HRMN 467 Week 3 - Expatriate adjustment and Global Compensation

Work Role Transitions: A Study of American Expatriate Managers in Japan Author(s): J. Stewart Black Source: Journal of International Business Studies, Vol. 19, No. 2 (Summer, 1988), pp. 277- 294 Published by: Palgrave Macmillan Journals Stable URL: http://www.jstor.org/stable/155026 Accessed: 21-09-2017 23:52 UTC

REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/155026?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

http://about.jstor.org/terms

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access to Journal of International Business Studies

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS: A STUDY OF AMERICAN EXPATRIATE MANAGERS IN JAPAN

J. Stewart Black*

University of California, Irvine

Abstract. Throughout the course of a career, an individual must

make numerous role transitions, instigated through such events as overseas transfers, domestic transfers, promotions, company reorganizations, and inter-company job changes. This paper examines the relationships between several variables and work

role transition in the case of an overseas assignment to Japan. Role ambiguity and role discretion were found to influence work

adjustment, while predeparture knowledge, association with local nationals, and family's adjustment were found to correlate with

general adjustment of American expatriate managers in Japan.

Although the topics of roles and role adjustment have been investigated for several years, only recently has a theory of work role transitions been presented [Nicholson 1984]. Work role transitions can be instigated through overseas transfers, domestic transfers, promotions, company reorganizations, and inter- company job changes. This study investigates work role transitions brought on by foreign assignments. This particular type of work role transition was selected because the growing internationalization of business requires increasing amounts of interaction between companies and business people of different countries and cultures. To implement this internationalization, often nationals of one country are sent to subsidiaries in foreign countries. Studies [Baker and Ivancevich 1971; Tung 1981] have found that between 20 to 40% of the expatriate managers (EXM) do not successfully make the transition and return early. These figures become larger if one includes the "brownouts" [Lanier 1979] or those who may not return early but perform at a decreased capacity by not being able to adjust adequately to their new work roles. In addition, of EXMs sent to Japan in the past, one source [Seward 1975] found that 90% of the EXMs were significantly less successful in Japan than they were in their previous assignment in their home countries. Another source [Adams and Kobayashi 1969] found that 80% of the EXMs in Japan were considered failures by their headquarters. These problems are significant to corporations because the associated costs are high, ranging from $50,000 to $150,000 per

* J. Stewart Black is currently a Ph.D. student at the University of California, Irvine, in the Graduate School of Management. While there, he was awarded a Regent's Fellowship for 1985-1986. He received his B.S. and Master of Organizational Behavior degrees from Brigham Young University, where he graduated with University Honors and Distinction, respectively.

Received: July 1987; Revised: October & December 1987; Accepted: December 1987.

277

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

278 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

person if the EXM returns from the international assignment early (Harris 1979). For companies with hundreds of expatriates, the total costs can easily reach several million dollars. Given the importance of such assignments, it is

useful to try to gain an understanding of how these types of work role transitions

are made and what factors influence their success and failure.

This study begins with an examination of the relevant theoretical and empirical literature. Based on this material, several hypotheses are presented and the

results of a survey of American expatriate managers making transitions to their

overseas assignments in Japan are presented. Finally, the theoretical and managerial implications are explored.

DIMENSIONS OF ADJUSTMENT IN WORK ROLE TRANSITIONS

Adjustment to the new work role can be viewed as having three primary dimensions: degree, mode, and facet. Each of these dimensions are briefly discussed.

Degree of Adjustment

Degree of adjustment can be viewed as both a subjective and objective concept.

Subjectively, it is the degree of comfort the incumbent feels in the new role

and the degree to which he or she feels adjusted to the role requirements.

Objectively, it is the degree to which the person has mastered the role

requirements and is able to demonstrate that adjustment via his or her

performance. In the past, scholars have largely relied on self-report responses to measure degree of felt adjustment [Torbion 1982] or to measure other degree

indicators such as time to proficiency [Pinder and Schroeder 1987].

In the case of managers sent on foreign assignments, degree of adjustment

has been measured through self-reports of adjustment to the new country and

culture [Gullahorn and Gullahorn 1962; Abe and Wiseman 1983; Torbion 1982]. An important aspect of the degree of adjustment to the culture is the stage of adjustment. Scholars [Lysgaad 1955; Oberg 1960; Torbion 1982] have argued that adjustment occurred in four phases often referred to as the

U-curve. The first phase occurs during the first few weeks after arrival. At this time the new arrival is fascinated with the new and different aspects of the foreign culture and country. Some have referred to this as the "honeymoon stage." Torbion [1982] adds that during the initial stage, the person has not had sufficient time and experience in the new country to discover that many of his or her past habits and behaviors are inappropriate in the new culture. This lack of negative feedback and the newness of the foreign culture combine to produce the honeymoon effect.

Once the newcomer begins to cope seriously with the real conditions of everyday life, the second stage begins. This stage is characterized by frustration and hostility toward the host country and its people. This is because the person discovers that her or his past behaviors are inappropriate in the new culture but as yet has not learned what to substitute in their stead [Torbion 1982]. In general, culture shock occurs at the transition between stage two and three

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 279

when the person has received the maximum amount of negative feedback but as yet has very little idea as to what the appropriate behaviors are.

The third stage begins as the individual acquires some language skills and

ability to move around on his or her own. In the third stage the person begins to learn not only how to "get around" but also some new appropriate behaviors. By the third stage the individual also has developed some proficiency in performing the new set of behaviors.

