Predictors affecting expatriation willingness family barriers to expatriation?

(1)

Predictors affecting Expatriation Willingness:

Family Barriers to Expatriation?

Master Thesis

MSc Business Studies

M.M.G. de la Motte 10219072 Michelle.delaMotte@student.uva.nl Supervisor: S.T. Mol Second Reader: A.H.B. de Hoogh

(2)

2

Abstract

Many multinational organizations send expatriates abroad in order to manage the subsidiaries operations effectively. According to Mol, Born, Willemsen, van der Molen and Derous (2009), due to high expatriate selection ratios, multinational organizations are having difficulties selecting potential expatriates that will perform well and who are willing to take on an international assignment. Applicants with expatriation aspirations are an effective approach to reduce the selection ratio. Mol et al. (2009) investigated predictors relative to increasing expatriation willingness and found that seven biodata items (about previous international experience) emerged as the most powerful predictor set in their study.

This current study replicated the study by Mol et al. (2009) by investigating whether the biodata items predicts expatriation willingness. Career- oriented is included to investigate whether this also predicts expatriation willingness. Moreover, multicultural competence and personal agency were used as mediators to see whether they could explain the variance between the predictors and expatriation willingness. Vitally, family plays an important decision making role whether one will decide to move abroad. The current study investigated whether family barriers will moderate the relationship between the predictors and expatriation willingness.

By means of a cross- sectional design, 292 participants (expatriates and domestic employees) from numerous nationalities took part in the survey. The results concluded that the biodata items and career- oriented are positively related to expatriation willingness. No mediation was found for multicultural competence and personal agency only mediated between two biodata items and expatriation willingness. Finally, this study found that higher family barriers moderate the relationship between the biodata items, but did not moderate career- oriented and expatriation willingness.

(4)

4

Introduction

As the world today is ever changing, and becoming increasingly diverse, many companies are challenged to expand their operations internationally. They are confronted with competitive challenges, which drive them to become more sustainable, global, and technologically competent (Noe, Hollenbeck, Gerhart & Wright, 2010). Since companies are expanding globally, the number of employees being sent on international assignments has increased considerably over the years. Employees that are assigned to live and work outside of their country of origin are known as expatriates. Aycan & Kanungo (1997) define the term expatriates as, “… employees of business and government organizations who are sent by their organization to a related unit in a country which is different from their own, to accomplish a job or organization- related goal for a pre-designated temporary time period of usually more than six months and less than five years in one terms” (p. 250).

Expatriates are considered to be a major investment for multinational enterprises, as it has been cited to cost between $300,000 and $1 million annually, which amounts to two to three times as much as local or host country managers earn annually (Black & Gregersen, 1999; Collings, Scullion & Dowling, 2009; Cullen & Parboteeah, 2013). The rise of studies done within the expatriate management field may be due to the fact that many expatriates’ foreign assignments abroad have not been successful (Black & Mendenhall, 1990). Expatriate assignments that are not considered to be successful are when expatriates return to their home country earlier than intended, due to their lack of ability to adjust to the foreign culture, and demonstrating poor job performance (Morris & Robie, 2006; Adler, 2001; Black & Mendenhall, 1990; Deshpande & Viswesvaran, 1992). The premature return of the expatriates or expatriate failure is defined as ‘the inability of an expatriate to perform according to the expectations of the organization’ (Harzing & Christensen, 2004, p. 7). Expatriate failure is not

(5)

5 only very costly for companies, but it may also damage the employee’s career (Callahan, 2007). On the other hand, Harzing (1995) argues that if an expatriate stays during the entire defined time period abroad and fails to perform their job adequately, this may cause more significant financial damage compared to expatriates that leave earlier from their assignment abroad. Harzing (1995) also concluded that expatriate failure rates are not as high as it is typically claimed within the expatriation literature.

Baruch & Altman (2002) proposed that HR managers should consider the management of expatriates, pre-posting preparation, the selection of potential expatriate candidates, maintenance of the relation with expatriates and the facilitation of repatriation in their strategic approach, in order to execute the expatriate practices effectively. The current study will focus on expatriate selection practices. Previous studies looked at the antecedents of expatriate job performance so as to choose and select suitable candidates (Mol, Born, Willemsen & van der Molen, 2005; Harris & Brewster, 1999; Anderson, 2005). The international strategy of organizations relied heavily on expatriate job performance as expatriates are the organization’s strategic resource, in order to enhance foreign direct investment and subsidiary performance (Wang, Chen, Tong & Kim, 2009). In light of the importance of selecting candidates based on valid predictors of expatriate job performance, going beyond that criterion, Selmer (2001) stated that companies are struggling to find employees who are willing to accept the assignment abroad. Due to high expatriate selection ratios, it isn’t about selecting the best and qualified candidates to do the job, but rather finding a candidate who is willing to take the job abroad (Mol et al., 2009; Selmer, 2001). As a matter of fact, Harvey (1995) suggested that initial motivation and willingness to go abroad for an assignment are important factors to consider. In other words, initial willingness and motivation are, in fact, strongly related to all other criteria of successful expatriation, such as overcoming cultural shock, lack of adjustment, and homesickness, and, ultimately improving

(6)

6 expatriate performance. Thus, companies should be sensitive to motivational issues early during the selection procedure to increase the expatriate candidate pool, and in order to avoid later costly failures. Thus, this study will look into valid candidate selection predictors of expatriation willingness.

Firstly, looking deeper into the concept of expatriation willingness, Mol, Born, Willemsen, van der Molen & Derous (2009) investigated a range of predictors geared toward the prediction of futureexpatriation willingness behavior. Expatriation willingness has been defined as: “the likelihood of accepting a job offer that requires living and working in a foreign country for a temporary period” (Mol et al. 2009, p. 2). Their study indicated that core self-evaluations, the Five Factor Model, expatriate specific predictors (e.g. cultural flexibility, intercultural sensitivity, tolerance for ambiguity and tolerance for uncertainty) and the biographical items accounted for 50% of the variance of expatriation willingness, with the biographical items emerging as the most powerful predictor set (Mol et al., 2009). The biographical items (i.e. biodata) that were investigated by Mol et al. (2009) included the following seven items: foreign number of countries visited on holiday, travel remoteness, foreign living experience, foreign travel liking, the number of friends living abroad, number of foreign friends living in their home country and foreign language ability. Replicating Mol et al. (2009) research, this study will demonstrate that expatriation willingness can be predicted on the basis of the biodata set. In addition, Mol et al. (2009) did not provide empirical evidence for why these biodata items are related to expatriation willingness. This being the case, this study will also try to add to the expatriation willingness literature by investigating the underlying mechanisms that might explain the variance between biodata and expatriation willingness.

