral identity construction. Because foreclosure includes a normative orientation, where the main concern is to conform to the prescriptions and expectations of others (Berzonsky, 1989), we suggest that when facing ethical dilemmas, individuals with this moral identity status will make decisions based on the values they have adopted from the work context. Therefore moral identity foreclosure can lead to adhering to organizational norms and demands without questioning them.
Because moratorium includes ongoing exploration without any present commitment (Marcia, 1966), we suggest that in moral identity moratorium the individual actively explores the personal values they would like to apply when facing moral issues at work. While diffusion is associated with low moral awareness and foreclosure with adhering unquestioningly to external values and norms, moratorium can associate with heightened moral awareness, and with criticizing current moral norms and rules in the organization or the personal values one has adopted before. It represents a developmental step towards a more mature moral identity, because the individual begins to explore for him-or herself those values that they would personally choose and want to commit to, thus indicating self-regulatory mechanisms and a more mature mode of psychosocial functioning (Bosma and Kunnen, 2001).
In achievement, the individual made value commitments after a period of personal exploration (Marcia, 1966). Achieved moral identity therefore represents the most mature status of moral identity development: going through an active developmental process of exploring and committing to certain values results in a central and salient moral identity. This identity status would be comparable to strong moral identity in Aquino and Reed's (2002) conceptualization, where moral values are an integrated part of the self (high centralization) and they are applied in personal actions (high symbolization).
Finally, it is important to consider what individual and/or social factors might affect moral identity development at work. Previous studies have shown that several personal factors, including religiosity, emotions, prior moral or immoral behavior, and demographics shape the moral self, and that the social context, such as role models (leaders and their characteristics), organizational characteristics, and workplace attitudes can either strengthen or weaken the moral self (for a review, see Jennings, Mitchell and Hannah, 2015). However, no previous studies have used identity statuses to represent individual morality at work, and so we lack empirical evidence as to what kinds of factors might predict belonging to different moral identity statuses. Our questionnaire study was cross-sectional and could not test how different antecedents affect moral identity development (e.g., predicting movement or stability between the statuses). However, we aimed to investigate whether and how personal and contextual factors differ between managers in each of the four statuses.

Research aims
Our first aim was to investigate how managers express their moral identity and how moral work identity relates to the ethical decision-making process in morally challenging situations. We explored the descriptions of moral dilemmas given by 16 managers who participated in focus group discussions (Study 1) and 180 managers who took part in a questionnaire study (Study 2), and used the moral identity statuses (diffusion, foreclosure, moratorium, and achieved) as a theoretical framework when analyzing the material. According to the logic of appropriateness (March 1994), the way individuals recognize, classify, and make decisions in a given situation is affected by their personal identity. Hiekkataipale and Lämsä (2017) used this framework to study the different strategies that middle managers use for handling ethical problems and found that managers who used different strategies also differed in terms of their identities. We analyzed the moral conflict stories in terms of how the managers talk about solving these dilemmas. As a secondary aim, in Study 2 we examined differences in individual (age, tenure, and gender) and contextual (work sector, organizational field, and manager's staff responsibility) background factors between the managers in different moral identity statuses.

Sample
The participants were 16 Finnish managers, who were participating in focus group discussions as a part of their executive MBA training (an educational program designed specifically for experienced business executives). Recruiting participants from the executive MBA course had two important advantages. First, using pre-existing groups reduces the risk of self-selection: all of the course participants took part in the data collection, not only those who might have been confident about presenting their views in a group discussion (see Cowton and Downs, 2015). Second, the course participants had already become acquainted with each other during the MBA training, and having a group discussion with familiar participants is likely to increase feelings of safety and confidence when talking about sensitive ethical issues.
Of the participants, nine were men and seven were women, and the mean age of the participants was 44 years (range 32-48). They were working in both the private and public sectors, and came from several different industries (including insurance, business services, public administration, commerce and trade, and education). They were working in middle management, and most of them were working as a supervisor.