In the fourth stage, the individual's adjustment is generally complete and the incremental degree of adjustment is minimal. In this stage, the individual now knows and can properly perform the necessary behaviors to function effectively and without anxiety due to cultural differences.

Mode of Adjustment

Mode of adjustment involves the manner in which the individual adjusts to the new role. Several scholars [Dawis and Lofquist 1984; Feldman and Brett 1983; Nicholson 1984; Van Maanen and Schein 1979] have essentially argued that the individuals can adjust by altering the new role to match better themselves or by altering their own attitudes and behaviors to match better the role expectations. Nicholson [1984] expands these two dimensions into a four cell matrix, which includes all the combinations of the two basic modes of adjustment. The first mode of adjustment Nicholson calls replication. When confronted with a new work role, a person using this adjustment strategy would make few adjustments in his or her identity or behaviors in order to fit into the new role. Also, the person would make few changes in the role. The second mode of adjustment is termed absorption. When confronted with a new role, a person using this adjustment strategy would make very few if any modifications in the role and would instead modify his or her own behavior and attitudes to fit the role requirements. The third mode of adjustment is termed determination. According to Nicholson, this mode of adjustment represents those instances in which the incumbent's adjustment to the demands of the role transition leaves the person relatively unaffected but alters the new role. The fourth mode of transition is termed exploration. This mode of transition represents those cases in which the person makes adjustments in his or herself and in the new role.

Facet of Adjustment

Although the primary focus of the literature on work role adjustment concerns adjustment to the "work role," in the case of overseas transfers, another facet of adjustment involves the manager's adjustment to the new customs and culture of the host country [Hawes and Kealey 1981; Ross 1985; Torbion 1982]. Even in the case of domestic transfers, there exists some theoretical and empirical support for the inclusion of adjustment to outside work factors as another facet of adjustment [Feldman 1976]. Thus, it seems that at a minimum there are at least two facets of adjustment: work adjustment and general adjustment.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

280 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

FACTORS HYPOTHESIZED TO INFLUENCE ADJUSTMENT

Because much of the theoretical writing on work role transitions is fairly recent (for example, see Nicholson 1984), much of the empirical literature in the next section has been approached from somewhat of an atheoretical perspective.

However, using the theoretical frameworks of Nicholson [1984], Dawis &

Lofquist [1984] and Feldman [1976], the empirical literature has been organized into two broad categories: adjustment inhibiting and adjustment facilitating

factors. These are then subdivided into three subcategories: individual factors, job-related factors, and outside factors.

Adjustment Facilitating Individual Factors

Several individual factors have been hypothesized and found to be important in expatriates' adjustment to new work roles during overseas assignments. In

formulating the hypotheses for this cross-sectional study, individual factors were not included. They were excluded because theoretically it is just as likely

that adjustment causes changes in the individual as it is that individual differences

influence adjustment. Although studies have found correlations between various individual factors and adjustment, a longitudinal design would be needed to provide a meaningful contribution to understanding the causal relationship

between any of the individual variables identified and adjustment. Consequently, they are not discussed in detail (for a review see Mendenhall and Oddou 1985). A summary of the individual factors that seem to facilitate adjustment includes

(1) the individual's desire to adjust [Brim 1966; Cogswell 1968; Howard 1974; Mortimer and Simmons 1978; Tung 1981], (2) technical or managerial competence [Bardo and Bardo 1980; Brim 1966; Harris 1973; Hawes and Kealey 1981; Hays 1971; Tung 1981], (3) a person's social relation skills orientation [Abe and Wiseman 1983; Hammer, Gudykunst and Wiseman 1978; Harris 1973; Hawes and Kealey 1981; Hays 1971; Louis, Posner and Powell 1983; Ratiu 1983; Ross 1985], (4) an individual's tolerance for ambiguity or "open mindedness" [Detweiler 1975; Gudykunst; Wiseman and Hammer 1977; Kahn 1964; Ratiu 1983; Ruben 1976; Ruben and Kealey 1979], (5) an individual's self-confidence [Kahn 1964; Fisher 1982; and Jones 1986], (6) met expectations [Dunnette, Arvey, and Banas 1973; Feldman 1976; Ilgen 1975; Toffler 1981], and (7) reinforcement substitution [Bren and David, 1971; David 1976; Mumford 1975; Tung 1981].

Adjustment Inhibiting Job Factors

Several job-related factors have been found to increase the uncertainty, unfamiliarity, unpredictability, or uncontrollability of the new work role and consequently inhibit the adjustment. These include role novelty, role ambiguity, role conflict, and role overload.

Role novelty really involves the difference between the past role and the new one. Role novelty essentially increases the degree of unfamiliarity with the new role, which likely decreases the degree of predictability. Pinder and Schroeder [1987] have found the greater the difference, the longer it takes the

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 281

person to reach a level of proficiency after a domestic transfer. Other researchers

have found a similar relationship between role novelty and adjustment [Burr 1972; George 1980; Minkler and Biller 1979; Pinder and Schroeder 1987; Sarbin and Allen 1968].

Hl: The higher the role novelty, the lower the adjustment to work responsibilities.

Most major role transitions have some associated role ambiguity. The greater the role ambiguity, the less the individual is able to predict the outcome of

various behaviors and the less the individual is able to utilize past successful

or determine appropriate new behaviors. Several studies have found the higher the associated ambiguity, the more difficult the transition [Harvey 1982; Misa and Fabricatore 1979; Pinder and Schroeder 1987].