Another study on expatriation willingness was conducted by Adler (1986), who found that one of the main reasons for individuals to expatriate was the attractive career benefits that

(7)

7 came with it such as the financial rewards, independence, challenge and unique lifestyle. Barrett & O’Connell (2001) showed that another central motivation for individuals to expatriate was most often for career opportunities as well as career advancement. When considering expatriation, individuals construct future career goals and expectations for their long-term career strategy (Tharenou, 2008). When individuals believe that expatriation will be instrumental to gain career advancement, this is likely to increase their willingness to take on an expatriate assignment. This study will also investigate whether career- oriented can predict expatriation willingness.

Looking closer at the aspect of interaction among employees from multiple cultures, nowadays it has become more common for individuals to be exposed to two or more different cultures due to changing mobility patterns and changing pattern in the world’s labor force (Pollock & van Reken, 1999; Brannen & Thomas, 2010). The concept of being exposed to two different cultures is known as biculturalism. LaFramboise, Coleman & Gerton (1993) defined biculturalism as: “belief, or confidence that one can live effectively, and in a satisfying manner, within two groups” (p. 402). The concept of biculturalism is limited when it comes to expatriates, as it is very likely that they are exposed to more than two cultures and are expected to interact or work with individuals from more than two different cultures at the same time. By reviewing the past literature and definitions of biculturalism, this study will propose an applicable definition to expatriation management and review the concept of multiculturalism which acknowledges the potential exposure to two or more than two cultures. Individuals with the ability to maintain the knowledge of their distinctive cultures (i.e. values, customs, beliefs and habits) and who can effectively integrate and interact with people from other cultures with little complication are likely to be more willing to accept an expatriate assignment (Bell & Harrison, 1996; LaFromboise, Coleman & Gerton, 1993). Moore (2009) argued that there is a gap in the expatriation literature that has yet to determine

(8)

8 whether bicultural or multicultural identity can be identified as a predictor of expatriation willingness. This paper will investigate whether multicultural individuals mediates the relationship between biodata and expatriation willingness. With regard to the biographical items, individuals with previous international experience have gained the knowledge of interacting with others from different cultures. This experience may be expected to lead to a higher capability of the individual to live effectively and adjust themselves accordingly to other foreign cultures. This study will therefore research whether individuals with previous international experience is related to a higher multicultural capability, and whether this, in turn, would influence their expatriation willingness.

According to Bhaskar- Shrinivas, Harrison, Shaffer & Luk (2005), an individual’s personal resources, are critical to meet the uncertainties and demands of living in a foreign culture. In other words, the individual should assess whether they can cope with the uncertainties and demands caused by the new circumstances they would have to face (Tharenou, 2008). Borstoff, Harris, Field & Giles (1997) claimed that it is likely for individuals with high personal agency to deal with the uncertainty, disruption, lack of familiarity and change of living abroad. Personal agency would be beneficial, in the prediction of behavior and attitudes associated with adapting to a foreign environment and culture. Personal agency can be defined as: “the belief that one can exercise control over the environment to be able to achieve desired goals and avoid undesired ones” (Bandura, 2001, p. 169). Tharenou (2008) in her study also confirmed that personal agency has a significant impact on expatriation willingness. Individuals with prior international experience have developed to be more self-confident and are more likely to overcome the challenges encountered when living in foreign countries (Tharenou, 2008). These individuals are able to control the environment and cope with the adjustment stressors associated with expatriation, which results in greater personal agency. In terms of career- oriented, when individuals

(9)

9 perceive that expatriation would lead to long-term career benefits and career advancement, this would increase their belief that they are able to cope with the uncertainties and demands of living abroad in order to reach their desired career goal and in turn increase their personal agency (Tharenou, 2008). A higher personal agency is likely to predict expatriation willingness. In this present study we will investigate whether personal agency mediates the relationship between past international experience and expatriation willingness and whether it also acts as a mediator between career- oriented and expatriation willingness.

Tharenou (2008) also emphasized the relevant role of family barriers. Tharenou (2008) indicated that no support of relocation from the family, a partner with an important career binding her/him at the home country, and children leads to high family barriers, which in turn diminishes the likelihood of expatriation (Tharenou, 2010). Overall, the role of family barriers cannot be ignored in terms of expatriation. Therefore, family barriers are proposed to moderate the relationship between the biodata items and expatriation willingness, as well as moderate the relationship between career- oriented and expatriation willingness.

The aim of this thesis is to research why biodata as well as career instrumentality are related to expatriation willingness. This paper will provide likely explanations as to why biodata and career instrumentality are related to willingness through possible mediating effects of personal agency and multicultural competence. In addition, this study will look into whether family barriers moderate the relationship between biodata and also career instrumentality on expatriation willingness.

(10)

10

Literature review and hypotheses development

As mentioned earlier, having the willingness to move and work abroad, could overcome the various problems that are associated with the adaptation processes (McFarland, 1997; Harrison & Shaffer, 2005). Harrison and Shaffer (2005) argued in their study that if the candidate shows that he or she is motivated to go abroad, this would have a significant impact on expatriate adaptation. The psychological motivation and adaptation to the international assignment and foreign culture has an impact on the effort regulation, job performance, and on the allocation of personal resources to specific daily job activities (Harrison & Shaffer, 2005). In other words, showing more effort and motivation to adapt to the host country’s culture, are important features to overcome adjustment barriers when living abroad.

Mol et al. (2009) argued in their study that it is difficult to find expatriate candidates due to high selection expatriation ratios. A selection ratio implies the ratio of the number of hired employees to the number of available applicants (Mol et al., 2009). The ratio of the number of employees to the number of employees actually willing to go abroad is too high, which thwarts the ability for international organizations to select potential expatriates (Mol et al., 2009). By establishing an internal candidate pool by preselecting candidates on their expatriation willingness, organizations would not only decrease the selection ratio, they would have the ability to subsequently choose candidates from that pool they expect will perform best during their assignment abroad and who are actually willing to embark on an assignment abroad. From a Human Resource Management perspective, this approach of selecting prospective expatriate candidates on their expatriation willingness, may be pursued as a more efficient than selecting highly qualified candidates that project a lack of expatriation willingness (Mol et al., 2009). Mol et al. (2009) studied which predictors are most likely to explain the variance in expatriation willingness. On basis of Mol et al. (2009), their research

(11)

11 has shown that biodata items set about previous international experience explained the most variance on expatriation willingness. Tharenou (2008) also found that people who perceive expatriation as an instrument to advance in their careers, are more willing to take on an international assignment. Given these points, the current research will look at the context of career- oriented and the biodata items as selected predictors that would lead to the individual’s willingness to expatriate, by also including mediating variables (i.e. personal agency and multicultural competence) and a moderating effect (i.e. family barriers).