Procedure
Focus groups have three main qualities (Denscombe, 2014): 1) focus on a particular topic of which all of the participants have some knowledge, 2) the interaction of the group is seen as a central way of acquiring information on that particular topic, and 3) the role of the researcher as the moderator of the discussion is limited to facilitating group interaction. Because the participants were in executive MBA training, they all had previous work experience as managers, which made a good starting point for sharing their authentic experiences of moral dilemmas at work.
During one executive MBA training session the participants were informed about the research project, after which they gave their written informed consent and began to participate in the study. The managers were assigned to four groups (the same groups that they had been in earlier in the training, so they could exchange their thoughts and ideas with people who were already familiar to them), and they were presented with the following question: 'Think about your work and its challenges from your own point of view. In what kind of situations do you have to think over the rightness of your actions before or after the situation?' The question was deliberately designed to be broad and general in order to generate different kinds of discussion about personal experiences on a range of moral dilemmas in the work context. The participants first wrote down their personal views on the given theme, and then discussed it with each other. This two-phase assignment was designed to give individuals the chance to think about their personal views before being affected by other people's ideas and viewpoints, and to generate new ideas during the group interaction. The format meant that each group had the opportunity to raise any kind of moral dilemma for discussion, without the researchers imposing any constraints on what moral issues they should focus on. The written answers were used only as personal prompts, to prepare the participants to take part in the group discussions; the participants kept them, and they were not used as study data.

Analysis
The discussions (mean duration 51 min.) were audio recorded and transcribed verbatim (66 pages with 12 pt. font and 1-line spacing) and the data were imported to the ATLAS.ti program (version 7.5.10; Friese, 2013) for further analysis. We used content analysis as our research technique (Krippendorff, 2012), reducing the text to meaningful categories with a theoretically driven approach in which we applied the identity status paradigm to the data.
The first author read and coded each of the four discussions. During this process, the researcher identified the main themes of the dilemmas that arose in each group's discussion. The next step was to identify individual participants' ways of describing their reflections, views, and justifications with regard to the dilemmas under discussion. At this point, the focus of analysis was on both whether the identity status paradigm was applicable to their experiences and how the managers described solving the different dilemmas.
Identifying the moral identity statuses was based on the managers' discussions, which were taken to reflect their thoughts and values when they faced and had to solve moral dilemmas at work. We aimed to identify those parts of the discussion that contained elements central to the four moral identity stages. That is, when the manager made a contribution to the discussion using elements that could be coded as reflecting 1) no personal moral reflection, 2) adopted extrinsic values, 3) ongoing value consideration, or 4) internalized personal values, these verbalizations were labeled using the four categories of diffusion, foreclosure, moratorium and achieved, respectively. Because of our deductive focus in this study, we did not code for other potential identity statements. Finally, the second researcher analyzed these preliminary codes and the two researchers then discussed any differences and agreed on the final coding. Altogether 116 text units were identified and coded: 3 units for diffusion, 46 for foreclosure, 47 for moratorium, and 21 for achieved (see Table 1).
Next, the authors read the data focusing on how the managers described solving the dilemmas. They found that the managers used different ethical principles when describing how they decided between right and wrong in a given moral situation. We identified four different orientations in the discussions (for a more thorough discussion of these ethical perspectives, see, e.g., Crane and Matten, 2007): the ethics of duty (making choices based on definite principles), virtue ethics (virtues are moral qualities that are manifested in practical actions), utilitarianism (decisions are based on their consequences), and ethics of care (highlighting kindness, empathy, and helping others who are in a vulnerable position). The first author began by identifying all the passages in the discussions that included content that was relevant to these principles. For example, some managers emphasized the possible consequences of their decisions for their employees (care ethics), while others emphasized the importance of achieving the greatest advantage for the organization (utilitarianism). The final coding was agreed on with the second researcher, who read the preliminary codes and discussed them with the first author. Altogether 132 units were identified, of which 27 were for virtue ethics, 28 for care ethics, 34 for utilitarianism, and 43 for duty ethics. For descriptive purposes, we examined whether the different principles of ethics co-occurred with the MWI statuses. The comparison was done with the χ 2 test, where all the ethical reasoning codes were cross tabulated with all the MWI status codes.