H2: The greater the role ambiguity, the less the degree of adjustment to the specific job responsibilities.

Additionally, often individuals in new roles experience conflicting signals about what is expected of them. When an individual experiences conflicting messages about expected behaviors, he or she is less able to determine which messages to ignore and which to follow and thereby execute the appropriate behaviors. Not surprisingly, researchers have found the greater the role conflict,

the greater the difficulty of the role transition (Kahn et al. 1964).

H3: The greater the role conflict, the less the degree of adjustment to the specific job responsibilities.

If managers have too many demands placed upon them, they will be less able to respond adequately to the demands. Thus, role overload will have a negative influence on work role adjustment. Role overload has been found to be negatively associated with successful role transitions [Kahn 1964; Karasek 1979; Tung 1982].

H4: The greater the role overload, the less the degree of adjustment to the

specific job responsibilities.

Adjustment Facilitating Job Factors

In addition to job factors that increase uncertainty, several job-related factors have the potential for reducing it and facilitating adjustment. These include the role discretion, previous transfers, and pre-departure knowledge.

Nicholson [1984] along with Dawis and Lofquist [1984] includes the notion that the greater the role flexibility or role discretion, the easier the adjustment. Some studies have found that role discretion enables individuals to more easily utilize successful past behaviors in the new role by having the freedom to modify the role to fit their abilities and habits and thus makes the role more familiar, predictable, and controllable and make the transition easier [Karasek 1979; Kahn et al. 1964].

H5: The greater the increase in role discretion, the greater the adjustment.

In the case of domestic or international transfers and the resulting role transitions, individuals with previous transfer experiences might be able to extrapolate

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

282 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

from these transfers and thus be more familiar with aspects of the new situation

and be better at predicting what to expect with the transfer [Louis 1980].

Despite the theoretical appeal of this variable, it has yet to be supported

empirically [Pinder and Das 1979; Pinder and Schroeder 1987].

H6: The greater the previous overseas work experience, the greater the

adjustment.

Because pre-departure knowledge has the potential of providing the individual with information about the impending transition, it can reduce the uncertainty

by increasing the predictability of the new situation as well as by increasing at least the individual's anticipatory familiarity. The fact that a majority of firms do not provide any training to facilitate pre-departure knowledge of the

target country perhaps indicates that they believe it does not facilitate adjustment [Baker and Ivancevich 1971; Tung 1981]. Some empirical work supports the assertion that pre-departure knowledge does facilitate adjustment [Tung 1984].

H7: The more knowledge the manager has about the target country of the overseas assignment prior to departure, the better will be the manager's

adjustment.

Adjustment Facilitating Outside Factors

When the work role transition requires a major change for the individual's

family, the family's ability to adjust to the changes has a significant impact on the individual's transition at work. Especially in the case of overseas assignments, an individual's family's inability to adjust is the biggest reason for the individual's inability to make the transition [Hays 1971; Misa and Fabricatore 1979; Torbion 1982; Tung 1981; Tung 1982]. However, it is possible that the causal direction is revised. It is possible that the EXM's adjustment has an influence on the family's adjustment. Consequently, only a correlation hypothesis can be made.

H8: The family's general adjustment will be positively correlated with the expatriate's adjustment.

Because association with host nationals can provide cues concerning appropriate behavior in the new situation, greater association with host nationals would reduce novelty and positively affect adjustment. However, greater association with host nationals will not necessarily provide information of how to perform specific job responsibilities. Thus, while association with host nationals likely will be associated with general adjustment, it likely is not associated with work adjustment. Unfortunately, in a non-longitudinal study, it is impossible to determine whether the greater interaction with host nationals leads to greater

adjustment or whether greater adjustment leads to increased association with host nationals. Thus, only a correlation hypothesis can be made. Nevertheless, this source of novelty reduction has been under-addressed and is worth examining even in a correlation context.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 283

H9: Interaction with host nationals will be positively correlated with

general adjustment.

METHODOLOGY

Sample

The sample for this study was drawn from those Americans registered with the American Chamber of Commerce in Japan. Although this registry is not exhaustive, the directory contains over 1,400 names and represents virtually all American companies operating in Japan. A sample of 195 individuals was randomly drawn by selecting every seventh individual in the alphabetized listing.

Of the 195 questionnaires sent, 77 were returned for a response rate of 40%. Sixty-seven of the returned questionnaires were usable. Given that limited funds prevented the sending of followup letters or questionnaires, the response rate was typical [Dillman 1978]. The respondents represented 25 of 33 different industries and were an adequate representation of the sample. All of the respondents were male and 80% were married. On average the respondents were 46 years old and had been in the firm for 14.5 years. In addition, 33.8% of the respondents held top management positions, 49.2% were department heads, 15.4% were middle managers, and 1.5% were technical specialists.

Definition and Operationalization of Variables

As discussed earlier, two methods can be utilized to measure adjustment. The first is the individual's assessment of how comfortable or adjusted he or she

feels. The second is an independent measure of the individual's adjustment or performance. Because response rates to mailed questionnaires which ask for

an independent measure of adjustment such as supervisor rating are usually unacceptably low [Dillman 1978] and lack of resources prohibited more direct methods of obtaining an independent measure of adjustment such as supervisor evaluation, only self-reported adjustment was measured.