Biographical items

As mentioned earlier, relevant past behavior is one of the best predictors of future behavior (Guion, 1998). Glanz, Williams & Hoeksema (2001) support the notion that previous international or expatriate experience will have a positive influence on the willingness of the individual to expatriate again. Borstorff, Harris, Field and Giles (1997) found that individuals with international experience are more able and competent to deal with certain barriers and uncertainties encountered when abroad for the reason that they learned relevant inter-cultural skills and knowledge. Moreover, Bhaskar- Shrinivas, Harrison, Shaffer & Luk (2005) showed that employees or individuals with international experience are expected to experience fewer stressors and encounter fewer problems when in foreign environments.

Mol et al. (2009) researched a set of predictors they expected would have a significant influence on expatriation willingness. Mol et al. (2009) found that a number of biographical items (biodata) about relevant previous international behavior, are the most powerful predictor set on predicting expatriation willingness of prospective expatriate applicants. Salgado, Viswesvaran & Ones (2001) stated the following: “biodata are one of the most valid predictors of personnel selection, and their validity can be generalized across organizations, occupations and samples”(p. 182). As mentioned before, the biodata items that were

(12)

12 investigated by Mol et al. (2009) included the following seven items: number of countries visited on holiday, travel remoteness, foreign living experience, foreign liking, number of friends abroad, number of foreign friends living in their home country and foreign language ability. In their research, they found that the number of friends abroad and foreign language ability were found to moderately to strongly correlate with expatriation willingness. Foreign travel liking was the only item that did not significantly correlate with expatriation willingness. Despite the fact that foreign travel liking did not significantly correlate with expatriation willingness, it will be included in this research to further investigate whether the results of Mol et al. (2009) replicates.

Therefore, we can expect a positive relationship between previous international experience (i.e. biodata) and expatriation willingness. In line with Mol et al. (2009) it is proposed that past international experience will be positively related to expatriation willingness.

Hypotheses 1a – 1g: Past international experience items (i.e. biodata items) such as number of countries visited on holiday (a), travel remoteness (b), foreign living experience (c), foreign

travel liking (d), number of friends abroad (e), number of foreign friends living in home country (f), foreign language ability (g) are positively related to expatriation willingness.

Career-Oriented

This current study will also investigate the relation between career-oriented and expatriation willingness. Maier, Rappensperger, Rosenstiel and Zwarg (1994) defined career-oriented individuals as: “attitudes expressed by super-ordinate intentions of an individual that will influence career-related decisions”(p. 4). In an early study on expatriation, Borstorff et al. (1997) showed that career focus variables were significant predictors of expatriation willingness. Career focus variables entails whether expatriation would offer employees the

(13)

13 opportunity for advancement and recognition, whether this would improve their career development and mobility, and whether this would prepare them for advancement towards executive top management positions (Borstorff et al. 1997). In addition, Feldman and Thomas (1992) also showed evidence for a correlation between career goals and expatriation willingness. Expatriation has its perks such as financial rewards, not only does it offer a unique lifestyle, but many employees perceive it as an attractive long- term career strategy (Adler, 1986; Cerdin & La Pargneux, 2009).

Richard & McKenna (2002) found three major motives as to why individuals embarked on an expatriation assignment, which were financial rewards, desire to travel and to change their lives for personal growth. However, Clegg and Gray (2002) found the opposite, namely that individuals are more inclined to take on an expatriate assignment for career opportunities and financial benefits more often than fulfilling their partner’s needs, to satisfy their desire to travel or to change the life. In other words, it is more likely that employees or individuals that are career- oriented would take on an international assignment in order to meet their goals and to provide a career ladder towards top management and executive positions (Przytula, 2011). Moreover, when employees or individuals consider expatriation, they are likely to construct career goals and long- term career expectations (Tharenou, 2008). Therefore, we can expect a positive relationship between career- oriented and expatriation willingness.

Hypothesis 2: Career- oriented is positively related to Expatriation Willingness.

Multicultural competence

Employees working for a large multicultural organization stand to gain many personal benefits such as: a greater personal ability to cope with the unexpected; greater tolerance and acceptance of others’ differences; greater flexibility in their own personal behavior,

(14)

14 communication and interaction styles; and enhanced self- insight (Trefry, 2006). More specifically, Bell and Harrison (1996) state in their research that individuals that have multicultural competence could be a positive attribute in this dynamic and global working environment. Working for a multinational company or in an international environment, it is very likely that one will encounter working with other people from different backgrounds and cultures. As these individuals are familiar with two or more different cultures, they are accustomed to recognizing different value systems and expectations (Bell & Harrison, 1996) in response to cultural cues (Hong, Morris, Chiu & Martinez, 2000). LaFramboise, Coleman & Gerton (1993) proposed in their study that, in order to live effectively between two or more cultures one should: respect the knowledge of cultural beliefs and values, positive attitude toward both majority and minority groups, multicultural cultural efficacy, communication ability, role repertoire and a sense of being grounded (LaFramboise, Coleman & Gerton, 1993). In addition, multicultural individuals possess unique cognitive, affective and behavioral repertoires that would appear to be important attributes for successful expatriate candidates (Bell & Harrison, 1996).

In light of individuals with previous international experience are likely to have accumulated knowledge as well as acquired relevant intercultural skills, in order to deal and overcome uncertainties when living abroad (Borstorff et al., 1997). The study by Domanico, Crawford & Wolfe (1994) supported the notion that multicultural adolescents are generally better equipped to adjust, and are more flexible and able to deal with acculturative stress in culturally ambiguous environments. Individuals with international experience would therefore be more self-confident when interacting with others from different cultures as opposed to individuals with no international experience (Bell & Harrison, 1996). It is very likely that individuals with previous international experience have been exposed to more than two cultures, which enables and influences their openness and acceptance to cultural diversity

(15)

15 (Yakunina, Weigold, Hercegovac & Elsayed, 2013). We can expect a positive relationship between previous international experience and multicultural competence.