Diffusion
We were able to identify only three places in the discussions when the participants showed no commitment to or reflection on any moral values when coming to an ethical decision. The first of these was a statement to the effect that the speaker had not consciously reflected on moral issues before the focus group task: 'I was really astonished actually, about how many things I finally wrote down when I started to think about situations where I had to think about the rightness of my actions' (Group 1, female 1). Emphasizes personal ethical evaluation in the situation and sees morality as a central part of one's work-related decisions.

Achieved
Group 3 Deciding about lay-offs among the staff Male 3 Previously I used to say really quickly to a person whose work didn't go as it should: 'Goodbye, you can leave', but now I consider these decisions for ages.
Decision-making has stopped being quick and simple and become more considered, with also increased feelings of uncertainty.

Moratorium
Male 2 You have to have good reasons for laying someone off: justifications, facts, and evidence. Then you're going by the book and not stepping into any gray zone.
Decision-making is based on formal guidelines, which are adhered to without personal judgment or questioning.

Foreclosure
The second was in a situation where all the other participants talked about how morally problematic it would be to change an employee's duties after the person returned from sick leave due to a serious illness, if the change was against the employee's wishes. One manager did not recognize the moral nature of this decision: 'I can't understand why you can't just tell the employee that (s) he needs to do another job' (Group 3, male 1). This shows that the manager paid no attention to the ethical aspect of the situation and his consideration of the question simply involved following the rules and exercising the rights of the manager. The third example is shown in Table 1 (Group 2, male 2).

Foreclosure
Foreclosure was used for individual moral evaluations that were based unquestioningly on external values adopted from, for example, the organization. In these cases, justifying one's moral decisions was often based on what was best for the company: 'In our field you have to have a good ethical reputation. You can't do good business if people think that you're being dishonest. 'I've recruited people and then laid them off after six months. It was ethical from the point of view of the organization, because the employees didn't fit in.' In these examples the managers saw that their decisions were beneficial to the organization, and any personal values were set aside. Some of the managers felt that acting according to the company's values and rules was the only option they had: 'Isn't that more of a question of functionality than ethicality? You take and accept the rules… It is actually really easy, working in an exchange-listed company, in the way that things are just judged according to the financial outcome' (Group 2, male 2).
For some managers, the expectations imposed by the organization or senior management created feelings of strain. Nevertheless, they did not express any alternative way of making decisions in these morally challenging situations: 'In our field it's important to have sufficient turnover of employees. Sometimes the organization pressures me to lay off employees who are good people but who don't make enough contribution to the business. These are tough situations, but I just have to take them unemotionally, not personally' (Group 3, male 1).

Moratorium
Managers in the moratorium status were actively thinking about ethical issues but had not yet clearly committed to any certain values.

Ethical principles in different MWI statuses
Phrases that included ethical reasoning based on certain principles and rules were coded as duty ethics. These included situations where the manager talked about his or her ethical decision-making in terms of principles such as equality, fairness, and trust. For example: 'For me, there was no choice, I would never tell anyone (the personal things an employee has said in confidence). Not even to the employee's supervisor, who's wondering why the employee's performance level has fallen'. Alternatively, the reasoning drew on legality: 'If you think about discrimination… the law states that even if you have a family relationship with the potential future employee, you should hire them if they're qualified for the job' (Group 1, female 1).
Virtue ethics (seen as individual moral qualities manifested in actual behavior) was coded in phrases that emphasized personal moral qualities and acting in line with them when making ethical decisions at work: 'It's your responsibility as a manager to intervene if you're aware of a conflict in the work community. You need to raise the issue with the employees. For me that's the ethics of being a manager: you tackle issues. Conflicts don't solve themselves if you keep quiet' (Group 2, male 2).
Managers' phrases that emphasized the consequences of their actions were coded under the third ethical principle, utilitarianism. These mostly focused on maximizing the benefits to the organization and solving ethical dilemmas efficiently: 'In all our decisions we have to follow one principle: what is best for the business.' In some cases the managers used utilitarian reasoning to cope with ethical decisions that went against their own moral beliefs: 'Even though I know we're facing lay-offs (after losing a client), I have to act positive to the employees. Convince them that we'll get new clients and make this profitable, even though I'm really skeptical about it… but if I would tell them the truth, that they're going to be laid off, their work input would fall for the remaining six months' (Group 4, male 2).
Finally, care ethics was coded when a manager emphasized maintaining relationships and considering the needs of others as the basis of their moral reasoning. These situations often involved making decisions that had direct consequences for their subordinates, as in the following example: 'There are many practical decisions, such as annual leave arrangements: everyone should get treated the same, get sufficiently long periods of leave and at good times. I think that taking the work-family interface into account is often hard: I have a lot of employees who have small children at home, which makes these decisions challenging; how to be fair' (Group 3, female 1).
We compared the co-occurrence of the phrases coded with the four ethical principles and the four MWI statuses.
Because of the small sample size, we present the full distribution between the ethical principles and MWI statuses instead of statistical tests (see Table 2). We found that utilitarianism typically coincided with the foreclosure status, while virtue ethics was typically associated with the achieved status and less frequently associated with the foreclosure status. Finally, ethics of care more typically co-occurred with the moratorium status. For duty ethics no clear differences were found between the MWI statuses.