For expatriate managers, it is hypothesized that there are three facets of

adjustment. It seems, expatriate managers adjust to (1) work roles, (2) interacting with host nationals (Japanese nationals in this study), and (3) the general culture and everyday life. An eleven-item scale was developed to measure these three facets of adjustment. The six items measuring adjustment to everyday life were based on the scales developed by Torbion [1982]. To measure work adjustment, respondents were asked to indicate the degree of adjustment they felt with their job and responsibilities, with interacting with Japanese peers and subordinates. (Adjustment to Japanese superiors was not included because preliminary interviews suggested that very few American expatriate managers had Japanese superiors.) To measure adjustment to interacting with Japanese in general, respondents were asked to indicate their degree of adjustment to

working with Japanese outside their company and to interacting with Japanese in general, everyday situations.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

284 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

TABLE 1 Factor Analysis of Adjustment

ITEM General Interaction Work

General living .90 Transportation .76 Food .75 Shopping .65 Weather .57 Entertainment .55

Working with Japanese outside .89 Interacting with Japanese general .38 .84 Job responsibility .77 Japanese co-workers .37 .47 .63 Japanese subordinate .32 .61

Only loadings greater than .30 are reported.

The eleven-item scale was factor analyzed using a principal component factor analysis procedure with a varimax rotation of factors. It was hypothesized that there were three factors within these eleven items, and three factors emerged with eigen values greater than one (see Table 1). These three factors explained 62.9% of the variance in the eleven-item set.

All the items in Factor 1 loaded strongly (above .50) on the first factor and below .30 on the other two factors. A reliability test of these six items in Factor 1 produced an alpha of .80, which is generally considered acceptable (Churchill and Peter 1984). This first factor of adjustment included adjustment to general living conditions, transportation system, food, shopping, weather, and entertainment and was termed general adjustment or adjustment to general living conditions and everyday life.

The two items of working with Japanese outside the company and interacting with Japanese in general loaded very strongly on the second factor (above .80). The items in this second factor had a relatively high reliability coefficient (alpha = .83). The factor essentially involved adjustment to interacting with Japanese.

The three remaining items of adjustment to interacting with Japanese peers and with Japanese subordinates, and to job responsibilities loaded on Factor 3. However, these three items had an unacceptably low reliability coefficient (alpha = .30). The items of interacting with Japanese peers and subordinates had complex loadings. They loaded above .30 on the second factor (interacting with Japanese) and above .60 on the third factor (adjustment to work). The item of adjustment to job responsibilities loaded negatively on the first and second factor and strongly (.77) on the third factor. The primary component of the third factor seemed to be adjustment to work responsibilities. Removing the items of adjustment to interacting with Japanese peers and subordinates from the third factor left a one-item measure of adjustment to work.

This procedure produced three factors or facets of adjustment. The first two factors (adjustment to general conditions, and adjustment to interacting with

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 285

Japanese in general) were very clear and strong; however, the two factors were significantly correlated (.47, p < .01). Therefore, they were not considered to be independent facets of adjustment. The eight items relating to these two factors were combined to form a scale of general adjustment (alpha = .83) The third factor (adjustment to job responsibilities) was conceptually different from and not significantly related to the factor of general adjustment (see Table 3).

Role novelty was defined as the degree to which the expected patterns of behaviors in the new role were different from those of past roles. Based on Stewart's [1982] categories of managerial demands and constraints, eleven items were used to measure role novelty. Respondents were asked to indicate

how similar or different each of the items was compared to their previous domestic assignment. A factor analysis of these eleven items produced four factors that accounted for 66.9% of the variance in the eleven-item set. Factor 1 involved the novelty of performance standards, degree of involvement in work unit, and outside work responsibilities. The second factor involved the novelty of the type of people in the work unit, the legal and governmental regulations, and employees resistance to change. The third factor consisted of the novelty of bureaucratic procedures and mandatory meetings. The fourth factor consisted of novelty of work relations, resource limitations, and technical limitations. Although the factor analysis produced four factors of role novelty, the four factors were highly intercorrelated (approximately r = .50, p < .001). Consequently, all eleven items were combined into one scale of role novelty which had an acceptable reliability (alpha = .78).

Role ambiguity was defined in terms of the existence or clarity of behavioral requirements [Rizzo, House and Lirtzman 1970]. The scale for measuring role ambiguity was taken from Rizzo, House and Lirtzman [1970], which has been used in other similar studies [Jones 1986], because of the objective wording of the items (for example, "clear, planned goals exist for my job"). The reliability of this scale was adequate (alpha = .76).

Role conflict is the congruency-incongruency or compatibility-incompatibility in the requirements of the role, where congruency or compatibility is judged relative to a set of standards or conditions which impinge upon role performance.

This scale was also taken from Rizzo, House and Lirtzman [1970]. This scale had moderately high reliability (alpha = .82).

Role overload was defined as the excessive demands placed upon the occupant of a particular role. The scale for measuring role overload was based on Kahn's [1964] description of role overload and included three items: (1) excessive work load, (2) excessive time demands, and (3) insufficient time to complete work. These three items had adequate reliability (alpha = .81).

Role discretion was defined as the individual's opportunities to alter the components and relationships of role demands. As a means of approximating this variable and assessing its impact on work role adjustment, respondents were asked on a one-item, Likert-type scale whether the degree of autonomy they currently had was more or less than that which they had in their previous assignment.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

286 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

TABLE 2 Factor Analysis of Family Adjustment

Item Factor 1 Factor 2

Schools .92 Transportation .66 .44

Shopping .82 Weather .43 Interacting w/Japanese .87 Entertainment .79

Only loadings greater than .30 are reported.