As a result, individuals with multicultural competence are more likely to expect fewer problems and stressors when interacting with other people from different backgrounds, even in an unfamiliar environment setting (Tharenou, 2008). Individuals with multicultural competence have the capability and relevant intercultural skills to adjust themselves to other foreign cultures than those with no international experience (Tharenou, 2008). Specifically, individuals with previous international experience would have more multicultural competence, which may help to reduce the uncertainties, overcome disruptions and challenges encountered when living abroad, thus they would be more willing to expatriate. Therefore we can expect to see that the relationship of the biodata items about relevant past international experience and expatriation willingness, is mediated by multicultural competence.

Given the concept of multiculturalism has been a particular interest within the psychology and HRM field, there has yet to be an applicable definition of multiculturalism in expatriation literature. As mentioned earlier, the most frequently cited definition is by LaFramboise, Coleman & Gerton (1993). In terms of expatriation, their definition is very limited as it only acknowledges two cultures and does not consider the interaction and exposure of more than two cultures that come with working in multinational organizations. The following paragraphs will review previous studies and definitions of multiculturalism in order to derive and propose an applicable definition in terms of expatriation.

A summary of the various multicultural competence definitions derived from different fields such as HRM, psychology, human development and cross- cultural management are presented in Table 1 below. For the reason that bicultural identity and bicultural competence has been more developed and researched than multiculturalism, this study will touch upon a

(16)

Table 1: Defining Multiculturalism

Authors, publication year Definition Content Construct

Bell & Harrison, 1996 The state in which individuals maintain their distinctive cultures, including values, attitudes, customs, beliefs, and habits while simultaneously interacting with and learning from those of other

cultures

General cultural awareness, acceptance, bicultural efficacy, dual fluency, broad role repertoire and groundedness are competencies which are needed for

expatriate effectiveness

Developed a conceptual model and suggested various propositions

LaFramboise, Coleman & Gerton, 1993 Belief, or confidence that one can live effectively, and in satisfying manner, within two or more groups

Knowledge of both cultures’ beliefs and values, positive attitudes toward both cultural groups, confidence that one can live effectively within two groups without compromising one’s cultural identity,

ability to communicate effectively with both groups, possession of a continuum of acceptable behaviors for both groups and stable social networks in both cultures

Measured dimensions of being bicultural on assimilation, acculturation, alternation, multicultural and fusion models. Emphasis

is given to alternation model

Schutz & Unger, 2010 Represents comfort and proficiency with both one’s heritage culture and the culture of the country or region in which one has

settled

Looks at the aspects around biculturalism: what it is, how it comes into being and when it is most adaptive

No construct: commentary on Mistry and Wu’s article

Benet- Martinez & Haritatos, 2005 Internalized more than one culture Examines bicultural identity by using the Bicultural Identity Integration (BII) construct, identify their

personality and acculturation predictors of BII

Unpacked the construct of Bicultural Identity Integration (BII)

David, Okazaki & Saw, 2009 Multicultural individuals are those persons who have been extensively exposed to two or more different cultures and may have internalized, or developed, multicultural knowledge systems

Research whether bicultural individuals perceived self-efficacy, or one’s perceived ability to perform personally significant tasks, is related to individuals’

psychological well-being and mental health

Developed a measurement of bicultural self- efficacy scale (BSES) and integrated

6 dimensions of bicultural competence

Moore, 2009 The psychological and emotional transition into dual competency, allegiance, and personal identification with the

home and host societies

Explores the bicultural identity of current American expatriates and the relationship between that identity

and assignment completion

Acculturation Index

Brannen & Thomas (2010) People who have internalized more than one cultural profile Provide a brief review of what we know about bicultural individuals, point out some implications of our current knowledge for organizations, and indentify

opportunities for further exploration.

No construct, review of bicultural identity concept

(17)

Benet- Martinez & Haritatos (2005) defined multiculturalism as: “internalized in more than one culture”(p.3), which is somewhat similar to the study by Brannen & Thomas (2009). We perceive their definition to be too simple and general in order to apply to expatriation management terms. Their definition lacks emphasis of the consequences that may result from internalizing from more than one culture. Moreover, the definition by David, Okazaki & Saw (2009) is more detailed and developed, however their definition only considers the exposure of only two cultures. This does not apply to expatriation management, as they are more likely to be exposed to more than two cultures.

Schwartz & Unger (2010) defined bicultural individuals as, “represents comfort… in [the] country or region in which one has settled” (p. 27). This definition would be suitable to expatriate studies, as these individuals need to adapt themselves to more cultures other than their own native culture. The definition covers the adaptive nature of bicultural individuals to other contexts. The definitions by Moore (2009) and Bell & Harrison (1996) also cover the important aspect of social interactions with others from different cultures. This aspect is pivotal in the investigation of expatriate effectiveness.

Evidently, the concept of biculturalism overlooks the experience of individuals that may have been influenced by more than two cultures. In other words, biculturalism does not seem fully apply to the competence and abilities that are critical to overcome the challenges encountered with expatriation. Conversely, multiculturalism embraces the community of cultural diversity and the exposure of multiple cultures (Parekh, 2000). It also refers to the fact that in a globalized society, individuals are increasingly exposed to many different cultures (Parekh, 2002). In the present study, extending upon previous definitions of biculturalism and multiculturalism altogether, we propose the following definition of multiculturalism that could be applicable to expatriation management:

(18)

18 In expatriation management, a multicultural individual is when his or her

‘personal’ culture is assimilated through the exposure to two or more cultures. He or she has the intrinsic ability to adapt to change, is open and accepting of different cultures, speaks more than one language and is inclined to adjust to foreign cultures.

This study offers the following hypothesis regarding the mediation of multicultural competence between the relationship of the biodata items (about previous international experience) and expatriation willingness.

Hypothesis 3: The positive relationship between biodata and expatriation willingness is mediated by multicultural competence.

Personal Agency

Lent, Brown & Hackett (2000) proposed that individual career interests and life goals stem partly from personal agency and partly from environmental opportunities. As mentioned earlier, personal agency, according to Bandura (2001, p. 169), is the belief that one is able to exercise control over the environment and act towards achieving desired goals and to avoid undesired ones. Bhaskar- Shrinivas, Harrison, Shaffer & Luk (2005) argued that an individual’s personal agency and motivation are vital for meeting the demands and uncertainties of moving to live in a foreign country and experience a new culture. Personal agency is a vital attribute to have for employees in order to anticipate whether or not they can successfully deal with the countless challenges of expatriation such as culture differences, living conditions and political systems (Tharenou, 2008). In the current research, personal agency is formed from two components derived from Tharenou’s (2008) research namely, self- efficacy and disruption. Tharenou (2008) stated that these components allow the exercise of forethought and self- control with respect to living abroad (p. 185). High personal agency depends on the individuals’ high self-efficacy and low disruption perception.