Sample
The participants were 180 managers who took part in a postal questionnaire in 2015. This is the fourth wave in a larger longitudinal study, and a more thorough description of the data collection procedure can be found in [name deleted to maintain the integrity of the review process]. Managers who had not declined to be contacted during previous rounds of data collection were sent a follow-up questionnaire and an invitation letter, which included a description of the aims of the study, the voluntary nature of participation, and information about confidentiality. This means that all of the managers who took part gave their informed consent to participation. We included only the final round of data as the previous questionnaires did not include items measuring moral identity.

Procedure
Moral identity statuses were coded on the basis of managers' self-generated assessments of their ethical decision-making. The questionnaire included the following open-ended question: 'When you face ethically challenging situations at your work (for example, situations where you have to think what is the right way to act), on what grounds do you make decisions in these situations?' followed by four empty lines for the managers' answer. The question was deliberately left at a general level so that it would prompt the participants to reflect on and summarize their thoughts about their moral decisions as they personally thought fit, without limiting them to any pre-defined concepts. Background variables included age, gender (male/female), tenure with the current employer (in years), work sector (private/public/own company), organizational field (industry/public administration/business services or renting/finance and insurance/telecommunications or data processing/commerce and trade/other, e.g., health care, public relations, and traffic), and staff responsibility (yes/no).

Analysis
Two researchers independently read all the open-ended answers, established that the contents were appropriate for the use of the moral identity framework, and proceeded to classify the answers in one or other of the four moral identity statuses. After this individual, preliminary coding phase, the researchers discussed together ambiguous cases and agreed upon a set of principles, as follows. (1) If the participant had written descriptions that would fit several different statuses, the researchers would try to identify the most salient viewpoint and base the final identity code on that. For example, 'I try to think what decision would be aligned with the company's values. Sometimes I might [emphasis added] also discuss with my supervisor or colleagues.' was coded for foreclosure, because company values were highlighted in the answer.
(2) All answers that included phrases such as 'I base my decisions on facts/common sense' or 'I try to look at the whole' were seen to indicate that the participant reduces ethical issues to factual questions without any moral reflection, and were thus coded in the diffusion status.
(3) Answers that emphasized discussing the situation with others (e.g., in order to get help and support) were coded in moratorium, because they included an element of uncertainty regarding personal decision-making. However, if the answer included an association between discussing with others and shifting responsibility for the decision to them, those answers were coded in diffusion (avoiding a personal decision). (4) Answers including the phrase 'values' were interpreted as reflecting the person's own values and were coded in achieved, unless the answer clearly indicated that the values referred to were the organization's values, in which case the answer was coded in foreclosure. (5) Answers in which the participant described balancing between personal and organizational values without emphasizing one over the other were coded in moratorium. Similarly, answers in which the manager acknowledged that several different and/or competing values and norms affect their decision-making (e.g., laws, ethical codes, personal values) but did not take a stand on which had the greatest influence on them were coded in moratorium. After agreeing on these coding principles, the two researchers coded all the answers individually. This resulted in 92% agreement between the two researchers.
Finally, age and tenure between moral identity statuses were tested with ANOVA, and gender, work sector, organizational field, and staff responsibility with crosstabulation using the χ 2 test.