Respondents indicated on a five-point Likert scale the degree of knowledge they had of Japan (language, culture, customs, political system, etc.) before the transfer. A factor analysis of this five-item scale was conducted and produced

one factor with moderately high loadings (above .80). The inter-item reliability of this scale of knowledge before departure was very high (alpha = .91).

Previous international transfers were operationalized as the number of years

the respondent had worked in overseas or foreign assignments.

Outside Factors

The outside factor measured in this study was family adjustment. Using a Likert-type scale, respondents were asked to rate their family's adjustment to schools, transportation system, shopping, weather, interacting with Japanese in general, and entertainment. A factor analysis of these items produced two factors (see Table 2). The first factor consisted of adjustment to schools and the transportation system and had an unacceptably low reliability (.40). The second factor included adjustment to shopping, weather, entertainment, and interacting with Japanese in general and had an adequate reliability coefficient (.75). Thus, only the items in the second factor were used as the scale of

measuring family adjustment.

Association with host nationals was operationalized as the percent of time during work and non-work hours that the expatriate manager spent with host nationals.

RESULTS

Adjustment Inhibiting Job Factors

Hypotheses 1 through 4 predicted that role novelty, role ambiguity, role conflict, and role overload would have a negative impact on work adjustment. The formulation of these hypotheses assumes that these four job factors will not have any significant relationship with general adjustment. This assumption was supported by the data (see Table 3). To test the influence of these job factors on work adjustment, a multiple regression was run by regressing work adjustment on role novelty, role ambiguity, role conflict, and role overload. Although the independent variables accounted for a significant amount of the

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 287

C\j 0

o "t

0 l- CC) o C)0

oC) C o C\!

CC) O~~~~~~~~~~~~ cr) 0,

0~~~~~ .0 0)

a CY ) \L .X~~~~~~~~~~C 00 ? ?? 0\ 0

C\j~~~~~~\

< S ~0 00 0 0 0 LO

0 C?) O I) CO t L ) 0 JU) nt a~OCY)C\] C0 ) 0 0

- k

0 C~~~~~) ~~00 CY) C] Y C \ ICY) '- LO) C

o O- LO 0,\ C\) C0 0

> _s

U CY) 9 O 9 1 C\ ) L CYO CY) U) ,- a o C CY? 0 C) CO C\)

0~~

C O IO) C(O. CO t) 0 0 NIC) co N-00 -0 O)O) C1) ' cO . . . C' O CY ) CO O CM c C\ O

C? CY? ~~~~~~CY?J C'J C ~ C\

CC\ No L c - C)N YC\I) 0\ 0 CY co CY) (00 N- CY)(OC\OO0\]" C\O 0- 0 0

co LO C\] (o o ')t cy r. o o rl LO r o r- o L O CY CY) C) C\ O

4- C\i~~~~~N C1)6

C/) C\? C\ )

C C\] rl C\] LO - C\ LOf ) O cO C\]C (1 C O - LO D CD) o 0 0 0 co O CD O ( 6 CD s C\i 4 CY)' CY54 CY5 CD cy L6 o6

4> Z3 -4 (D0 cm ? ??????o o 10 CM CrLD. CD o ?? C C CM

(1 ( ) t 0 c

0)0)0) E 99~E 3 L -0

C\] li t6 6 r<c6 & 6 C\'

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

288 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

TABLE 4

Results of Multiple Regression Adjustment Facilitating Job Factors

Independent Variables Betas Values for t

Previous Overseas Experience .33 2.86* Pre-departure Knowledge -.31 -2.69* Role Discretion .31 2.62*

R square = .27; F = 7.14; p < .001; Dependent Variable: Work Adjustment

* p <.01

Independent Variables Betas Values for t

Previous Overseas Experience .18 1.50 Pre-departure Knowledge .38 3.14* Role Discretion -.01 -.07

R square = .15; F = 3.29; p<.05; Dependent Variable: General Adjustment

*p<.01

variance in work adjustment (R square = .19, F = 2.94, p < .05), only role ambiguity had a significant impact on work adjustment (beta .44, p < .005).

Adjustment Facilitating Job Factors

Hypotheses 5 through 7 predicted that role discretion, previous overseas transfers, and pre-departure knowledge would facilitate adjustment. In general, it remained an empirical question as to whether the three factors would have a positive influence on both facets of adjustment. To test the impact of these variables

on adjustment, two separate multiple regression equations were conducted. In the first equation, work adjustment was regressed on role discretion, pre- departure knowledge, and previous work experience. These three factors explained a significant amount of the variance in work adjustment (see Table

4). As predicted both role discretion and previous overseas work experience had a significant and positive impact on work adjustment. However, pre-

departure knowledge had a significant and negative impact on work adjustment. This unexpected finding is discussed later.

The second regression equation examined the impact of these same three independent variables on general adjustment. Again the three independent variables accounted for a significant portion of the variance in general adjustment

(see Table 4). However, in the case of this facet of adjustment, only pre- departure knowledge had a significant impact.