(19)

19 Looking closer at the components of personal agency, self-efficacy in terms of working abroad is defined by Cianni & Tharenou (2000) as: “the certainty of individuals that they are able to master the challenges related to working and living in a country with a culture different than their own” (p. 185). Individuals with high self- efficacy indicate that they are confident in the ability to handle work- related problems and challenges in a foreign country. Moreover, individuals high on self-efficacy can cope with stressful demands such as cultural adjustment and are able to control their responses in order to achieve desired outcomes (Tharenou, 2008). On the other hand, disruption entails whether an individuals anticipates whether relocating to a job overseas would disrupt routines such as work or social/ family routines. Higher levels of perceived disruption of living abroad leads to a greater uncertainty, less control over the environment and are likely to have adjustment issues (Tharenou, 2008). Therefore, anticipating low disruption and having high self-efficacy levels result in high personal agency.

Reflecting back to the biographical items, it has been mentioned that individuals who have had prior international experience will be more willing to embark on an expatriate assignment. As a result of having previous international experience, such individuals have gained self-confidence, acquired unique knowledge and skills, and are more able to anticipate and deal with the uncertainties and challenges that are associated with living abroad (Tharenou, 2008). Consequently, it can be expected that individuals with prior international experience are less likely to face problems or uncertainties abroad, and more likely to have increased feelings of control (Bhaskar- Shrinivas et al., 2005). For this reason, it may be expected that previous international experience have higher levels of personal agency (Tharenou, 2008). Thus, we can hypothesize the following:

Hypothesis 4: The positive relationship between biodata and expatriation willingness is mediated by personal agency.

(20)

20 As for career- oriented, individuals with the intrinsic motivation to advance in their careers are most likely to perceive expatriation as an instrument to reach their goal (Tharenou, 2008). If career- oriented people believe that taking on an assignment abroad would be

strategically resourceful in order to pursue their ambition, this would increase their personal agency (Tharenou, 2008). In other words, high career- oriented implies that by taking an extra step in order to achieve their goal, would contribute to the anticipation and belief that one can control unexpected events while living abroad, in turn increasing personal agency (Noe, Noe & Bachhuber, 1990). Individuals with high personal agency are confident in dealing with the uncertainties and challenges of living abroad, and are therefore more willing to take on an international assignment as the intention to reach their desired career (Tharenou, 2008).

Hypothesis 5: Personal agency mediates the relationship between career- oriented and expatriation willingness.

Family Barriers

Past studies showed that family circumstances and concerns are, amongst others, the main reasons why employees refuse to take on an international assignment (Baruch & Altman (2002; Cendant, 2004). In other words, employees who have more responsibility in their home country such as taking care of family or friends may be more reluctant to agree to an expatriate assignment, in turn having a negative influence on their willingness to expatriate (Aryee, Chay & Chew, 1996). Other studies have shown that many expatriates withdraw from completing their assignment abroad due to reasons such as the trailing spouse or family’s inability to adapt to the foreign country’s culture (Shaffer & Harrison, 1998; Johnson, Lenartowicz & Apud, 2006).

Tharenou (2008) emphasizes the relevance of family barriers in disrupting expatriation. The core components of family barriers consist of family support, whether the

(21)

21 partner or spouse has a career, and number of children (Tharenou, 2008). It is relevant to consider the expatriate’s spouse and family during the selection process as well, in recognition that expatriate employees are most likely to relocate with their trailing spouse and children (de Cieri, Cox & Fenwick, 1998). Przytula (2011) found a positive relationship between the support of the family and decision to expatriate. Individuals who are married and have children are more likely to accept an international assignment when the family supports the decision to relocate as well (Landau, Shamir & Arthur, 1992). Landau et al. (1992) concluded that having many children, negatively relates to expatriation willingness.

The current study will look at whether the role of family barriers has a moderating influence on the relationship between individuals with past international experience on expatriation willingness as well as the relationship between career- oriented and expatriation willingness in such a way, that these relationships will be weaker for those individuals with high family barriers.

H6a: Family barriers moderate the relationship between biodata and expatriation willingness, so this relationship is weaker for people with high family barriers.

H6b: Family barriers moderate the positive relationship between career- oriented and expatriation willingness, so this relationship is weaker for people with high family barriers.

(22)

22

Theoretical Model

(23)

23

Method

The previous section covered a review of past theoretical literature, leading to the formation of hypotheses and a conceptual model to demonstrate the relationship between the constructs that will be tested in this study. In the following section, the sample and data collection approach of this study will be presented. Furthermore, this section will describe the measurement scales of each construct utilized in the study.

Sampling, Population & Survey

The data collected in the current research were gathered through a self-administered web-based questionnaire created with an online survey administration software ‘Qualtrics’. Advantages of using web-based questionnaires include low cost, time efficiency and fast data representation. The entire survey was in English. The findings in this study are based on two sample groups in an effort to represent the general working population worldwide. The current study consisted of the following two sample groups: domestic employees (working and living in their home country) and expatriates (working and living in a host country). The reason to also including domestic employees is to ensure variance on each characteristic (domestic vs. expatriates) as reflected on their willingness to expatriate. In other words, studying expatriation willingness would only be valid if we also include the effects of people that are unwilling to expatriate in the sample, in this case these individuals might be the domestic employees.

The sample consists of participants from companies of various industries, such as consultancy firms, telecommunications, banking, marketing agencies and consumer goods companies worldwide. The majority of these participants were recruited through personal and social networks. In addition, also asking the participant by means of “snowball sampling”, whether he or she can recommend other potential participants to take part in the survey was

(24)

24 useful in the recruitment process. Human Resource departments from five large consumer goods companies were also contacted in order to request emails of potential candidates and to ask whether they could send out the survey to their employees internally. Subsequently, participants were also approached through online forums and web blogs such as expat-blog.com or Iamexpat.nl, personal blog pages, expat community forums, websites and expatriate Facebook groups. Around thirty-six expat bloggers have been approached after finding their contact information in their blogs and only six have agreed to take part in the survey. Invitations to join the survey were posted in twenty-two Facebook groups such as Expatica (4,206 members), Young Expats (7, 250 members) and Expats in Amsterdam (9,436 members).