Results
The final classification of the managers into the four moral identity statuses is presented in Table 3. As described in the conduct of the analysis, not all of the answers were straightforwardly classifiable into the theorized identity statuses, but after agreeing on the aforementioned coding principles we were able to interpret and identify one predominant identity status for each manager. In other words, we found no new dimensions from the data that did not fit into the identity status paradigm. There were no significant differences between the statuses in terms of age (F(3, 176) = 0.318, p = .81), tenure (F(3, 176) = 0.826, p = .48), work sector (χ 2 (12) = 19.55, p = .08), organizational field (χ 2 (21) = 31.24, p = .07), staff responsibility (χ 2 (3) = 1.85, p = .61), or gender (χ 2 (3) = 4.55, p = .21).

General Discussion
The main conclusion that we draw from our two studies is that Marcia's (1966) identity status typology is applicable to managers' moral identity in the work context. Study 1 showed that the moral identity status typology is applicable to managers' reflections about their moral values when they talk about facing moral challenges at work. However, the context might have affected our findings, as the results derive from 16 managers in eMBA training and their focus group discussions: social conformity could have led the managers to express opinions that were in line with those of the majority of the group, and the small and somewhat homogeneous sample may have led to results that might not hold with a more varied group of managers. We therefore went on to make a second study, which was conducted among a larger and more heterogeneous sample of managers. The benefits of this kind of mixed-method study design are that it provides both triangulation (i.e., seeking convergence of results from different methods studying the same phenomenon -in our study, qualitative coding and further statistical testing) and complementarity (i.e., seeking elaboration and clarification of the results from one method with results from the other method -in our case, continuing from a focus group study to a questionnaire study) (see, e.g., Johnson, Onwuegbuzie and Turner, 2007). Study 2 provided evidence of the transferability of our findings from Study 1, showing that the moral identity statuses were applicable among a more varied sample of managers. However, we did not find any significant differences between the status groups in relation to any of the individual or contextual background factors. If managers' age, tenure, gender, or general work characteristics do not explain the variation in individual moral identity maturity, we need to look for other potential antecedents. It is possible that managers differ in the degree to which they have had experiences that develop their ability to solve moral problems (Dane and Sonenshein, 2015). This can further relate to the differences in the stage of moral identity they have reached through their exploration of moral values and their degree of commitment. This implies that the statuses are not stable but susceptible to change.
Our conclusion from the two studies is that the achieved status associates with a highly internalized moral identity which is the result of a personal process of value exploration and commitment. The other statuses represent less mature moral identities, where personal values remain unrecognized (diffusion), are unquestioningly adopted from external agents (foreclosure), or are under personal exploration but still lack any strong commitment (moratorium).
Looking at each of the four different moral identity statuses in closer detail we find, first of all, that the diffusion status got only very tentative support in our focus group study. It is possible that because the managers in this study were participating in the executive MBA training, they had already spent time reflecting on their ways of working from different perspectives. Therefore there might have been nobody in this group who was totally unaware of the moral aspects of their work and how their personal values apply in these moral dilemmas. However, Study 2 provided more evidence for the applicability of the moral identity diffusion: almost one fourth of the managers were coded in this status. These managers described their most typical ways of solving ethical dilemmas as fading any ethical aspect, focusing on facts and using common sense, or passing responsibility for the solution to others (e.g., to upper management). This may reflect managers' attempts to avoid confronting the moral nature of problematic decisions they have to make as a part of their job, which can also be a cognitive coping strategy (Detert, Treviño and Sweitzer, 2008;Tenbrunsel and Messick, 2004).