Adjustment Facilitating Outside Factors

Hypotheses 8 and 9 asserted a positive correlation between family's adjustment, association with host nationals and the EXM's adjustment. Although researchers [Hays 1971; Tung 1981] have found the family's inability to adjust as well as the EXM's inability to adjust were related to early returns, it remained an empirical question of how family adjustment related to different facets of EXM adjustment. In testing these hypotheses, only correlational analysis was used. Theoretically it is just as likely that adjustment of the EXM leads to family

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 289

adjustment as the reverse. Likewise, it is equally possible that adjustment leads to greater association with host nationals or that association with host nationals facilitates adjustment. Torbion [1982] has argued that both family adjustment and association with host nationals leads to adjustment; however, neither the design of his study nor this study provide the means of testing the causal relationship. Hypothesis 8 asserted that association with host nationals during work and non-work hours would be positively correlated with adjustment. Time spent with host nationals during work and non-work hours was positively, and significantly correlated with general adjustment (see Table 3) but not with work adjustment. Likewise, hypothesis 9 predicted that family adjustment would be positively correlated with EXM adjustment. Family's adjustment was positively and significantly correlated with general adjustment (.85, p < .001) but not with work adjustment.

DISCUSSION

The results of this study lend some support to the theoretical argument that there are at least two distinct facets of adjustment. This is further supported by the fact that work adjustment is related to role ambiguity and role discretion, while general adjustment is related to pre-departure knowledge, to association with host nationals, and to family's adjustment. This study indicates that past practices of thinking and measuring adjustment particularly in overseas assignment as a generic or unitary phenomenon perhaps masks the different impacts of variables on adjustment. Much more research needs to take place to examine the various dimensions of the work roles, related outside roles and the adjustment process.

This study found no relationship between role novelty and work adjustment. In their study of domestic transfers, Pinder and Schroeder [1987] did find a relationship between role novelty and self-reported time to proficiency. Two methodological considerations may account for the different findings. First, Pinder and Schroeder measured work adjustment by asking respondents to indicate how many months after the transfer it took them to become effective in their job. This study asked respondents to indicate the degree of adjustment they felt concerning their job responsibilities. Second, Pinder and Schroeder used a one-item measure of role novelty ("overall how similar is your current job to the one before the transfer"). This study used an eleven-item scale to measure role novelty. It may be that when asked to rate role novelty overall managers incorporate aspects and weightings not included in the multiple measure. The nature of the work role transitions may also be an important explanation. It may be that in an international transfer so much is novel (the job, the people, the culture, etc.) that the impact of role novelty is diluted. Future studies might address both role novelty and culture novelty to test this possibility. In addition, future studies might rely less on subjective measures of role novelty. Possible method covariation problems could be reduced in future studies by using objective proxies of role novelty such as actual changes in functional area (for example moving from sales to production).

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

290 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

In addition to limitations concerning conclusions about role novelty, some

limitations exist regarding conclusions about role ambiguity as well. Although effort was made to use objectively worded items to measure role ambiguity,

the relationship found between role ambiguity and work adjustment might be

the result of method covariation. However, items relating to role ambiguity

and work adjustment were placed in separate parts of the questionnaire in

order to reduce this limitation. In summary, of the adjustment inhibiting job factors, only role ambiguity was significantly related to work adjustment, and though steps were taken to reduce method covariation problems, this finding should be viewed with some caution.

Concerning adjustment facilitating job factors, as hypothesized, previous overseas work experience and role discretion were related to work adjustment. However,

neither were related to general adjustment. The lack of significant relationship

between previous overseas work experience and general adjustment conditions

and to interacting with host nationals perhaps suggests that many aspects of

overseas assignments are not readily generalizable from one overseas assignment to another or that EXMs are not able to transfer or extrapolate their learnings concerning adjustment in one country to another.

Pre-departure knowledge did have the expected significant and positive relationship with general adjustment, but it had an unexpected significant and

negative relationship with work adjustment. Recent follow-up interviews suggest

one possible explanation for this unexpected relationship. Several EXMs in

Japan suggested that even though they believed pre-departure knowledge facilitated work adjustment, those that had pre-departure knowledge concerning the country and culture would report a lower level of work adjustment because

they were more aware of the cultural elements in the job responsibilities than those who had a low level of pre-departure knowledge. In other words, they suggested that those with little knowledge could not ignore and had to be aware of the negative impact of lack of knowledge on general adjustment (i.e., living in Japan and interacting with Japanese), but because certain elements of the job were similar to those in the U.S., they could pay attention to more familiar aspects and ignore or just not be aware of those aspects in which culture had an impact on job responsibilities. Consequently, they would report higher levels of work adjustment than those made more aware because of their pre-departure knowledge.

Although not investigated at the time, another possible explanation for the positive relationship between pre-departure knowledge and general adjustment and the negative relationship between pre-departure knowledge and work adjustment is that the pre-departure knowledge was more accurate and relevant to living in Japan and was less accurate and relevant to working in Japan. This possibility would suggest that future research might explore the exact content of the pre-departure knowledge EXMs have to determine relevant information concerning living in Japan versus working in Japan. Since this study suggests that adjustment to living versus working in Japan is different, it may be that some pre-departure knowledge facilitates adjustment to living

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 291

in Japan while other pre-departure knowledge facilitates adjustment to working in Japan.

In terms of adjustment facilitating outside factors, general support was found for the hypotheses that family adjustment and association with host nationals were correlated with EXM general adjustment. A non-significant relationship was found between these two variables and EXM work adjustment. As argued earlier, it is unlikely that association with host nationals would provide cues to facilitate work adjustment or that work adjustment would lead one to associate more with host nationals. However, an explanation for the lack of correlation between family adjustment and EXM work adjustment is not as obvious. It may be that the significant correlation between family adjustment and EXM general adjustment conditions and interacting with host nationals is due to their similar nature-adjustment to weather, shopping, interpersonal interactions, etc. The lack of a significant correlation between family adjustment and EXM work adjustment may be because the nature of work adjustment is different from general adjustment. Therefore, even if the family does not adjust to living in the foreign country, the EXM can still adjust to the job because it to some degree is similar to responsibilities held prior to the transfer and therefore independent of its current foreign context.