All participants were sent a cover letter explaining the purpose of the research and a hyperlink to the questionnaire itself to their electronic mail addresses. The reason why electronic mail and electronic communication were utilized to recruit the participants is due to the physical and international locations of the participants. In order to increase the response rate, the invitation to the survey stated that it would take no longer than 15 minutes of their time. Furthermore, email reminders were set to also assist in increasing response rates. To encourage prompt completion of the web-based survey, the participants were offered the opportunity to draw in a raffle, where one of them won the prize of a $25 value iTunes gift card.

A total of 576 e-mails were sent to potential candidates worldwide, either living in their home country or currently living abroad. Eight survey invitations bounced back for the reason that these e-mail addresses were no longer valid or incorrect, which resulted in 568 successful invites. 364 participants started filling in the survey and partly completed the survey, whereas only 312 participants completed the whole survey. Around 72 candidates were excluded from the analysis for the reason that these did not fill in all the questions for

(25)

25 more than one construct or if their results revealed that they did not fill in the answers to the questions seriously. This was concluded when the respondents had provided the same answer throughout the whole survey, even when a certain question indicated reverse scaled items.

The sample consisted of 39.7% males (N= 116) and 60.3% females (N=176) resulting in a total of 292 respondents. The age of the respondents varied from 19 years to 65 years, where the mean age was 29 (M= 29.698, SD= 10.34). The majority (49.3%) of the respondents are well educated, as the majority of the sample possessed a Bachelor’s degree (M= 4.01, SD= 1.01). In total 39.4% of the participants are single, whereas 60.6% of the participants have a partner or a spouse. In total 23.6% revealed that the participants with a partner/ spouse, their partner’s have important careers of their own (M=2.80, SD= 1.66). Moreover, looking at whether the respondents have children, 79.8% of the sample indicated to have no children (M=1.41, SD= 0.89).

Measures

The previous section looked at the data collection and the sample of respondents that took part in the survey. The next section will look into the measures of the constructs that have been assessed in the survey. All of the items used to measure the constructs are obtained from previous research and studies. As was previously stated, the dependent variables of the current study are previous international experience and career- oriented. Multicultural competence and personal agency are the mediating variables. The moderating variable is family barriers and the outcome variable is expatriation willingness.

Dependent Variable

Expatriation Willingness

Expatriation willingness was be measured by using seven items derived from the study by Mol et al. (2009). The authors obtained two items from the studies by Van der Zee and Van

(26)

26 Oudenhoven (2000) one item from the content domains of the DOLE construct by Douthitt, Eby and Simon (1999) and lastly three items were developed by Mol et al. (2009).

The items by Van der Zee and Van Oudenhoven (2000) asked whether or not the subject considers his or herself capable enough to take on an international career. This item was assessed on a 5-point scale where: 1= Incapable, 5= Very Much Capable. The other item assessed the subjective probability of the subject to start an international career. This item was assessed on a 5-point scale where 1=Very Unlikely, 5= Very Likely.

The following item by Douthitt et al. (1999) assessed the feelings of the subject about having a job that requires travel to non- English speaking foreign countries. These items were assessed on a 5-point scale where: 1= would not like it at all and 5= would like everything about it.

Mol et al. (2009) developed the remaining three items, which were: ‘it is my goal to have the experience to live and work in a foreign country’, ‘in case I was expatriated… I would use an expatriate assignment as a means to further my career/ increase my promotional opportunities’ and ‘ I would only take on an expatriate assignment if it was forced upon me’. These items were assessed on a 5 point scale where 1= Strongly Disagree and 5= Strongly Agree.

Independent Variables

Past International Experience (Biodata)

The seven biographical items that were investigated included: a) number of countries visited on holiday, b) travel remoteness, c) foreign living experience, d) foreign travel liking e) number of friends abroad, f) number of foreign friends living in the home country and g) foreign language ability. Two items were derived from one of the DOLE experiments “Experiencing different cultures through travel” (Douthitt et al., 1999). One item measures the number of times the respondent has travelled, as rated on a 4-point scale: 1= Never, 2=

(27)

27 Once or twice, 3= Three to Five times, and 4= Five or More Times. The second item has been adapted from Mol et al. (2009), and measured how far from home the subject has traveled to on a 5-point scale: 1= Only within my country of residence, 2= Only to countries neighboring my country of residence, 3= To different regions of the continent on which I reside, 4= To one other continent and 5= To more than one continent.

Mol et al. (2009) obtained four items from another content domain namely ‘Diversity of geographic residence’ (Douthitt et al. 1999). One item inquires whether the subject has ever lived or stayed in a country for an extended amount of time where they were a member of a minority group. This item was rated on a reverse coded 2-point scale where 1= Yes and 2= No. The fourth item asked whether the subject liked being in a foreign country. This item was rated on a 6-point scale from 1= Did not like it at all, 2= Disliked it to some extent, 3= Neutral feelings about it, 4= Liked it for the most part, 5= Liked Everything About It, and 6= Never Been to Such a Foreign Country (this was scored as zero). The other two items asked whether the subject has foreign friends. These items were rated 1= No, and 2= Yes, and specify their nationalities. The numbers of nationalities they list are all summed up to one single score.

The last biodata item, was derived from the study by Van der Zee and Van Oudenhoven (2000) where the subject had to indicate their language proficiency. The 60 most common languages in the world were included, and based on each language the respondent had to indicate their proficiency per drop-down menus in speaking, comprehending, reading and writing considering the levels 1= None, 2= Elementary, 3= Intermediate, 4= Advanced, 5= Near Native and 6= Native. Per language the respondent indicated, the levels of proficiency were summed up and computed to one score.

(28)

28

Career- oriented

Career- oriented was measured by three items, which represented the likelihood of certain career outcomes such as a promotion, would occur if one moved abroad for a job (Tharenou, 2008). An example of an item included, “How likely is it that the following things would happen to you if you relocate overseas for a job? Be promoted at a faster rate/given greater advancement.” Each item was scored on a 5-point Likert scale from 1= Very Unlikely, 2= Somewhat Unlikely, 3= Neither Likely nor Unlikely, 4= Somewhat Likely and 5= Very Likely.

Mediating Variables Multiculturalism

In order to measure the participant’s multicultural capacity, they were asked to rate ten statements derived from the Multicultural Ideology Scale, where five statements are worded in a negative direction (Berry & Kalin, 1995). The scale was initially aimed to assess the attitude towards a culturally diverse society in Canada. Since this study will approach individuals with different cultural backgrounds, the items have been adapted so that the statements would apply to everyone regardless of their cultural background. An example of a positively worded item is, “It is best for the people of my society if they all forget their different cultural backgrounds as soon as possible”. An example of a negatively worded item is, “immigrants who come to live in my home country should change their behavior and be more like local citizens”. All responses were given on a 7- point Likert scale, where the answers range from 1= Strongly Disagree, 4= Neutral and 7= Strongly Agree.