We were more clearly able to identify managers whose ethical decision-making was characterized by internalized values that had been adopted from the outside (foreclosure). In both studies, managers in foreclosure highlighted the values and norms of the organization in which one works. They emphasized their feeling of obligation to follow the norms and rules set by the organization when making these decisions. Also their ethical reasoning focused on the consequences of moral decisions (utilitarianism) and they tended to prioritize their organization when making ethical decisions. Thus, managers in foreclosure can see ethical decision-making as more straightforward, leading to more consistent decision-making in different dilemmas. As Bosma and Kunnen (2001: 44) conclude, ' assigned commitments can also be strong, but they lack the flexibility and adaptiveness of self-chosen commitments'.
Moratorium was the most common identity status in our focus group study, but the rarest status in the questionnaire study. It should be noted that the answers that were interpreted as moratorium (e.g., answers describing several viewpoints that affected the manager's decision-making, but without emphasizing any of them) did not include an explicit description of any active measures that the managers would take in order to weigh up and explore the different possible options when solving ethical dilemmas. It is possible that the short, openended question, with its focus on the most common way of solving ethical problems, did not adequately capture this active exploration period. In Study 1, managers who were coded for moratorium often described work-related ethical dilemmas as demanding and wearing, because they were uncertain about their own values and how to solve the dilemmas. This may have been due not only to the ongoing process of value exploration with experiences of uncertainty, but also to their critical evaluation and questioning of the prevailing values and practices in the organization. This would be in line with the original identity status theory (see Kroger and Marcia, 2011), which states that moratorium may bring out feelings of anxiety. Managers in moratorium also used ethics of care more often than did other managers. Our analysis suggests that the situations these managers referred to were often related to well-being, equality, and the needs of others. It may be that when such complex and high intensity social and moral questions are involved in work situations, more managers use uncertain moral reasoning (i.e., are in the status of moratorium) and they also apply the ethics of care.
Some managers clearly described following their personal moral values, which they felt were an important part of themselves. For example, managers in both studies described how they wanted to be fair, just, honest, and even-handed in their decisions. These managers belonged to the achieved moral identity status, characterized by selfchosen commitments that indicate internalized morality. They also used virtue ethics more than others did when describing their ethical reasoning. According to the identity development model (see Bosma and Kunnen, 2001), once a person has gone through the processes of exploration of and experimentation with different alternatives (in this case, moral values), they tend to be more flexible and adaptable in their solutions, and thus represent a more mature mode of psychosocial functioning. Thus, achieved moral identity is the most mature identity status, which is likely to capture the dimension of strong internalization in Aquino and Reed's (2002) moral identity concept.