However, an important limitation should be noted. The EXMs gave the scores for the adjustment of their families. Even though items referring to family adjustment were placed in a separate portion of the questionnaire to avoid response set bias, the association between EXM general adjustment conditions and to interacting with host nationals may be a function of response set bias. Future research might test this by obtaining family self-reports and comparing them with EXM reports to see if EXM reports of the family's adjustment are similar to those provided by the family members (primarily the spouse).

In considering the findings and suggestions of this study, two sampling problems should be considered. Because of the logistics of updating the large registry from which this sample was drawn, only those individuals that had been in the country more than six to eight months were listed in the directory. Thus, the first six months of adjustment were not directly measured. This is important

because respondents indicated that the low point for their adjustment occurred approximately six months after their arrival. If those who had trouble making the transitions returned to the U.S. at this average low point, then the sample would be overly represented by relatively adjusted managers. Second, even though the directory had multiple names for a given company, the names listed for a given company were not exhaustive, and the names listed tended to be the higher level executives. Thus, the sample, while representing a large number of companies and industries, was overly represented by higher level executives and may not be generalizable to lower level expatriates such as technical specialists.

In conclusion, it seems firms might facilitate international transfers and adjustments by providing overlap time between the returning manager and the new replacement as well as providing clear job descriptions in order to reduce

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

292 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

role ambiguity. Also, firms not providing training to facilitate pre-departure

knowledge of the country and culture might facilitate adjustment by changing this practice, especially if the manager is expected to regularly interact with host nationals. Several research issues have been mentioned, but it seems particularly important to investigate the causal relationship between family adjustment and EXM adjustment and between association with host nationals and EXM adjustment. The increased internationalization of business makes the more precise understanding of all the dimensions of adjustment to international transfers increasingly important.

APPENDIX

Respondents were asked to indicate on a scale from 1 to 7 (1= Not Adjusted At All; 7 = Very Well Adjusted) the degree to which they are adjusted or not on the following eleven items.

1. How adjusted are you to your job and responsibilities?

2. How adjusted are you to working with Japanese co-workers?

3. How adjusted are you to the transportation system in Japan?

4. How adjusted are you to working with Japanese outside your company?

5. How adjusted are you to the food in Japan?

6. How adjusted are you to the weather in Japan?

7. How adjusted are you to interacting with Japanese in general?

8. How adjusted are you to shopping in Japan?

9. How adjusted are you to supervising Japanese subordinates?

10. How adjusted are you to generally living in Japan?

11. How adjusted are you to the entertainment available in Japan?

RREFERENCES

Abe, H. & R.L. Wiseman. 1983. A cross-cultural confirmation of intercultural effectiveness. International

Journal of Intercultural Relations, 7:53-68.

Adams, T.E. & N. Kobayashi. 1969. The world of Japanese business. Tokyo: Kodansha International.

Baker, J.C. & J.M. Ivancevich. 1971. The assignment of American executives abroad: Systematic, haphazard,

or chaotic? California Management Review, 13:39-44.

Bardo, J.W. & D.J. Bardo. 1980. Dimensions of adjustment of American settlers in Melbourne, Australia. Multivariate Experimental Clinical Research, 5:23-28.

Brein, M. & K.H. David. 1971. Intercultural communication and adjustment of the sojourner. Psychology

Bulletin, 76:215-30.

Brim, O.G. 1966. Socialization through the life cycle, In O.G. Brim & S. Wheeler, eds., Socialization after childhood: Two essays, 1-45. New York: Wiley.

Burr, W.R. 1972. Role transitions: A reformulation of theory. Journal of Mariage and the Family, 34:407- 16.

Churchill, G.A. & J.P. Peter. 1984. Research design effects on the reliability of rating scales: A meta analysis. Journal of Marketing Research, 21:360-75.

Cogswell, B.E. 1967. Rehabilitation of the paraplegic: Process of socialization. Social Inquiry, 13:417- 40.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

WORK ROLE TRANSITIONS 293

David, K.H. 1976. The use of social learning theory in preventing intercultural adjustment problems. In

P. Pederson, W.J. Lonner & J. Draguns, eds., Counselling across cultures, 123-37. Honolulu, Hi.:

University of Hawaii Press.

Dawis, R.V. & L.H. Lofquist. 1984. A psychological theory of work adjustment. Minneapolis, Minn.: University of Minnesota Press.

Detweiler, R. 1975. On inferring the intentions of a person from another country. Journal of Personality, 43:591-611.

Dillman, D.A. 1978. Mail and telephone surveys. New York: Wiley.

Dunnette, M.D., R.D. Arvey & P.A. Banas. 1973. Why do they leave? Personnel, 50:25-39.

Feldman, D. 1976. A contingency theory of socialization. Administrative Science Quarterly, 21:433-452.

& J.M. Brett. 1983. Coping with new jobs: A comparative study of new hires and job changers. Academy of Management Journal, 26:258-72.

Fisher, C.D. 1982. Identifying the outcomes of socialization: Two studies. Technical Report No. TRONR- 8. College Station, Tex.: Texas A & M University.

George, L. 1980. Role transition in later life. Belmont, Calif.: Wadsworth.