Personal Agency

As mentioned earlier, personal agency looks into the levels of self-efficacy and disruption. Self- efficacy was measured with Cianni and Tharenou’s (2000) 3- item scale. This scale measured the confidence and the certainty of an individual to live and work in a country

(29)

29 different to their own (Cianni & Tharenou, 2000). An example of an item was, “I am certain of my ability to handle work-related problems when working in a country with a different culture to my own”. These items were scored on an 11-point scale ranging from 0% = High uncertainty, to 100% = Total certainty.

Secondly, the scale to measure disruption was derived from the study by Black et. al (1991). Three items measured disruption, asking the participants to what extent would relocating abroad for a job (longer than 12 months) disrupt their work routines, their non- work routines and subsequently result to cause uncertainty. These items were all rated on a 7-point scale where 1= Not At All, to 7= Always.

Moderating Variable

Family Barriers

Family Barriers was assessed based on three components: Family Involvement, Perceived Partner’s Career and Children. Two items derived from the study by Tharenou (2003), were asked to measure the family’s disapproval of the employee expatriating. When the family disapproves, the likeliness of expatriation is reduced (Tharenou, 2008). These items were, “my family has a strong influence on whether I take up international work” and, “my family’s approval and support would be important to my taking international work”. The items were both scored on a 7-point scale from 1. Strongly Disagree, to 7. Strongly Agree.

Perceived partner’s career was measured based on one item, which was “Please indicate whether you have a partner/spouse or not and if so, whether he or she has a job or career”. This item was assessed on a 5-point Likert scale, where: 1. No partner/ spouse; 2. Partner/ spouse not employed; 3. No, regards job just as a job (e.g. for income); 4. Yes, regards job partly as a career; and 5. Yes, regards job fully as a career.

(30)

30 The last component looked at whether the participant’s had any children. This was measured by asking the following item, “How many children do you have?” The participant’s rated their answer on a 5-point scale: 1. None; 2. One; 3. Two; 4. Three; and 5. Four or more.

Control Variables

Control variables were included in this study in order to eliminate any possible factors that may influence the data results. The control variables are usually held as a constant variable during the data analysis and research. The control variables in the current study are age, gender and education.

(31)

31

Results

The current section will focus on the analysis of the results that were obtained from the data collection mentioned in the previous section. Firstly, the results and description of the correlation matrix will be discussed. Secondly, the section will move along to describing the results found for each hypothesis mentioned earlier in the study.

Descriptive Statistics of the Data

The results of the correlation matrix are presented in Table 2 below. The correlation matrix presents the descriptive statistics such as information about the alpha reliability of the scales (shown along the diagonal), means, standard deviation, and correlation between different constructs. The reliability of the scales are significant when they are greater or equal to α= .7 at a 95% level. When the reliability of the scale is lower than α= .7 implies that the scale is not reliable.

In general, the alphas of all the construct scales used in this study indicate that they are reliable. Language ability is the only Biodata item measured for reliability, which resulted in a very high alpha (α=. 92). The multiculturalism scale had a very low reliability (α= .54). The low reliability may have resulted due to the lengthiness of each statement. However, looking at the scale analysis, it showed that under the ‘Cronbach’s alpha if item deleted’ column, that if the item “It is best for the people of my society if they all forget their different cultural backgrounds as soon as possible” were deleted, the alpha would increase to α=. 70. For this reason, this item has been removed from the multicultural scale. Furthermore, the reliability of the three items that measured career- oriented is moderately high (α=.72).

The two components of personal agency also showed moderately high reliabilities (α= .79), which indicates a rather high level of internal consistency. Family Barriers had the

(32)

32 reliability of (α=.62), however, by removing the item ‘please indicate whether you have a spouse and if he/she has a job or career’, the Cronbach’s alpha would raise to α= . 63. For the reason that the reliability of removing one item from Family Barriers did not greatly improve the scale, the item has therefore not been deleted. In addition, Expatriation willingness had a moderate reliability of (α=.71).

(33)

Table 2. Scale Means, SD's, Inter-item correlations and Reliabilities Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Age 29.69 10.35 (-) 2. Gender 1.61 0.49 -.161** (-) 3. Education 4.01 1.01 .031 -.027 (-) 4. Traveled foreign countries 3.87 0.47 -.055 -.047 .076 (-) 5. Travel Remoteness 4.46 0.81 .076 -.067 .007 .304** (-) 6. Minority Group 1.57 0.52 -.034 .048 .286** .192** .158** (-) 7. Experience foreign country 2.27 2.15 .025 -.003 .142** .138** .178** .818** (-) 8. Friends abroad 5.77 5.22 -.014 .148** .057 .125* .185** .229** .240** (-) 9. Foreign friends 2.27 3.14 -.04 .081 .019 .053 .132* .118* .127* .450** (-) 10. Language Ability 54.73 17.06 -.014 .115* .071 .072 .191** .066 .136* .252** .178** (.92) 11. Career- Oriented 3.84 0.76 -.064 .023 .145** -.061 .031 .126* .120* .180** .067 .142** (.72) 12. Personal Agency 4.38 1.02 -.001 .007 .106* .122* .125* .135* .224** .235** .134* .03 .142** (.79) 13. Family Barriers 3.62 1.02 -.279** -.113* .055 .020 -.068 -.017 -.009 .073 -.008 -.037 .026 .051 (.62) 14. Multiculturalism 4.38 0.59 -.001 -.117* -.059 -.048 -.046 -.021 .037 .025 -.082 -.024 .146** .087 -.098* (.70) 15. Expatriation Willingness 4.01 0.61 -.105* .117* .138** .172** .153** .283** .291** .335** .198** .201** .385** .364** .126* .086 (.71)

** Correlation is significant at the p<0.01 level (1-tailed). * Correlation is significant at the p<0.05 level (1-tailed).

(34)

Hypotheses Testing

In this section, we test and analyze the hypotheses mentioned earlier based on the data collected in order to obtain statistical inference for this research. This section will state each hypothesis, following the analytical method used for each hypothesis and discussing the outcomes.