Implications for theory and practice
Our findings have a range of implications for both theory and practice. First, we found that there are significant differences in how adults describe their personal values when it comes to making moral decisions at work. In other words, our findings show that there is variability in the identity status of working adults, which gives preliminary support to the notion that moral identity development extends into adulthood (Bergman, 2002). Managers differ in their moral identity maturity, which can be captured by looking at whether the individual has gone through the processes of personal moral value exploration and commitment. Identity status research has shown that individuals who have achieved identities have taken ownership of their own identity development (unlike those in e.g., Foreclosure), and may be better able to respond to rapidly and unexpectedly changing circumstances (Schwartz et al., 2013). Supporting individual moral reflection (commitment to self-chosen moral values) could therefore help to prevent unethical behavior and promote moral action in the work place. This is because achieved moral identity is likely to include a strong sense of being a moral person, which is an important source of moral motivation and provides a link from what one believes is right or wrong to actual efforts to do 'the right thing' (Mayer et al., 2012).
We also found that individual factors (age, gender, tenure) did not show any significant differences between the moral identity statuses. This means that other factors must explain the level of moral identity maturity at work. The general model of identity development posits that different life events that are experienced as conflicts can trigger identity change (Bosma and Kunnen, 2001). This leads us to suggest that having the opportunity to experience, acknowledge, and reflect on different moral dilemmas can promote moral identity development. For example, experiences that are specific to the work context, such as moral dilemmas that one has faced in the past (Dane and Sonenshein, 2015), could bring about changes in these statuses. It may be, too, that especially high intensity dilemmas that provoke moral emotions (Ford et al., 2017;Zeelenberg et al., 2008) can promote personal reflection and lead to changes in moral identity statuses through the processes of value exploration and commitment.
Second, we found that moral identities can be based on organizational values and norms that guide moral decisions at work (foreclosure). This highlights the role of the organizational context. Although our findings on the differences in terms of work sectors and organizational fields between the identity statuses did not reach statistical significance, these p-values were close to the significance threshold (0.08 and 0.07, respectively). The non-significant results might have been affected by our small sample size, which was further divided into several subgroups based on the identity statuses, leading to low statistical power. With this reservation in mind, our tentative findings showed that achieved leaders were under-represented in the industrial field, while diffused identities were over-represented among managers there. In addition, foreclosed identity statuses were overrepresented among leaders working in the private sector, whereas achieved moral identities were over-represented among entrepreneurs who owned their own company. This suggests that moral issues may be less acknowledged in industry and that private sector leaders are more likely to justify their moral decisions on the grounds of the benefits to the company they work for. Founding one's own enterprise, on the other hand, coincides with values one has personally chosen. Future studies should therefore explore whether certain moral identity statuses might be more or less pronounced in different work environments.
According to Weaver's (2006) model of organizational influences on moral identity, organizational values can also affect moral identity development. For example, ethical culture has been found to positively influence employees' moral efficacy and ethical sensitivity (Jennings, Mitchell and Hannah, 2015). This in turn could affect one's development toward moratorium and/or achieved status, by providing opportunities to explore a broader range of moral values that could be applied in moral dilemmas at work. Working in a less ethical environment, on the other hand, can lead to downplaying ethics in different situations (Kvalnes, 2014), consequently leading to diffusion (e.g., by reducing ethically challenging situations to purely factual questions in order to avoid the need for any deeper moral consideration) or foreclosure (e.g., adopting the organization's amoral values in order to avoid moral conflicts). More research is needed on how the experience of previous moral encounters at work can bring about changes in personal moral identity development, and what kinds of previous moral encounters could do this.
Third, we found that the ethical principles that the managers used differed between moral identity statuses, pointing toward identity having a role in recognizing and solving ethical dilemmas at work. A previous study has shown that internalized moral identity reduces unethical behavior, while moral identity based on external factors (symbolization) increases it (Reynolds and Ceranic, 2007). This suggests that the motivation to act morally comes from within, and when an individual has strongly integrated moral values (such as achieved identity), this creates a need to act consistently with this identity, thus leading to moral actions. Adopting external values (foreclosure), on the other hand, can lead to acting according to company expectations without questioning them. The co-occurrence of utilitarianism with foreclosure or diffusion can further heighten the motivation to make decisions that benefit the organization, even though this could mean acting unethically or in the "moral gray zone".
Fourth, it should be recognized that if individuals have not given time and thought to their own moral values, they may play down to themselves the importance of noticing the ethical aspects of a problem they face (as in diffusion status) or they might take on the organization's values without questioning them (foreclosure) -which might not always be ethical. In these cases, individuals may be more inclined to make decisions based solely on the benefits to the organization, which can lead to the risk of unethical behavior (e.g., if the person aims to maximize profit and performance with no reference to critical and moral reasoning). Understanding managers' moral identities is especially important, as moral identity has been shown to be an important predictor of ethical leadership (Skubinn and Herzog, 2016;Zhu, Treviño and Zheng, 2016). Helping managers to develop more mature forms of moral identity and integrate moral values as a central part of their self-concept would enable them to demonstrate more ethical leadership practices.
This leads us to our final practical implication: the moral identity statuses can be considered a kind of personal journey towards moral values that the individual has himor herself chosen and that are personally meaningful. It might be that the organizational context is particularly significant for this journey, either facilitating or hindering moral identity acquisition. Individuals integrate identity elements from their social environment (Adams and Marshall, 1996), and therefore it is possible that working in a competitive, profit-oriented organization might provide less support for practicing moral virtues and developing ones' moral identity than working, for example, in the service of human welfare. Organizations therefore need to consider their role in influencing (by fostering or inhibiting) employees' and managers' moral identities (see Weaver, 2006).