Gudykunst, W.B., M.R. Hammer & R.L. Wiseman. 1977. An analysis of an integrated approach to cross- cultural training. International Journal of Intercultural Relations, 1:99-110.

Gullahorn, J.R. & J.E. Gullahorn. 1962. An extension of the u-curve hypothesis. Journal of Social Issues, 3:33-47.

Hammer, M.R., W.B. Gudykunst & R.L. Wiseman. 1978. Dimensions of intercultural effectiveness: An

exploratory study. International Journal of Intercultural Relations, 2:382-93.

Harris, J.G. 1973. A science of the south pacific: Analysis of the character structure of the Peace Corps

volunteer. Annual Psychologist, 28:232-47.

Harris, P. 1979. The unhappy world of the expatriate. International Management, July: 49-50.

Harvey, M.C. 1982. The other side of the foreign assignment; dealing with the repatriation problem. Columbia Journal of World Business, 17:53-59.

Hawes, F. & D.J. Kealey. 1981. An empirical study of Canadian technical assistance. International Journal of Intercultural Relations, 5:239-58.

Hays, R.D. 1971. Ascribed behavioral determinates of success-failure among U.S. expatriate managers. Journal of International Business Studies, 2:40-46.

Howard, C.G. 1974. A model for the design of a selection program for multinational executives. Public Personnel Management, March: 138-45.

Ilgen, D.R. 1975. The psychological impact of realistic job previews. Technical Report No. 2, Department

of Psychological Science, Purdue University.

Jones, G.R. 1986. Socialization tactics, self-efficacy, and newcomers' adjustment to the organization. Academy of Management Journal, 2:262-79.

Kahn, R.L., D.M. Wolfe, R.P. Quinn & J.D. Snoek. 1964. Organizational stress. New York: Wiley.

Karasek, R. 1979. Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly, 2:215-308.

Lanier, A.R. 1979. Selecting and preparing personnel for overseas transfers. Personnel Journal, 58:160- 63.

Louis, M.R. 1980. Surprise and sense making: What newcomers experience in entering unfamiliar organizational settings. Administrative Science Quarterly, 25:226-51.

, B. Posner & G. Powell. 1983. The availability and helpfulness of socialization practices. Personal Psychology, 36:857-66.

Lysgaard, S. 1955. Adjustment in a foreign society: Norwegian Fullbright grantees visiting the United States. International Social Science Bulletin, 7:45-105.

Mendenhall, M. & G. Oddou 1985. The dimensions of expatriate acculturation. Academy of Management Review, 10:39-47.

Minkler, M. & R.P. Biller. 1979. Role shock. Human Relations, 32:125-40.

Misa, K.F. & J.M. Fabricatore. 1979. Return on investment of overseas personnel. Financial Executive, April: 42X-6.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms

294 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SUMMER 1988

Mortimer, J.T. & R.C. Simmons. 1978. Adult socialization. Annual Review of Sociology, 4:421-54.

Mumford, S.J. 1975. Overseas adjustment as measured by mixed standard scales. Paper presented at Western Psychology Association, Sacramento.

Nicholson, N. 1984. A theory of work role transitions. Administrative Science Quarterly, 29:172-91.

Oberg, K. 1960. Culture shock: Adjustment to new cultural environment. Practical Anthropologist, 7:177- 82.

Pinder, C.C. & H. Das. 1979. Personnel transfers and employee development. In K.M. Roland & G. R. Ferris, eds, Research in personnel and human resource management, 2:187-218. Greenwich, Conn.: JAI Press.

& K.G. Schroeder. 1987. Time to proficiency following transfers. Academy of Management Journal, 30:336-53.

Ratiu, I. 1983. Thinking internationally: A comparison of how international executives learn. International

Study of Management and Organizations, 13:139-50.

Rizzo, J.R., R.J. House & S.I. Lirtzman. 1970. Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 2:150-63.

Ross, C.W. 1985. Americans abroad. Unpublished dissertation, Fielding Institute.

Ruben, I. 1967. The reduction of prejudice through lab training. Journal of Applied Behavioral Science, 3:29-50.

& D.J. Kealey. 1979. Behavioral assessment of communication competency and the prediction of cross-cultural adaptation. International Journal of Intercultural Relations, 3:15-47.

Sarbin, T.R. & V.L. Allen. 1968. Role theory. In G. Lindzey and A. Aronson, eds., Handbook of social psychology, 488-567. Reading, Mass.: Addison-Wesley.

Seward, J. 1975. Speaking the Japanese business language. European Business, Winter, 40-47.

Stewart, R. 1982. Choices for the manager. Englewood Cliffs, N.J.: Prentice-Hall.

Toffler, B.L. 1981. Occupational role development: The changing determinants of outcomes for the individual.

Administrative Science Quarterly, 26:396-418.

Torbion,, I. 1982. Living abroad. New York: Wiley.

Tung, R. 1981. Selecting and training of personnel for overseas assignments. Columbia Journal of World Business, 16:68-78.

. 1982. Selecting and training procedures of U.S., European, and Japanese multinational corporations. California Management Review, 25:57-71.

1984. Key to Japan's economic strength: Human power. Lexington, Mass.: Lexington Books.

Van Maanen, J. & E. Schein. 1979. Toward a theory of organizational socialization. In B. Staw, ed., Research in organizational behavior, 1:209-64. Greenwich, Conn.: JAI Press.

This content downloaded from 54.84.104.155 on Thu, 21 Sep 2017 23:52:37 UTC All use subject to http://about.jstor.org/terms