Hypothesis 1 and 2

Hypotheses 1a – 1g: Past international experience items such as number of countries visited on holiday (a), travel remoteness (b), foreign living experience (c), foreign travel liking (d),

number of friends abroad (e), number of foreign friends living in home country (f), foreign language ability (g) are positively related to expatriation willingness.

Hypothesis 2: Career- oriented is positively related to Expatriation Willingness.

In order to test these two hypotheses, the correlation matrix was used and a linear regression analysis was conducted. The results of the linear regression analysis are presented in table 1 (appendix I) and the results of the correlation matrix are presented in Table 2 (above). The results show the relationship between each predictor with expatriation willingness. The variables age, gender and education were entered as control variables.

By looking at the correlation matrix, it is evident that all biodata items, are all significantly related to expatriation willingness: traveled to foreign countries (r=.17, p <.05), number of countries traveled (r=.15, p <.05), foreign living experience (r=.28, p <0.05), foreign travel liking (r=.29, p <.05), number of friends abroad (r=.34, p < .05), number of foreign friends living in home country (r=.20, p < .05) and language ability (r=.20, p < .05). Moreover, by looking at the correlation between career- oriented and expatriation willingness is also significant (r=.39, p <.05).

(35)

35 The linear regression results (Appendix I- table 1) shows two models. The first model only included the control variables age, gender and education. The second model shows the results of relationship between the seven biodata items and career- oriented. The F value of the second model is significant, which indicates that the model makes sense. By looking at the table, it is evident that the biodata items are all related to expatriation willingness: number of countries visited (B= .213, p<0.01), travel remoteness (B= .126, p<0.01), foreign living experience (B=.301, p<0.01), foreign travel liking (B=.079, p<0.01), friends abroad (B= .037, p<0.01), foreign friends (B= .036, p<0.01) and foreign language ability (B= .010, p<.01). Career- Oriented was also found to be related to expatriation willingness (B= .293, p<.0.01). The R-square of the model is .54, which indicates that the variables in model 2 explain 54% of the variance in expatriation willingness. These results support hypothesis 1a- 1g and hypothesis 2.

Hypothesis 3, 4 and 5

Hypothesis 3: The positive relationship between biodata and expatriation willingness is mediated by multicultural competence.

Hypothesis 4: The positive relationship between biodata and expatriation willingness is mediated by personal agency.

Hypothesis 5: The positive relationship between career- oriented and expatriation willingness is mediated by personal agency.

(36)

36 Hypotheses 3, 4 and 5 have been tested with the use of the PROCESS tool developed by Hayes (2012). Since these hypotheses inquire indirect effects, model 4 of the PROCESS tool will be used and bootstrapped 1000 times at a 95% confidence interval (Hayes, 2012).

The simple mediation model has been used in this research in order to find out to what extent Biodata (x) and career- oriented (x) influence expatriation willingness (y) through the mediating variables (m) multicultural competence and personal agency. The results of the mediation analyses are presented in Appendix I – table 2. Regression 1 on the table shows whether the independent variable (x) is related to the mediating variable (m). Regression 2 shows the direct effect of the independent variable (x) with the dependent variable (y) and the relationship of the mediating variable (m) with the dependent variable (y). The column next to regression 2 shows the total, direct and the indirect effects.

Because only one independent variable may be used at a time, the other independent variables were inserted as covariates. This was done for each one IV at a time. This analysis does not produce a single estimate of the total indirect effect throughout for all of the independent variables, but the estimates and tests for each of the independent variables (Hayes, 2012). In addition, the control variables age, gender and education were also added as covariates.

In order to establish a full mediation effect, the direct effect of the independent variable (x) on the dependent variable (y) becomes non-significant, whereas the indirect effect

(37)

37 of the mediator (m) of the relationship becomes significant (Hayes, 2012). Moreover, partial mediation occurs when the pathway shows that the total effect, direct effect and indirect effect are all significant

All 15 pathways are presented in the table (Table 2- Appendix I) for all three hypotheses. The biodata items are listed under the independent variables (x) from a – g. Hypothesis 3 entails the indirect effect of biodata items (x) and expatriation willingness (y) via of multicultural competence (m) By looking at the table, none of the pathways shows any significant results of the indirect effect. The total effects of the model and direct effects all show significant results. However, the indirect effects are all insignificant because zero is included between the Lower (LLCI) endpoints of the confidence interval and the Upper levels (ULCI) of the confidence interval. Thus, contrary to our expectations, hypothesis 3 is rejected.

Hypothesis 4 entails the indirect effect of biodata items (x) on expatriation willingness (y) via personal agency (m). By looking at the table, only two pathways provided significant results. Firstly, the relationship between travel remoteness and personal agency is significant (Beta= .18, p <. 01), and personal agency is also positively related to expatriation willingness (Beta= .12, p<. 01). Furthermore, the total effect and direct effect are also significant which indicates that the model is partially mediated (Effect= .02, p<. 05). The analysis also reveals that the indirect effect is significant as the LLCI and ULCI of this pathway excludes zero. Secondly, foreign language ability is significantly related to personal agency (Beta= .01, p<. 05), and personal agency is related to expatriation willingness (Beta= .12, p<. 01). This pathway indicates a partially mediated effect (Effect=. 01, p<. 05) for the reason that the total and direct effects are also significant. Moreover, the LLCI and ULCI excludes zero, which indicates a significant indirect effect. These results partially support hypothesis 4.

(38)

38 Hypothesis 5 entails the indirect effect of career- oriented (x) on expatriation willingness (y) via personal agency (m). This pathway does not show a significant indirect effect, since the LLCI and ULCI includes zero. As expected from the correlation matrix (r= .39, p<. 01), career- oriented is related to expatriation willingness (Beta= .33, p<. 01). Altogether, these results do not support hypothesis 5.

Hypothesis 6

H6a: Family barriers moderate the relationship between biodata and expatriation willingness, so this relationship will be weaker for people with high family barriers.

H6b: Family barriers moderate the positive relationship between career- oriented and expatriation willingness, so this relationship will be weaker for people with high family

barriers.

Hypotheses 6a and 6b were also tested with the use of the PROCESS tool (Hayes, 2012). Since these hypotheses require moderation effects, model 1 of the PROCESS tool will be used and bootstrapped 1000 times at a 95% confidence interval (Hayes, 2012). The control variables age, gender and education are also included in the analyses.

The results of hypotheses 6a and 6b are presented in Appendix I table 2. The simple mediation model has been used in this research to test the effect of Family barriers as the moderating variable (m) on the relationship between the biodata items (x) and expatriation

Predictors affecting expatriation willingness family barriers to expatriation?