Limitations of the study and future directions
Our cross-sectional studies could not capture the actual developmental shifts in moral identity from one status to another. It should also be emphasized here that the identity status paradigm (Marcia, 1966) is not a developmental model, as it only provides a typology that represents different levels of identity maturity. Therefore future studies should investigate how moral identities are developed within the context of work and consider how status shifts from one to another might take place. One fruitful approach would be to focus on changes in commitment: over time, the content of commitments can change, and commitments can become stronger or weaker, more rigid or more flexible (Bosma and Kunnen, 2001).
These changes in strength and quality of commitments should be considered when investigating moral identity construction among working adults.
There may also be considerable variation in what kind of events and sources of influence lead to identity status transitions. For example, environmental demands, such as organizational norms or expectations that are in conflict with one's personal values, can create pressure to evaluate and revise one's personal moral values. This can be experienced as an identity crisis, leading to the renegotiation of one's moral commitments. However, individuals differ from each other in their readiness to re-evaluate their personal values or their way of recognizing and responding to these situations: some people will revise their personal values, while others will create new coping skills such as fading the ethics and morally disengaging from the situation, or leave the organization. Some individuals are also more open than others to feedback and readier to change their commitments in response to situational information (Berzonsky, 1989;Dane and Sonenshein, 2015). Others will be more inclined to defend their normative core beliefs (foreclosure), leading to a more rigid attitude towards feedback.
The moral identity category of diffusion can lead to quick and reactive behavior based on environmental expectations or cues (Berzonsky, 1989), because the individual does not have a coherent sense of their own personal moral values in the situation. In these cases, changes can appear as superficial adjustments made in response to the ethical challenge at hand, without their leading to any fundamental changes in personal values. This is in line with Bergman (2002), who proposed that sometimes situational influences or other factors can override personal moral values and the motivation to act according to what one regards as right. Personal development can also be multidirectional: different moral values can become more or less central to one's identity with experience (Bergman, 2002). Future studies are needed to increase our understanding of the changes between MWI statuses.
Finally, regarding the methodology used here to study and capture moral identity processes, although using focus group discussions has the advantage that it can enable a more personal and comfortable atmosphere, without the researcher being present, observing and potentially influencing the content of the discussion, there could have been a pre-established social hierarchy within the familiar groups which may have influenced the themes that arose in the discussions. The group dynamics might have defined the themes that were seen as "appropriate" to talk about if the more dominant participants chose the topics. For example, most of the discussions revolved around situations related to the workplace (e.g., interpersonal relationships or organizational norms) and did not touch on other stakeholders, such as customers, suppliers, or media. Future research might therefore consider also using individual interviews in addition to focus groups. This could give greater access to minority opinions and provide a safe space where different viewpoints could be expressed and more personal and sensitive issues raised.
We should also note that there are limitations in trying to assess moral identity statuses by means of one openended question (as we did in Study 2). Facing and solving moral conflicts is complex. However, we chose to ask the participants to summarize their thoughts in a fairly limited space, hoping to tap into their typical way of handling such issues. The risk is that forcing the managers to give short responses could have led to over-simplified answers or descriptions that do not cover the full range of issues and the nuances in their moral decision-making and identity-related mechanisms. Therefore future studies should consider applying existing identity measures, such as broader questionnaires (e.g., the identity style instrument; Berzonsky, 1989) or developmental interviews (e.g., the identity status interview, see Fadjukoff, Pulkkinen and Kokko, 2005), in the moral domain.

Conclusion
Based on our two studies we can conclude that moral work identity is a promising concept that can provide insight to working adults' moral maturity in the identity domain. The four moral identity statuses help to illustrate the variety of moral value exploration and commitment. These statuses reflect the basis of individual ethical decision-making at work: how much it is based on a personally chosen value framework, on adopted organizational norms, on an on-going search for 'the right way', or on a lack of established moral values, which can lead to the avoidance of moral questions. We hope that future studies will shed more light on the processes involved in the development of moral identity and will bring more understanding of how moral identity is related to actual moral behavior at work.

Competing Interests
MH was financially supported by the Academy of Finland (Grant No. 294428). All other authors have no competing interests.