International Business Review Volume 24 Issue 3 2015 (Doi 10.1016/j.ibusrev.2014.08.009) Katsikea, Evangelia Theodosiou, Marios Morgan, Robert E. - Why People Quit - Exp

G Model
IBR-1140; No. of Pages 13

International Business Review xxx (2014) xxxxxx
Contents lists available at ScienceDirect
International Business Review

journal homepage: www.elsevier.com/locate/ibusrev
Why people quit: Explaining employee turnover intentions among

export sales managers
Evangelia Katsikea a,1, Marios Theodosiou b,*, Robert E. Morgan c,2
a
Assistant Professor of Marketing, Department of Marketing and Communication, Athens University of Economics and Business, 76 Patission Street,
10434 Athens, Greece
b
Associate Professor of Marketing, Department of Business and Public Administration, School of Economics and Management, University of Cyprus,
Kallipoleos 75, PO Box 20537, CY-1678 Nicosia, Cyprus
c
Sir Julian Hodge Chair of Marketing and Strategy, Cardiff Business School, Cardiff University, Colum Drive, Cardiff CF10 3EU, UK
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 26 November 2012
Received in revised form 11 July 2014
Accepted 13 August 2014
Available online xxx
Predictably, sales organizations invest substantial nancial and human resources in developing effective
salesforces and yet salespeople are among the highest risk group in terms of staff turnover. For export
sales settings, the organizational consequences of this form of turnover are even more severe. This study
develops a comprehensive conceptual model of seventeen hypothesized relationships among key
structural, supervisory-related, and psychological factors, and examines this nomological network that
leads to explaining export sales managers intentions to quit. The ndings reveal the favorable impact of
formalization and the unfavorable impact of centralization upon both role ambiguity and role conict.
The study nds that both formalization and centralization relate positively to the export sales
management behavior control system. Role stressors deleteriously affect export sales managers job
satisfaction, which in turn affects negatively intentions to quit. The study also discovers moderation
effects of psychic distance and export sales managers experience.
2014 Elsevier Ltd. All rights reserved.
Keywords:
Employee turnover
Export sales
Job satisfaction
Management control
Organizational structure
From afar it can appear to be glamorous, jetting off a couple of

times per month to faraway, exotic destinations to close major
deals. In truth it can be extremely hard work and is not a role for
the faint hearted. Many export sales professionals that weve
worked with are renowned experts in airport departure lounges
and unfortunately see little of the countries to which they travel
(an International Recruitment Consultancy, 2011).
1. Introduction
Sales organizations invest substantial nancial and human
resources in developing effective salesforces (Ahearne, Rapp,
Hughes, & Jindal, 2010). To gain maximum return from this
investment, organizations must be able to retain competent and
* Corresponding author. Tel.: +357 22893606; fax: +357 22895030.

E-mail addresses: ekt@aueb.gr (E. Katsikea), mariosth@ucy.ac.cy
(M. Theodosiou), morganre@cardiff.ac.uk (R.E. Morgan).
1
Tel.: +30 2108203417; fax: +30 2108203417.
2
Tel.: +44 2920870001.
experienced salespeople on a long-term basis (Homburg & Stock,

2004). In practice, however, all sales organizations face a certain
degree of salesperson turnover. Importantly though, evidence
suggests that salespeople are the highest risk group in terms of
staff turnover with some estimates between 30% to 70% per annum
(JDH Group, 2011). This undesirable phenomenon results in the
loss of knowledgeable and experienced salespeople, causes
signicant replacement costs, and has many other detrimental
effects for the organizations (Noble, 2008).
The deleterious effects of salespeoples turnover are even more
severe in the case of export sales organizations. In todays highly
globalised and competitive international business environment,
export sales managers play a critically important role in the
development and implementation of effective export sales
strategies and tactics. In small and medium sized rms in
particular, export sales managers are commonly the only company
personnel that travel overseas and make direct contact with
foreign customers. Personal visits to foreign markets enable export
sales managers to acquire valuable knowledge and rst-hand
experience regarding local market conditions and export customer
requirements and preferences. The literature that informs existing
knowledge regarding international sales personnel turnover is
limited. Insights exist from expatriate adjustment theories,
http://dx.doi.org/10.1016/j.ibusrev.2014.08.009
0969-5931/ 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Katsikea, E., et al. Why people quit: Explaining employee turnover intentions among export sales
managers. International Business Review (2014), http://dx.doi.org/10.1016/j.ibusrev.2014.08.009
G Model

2
E. Katsikea et al. / International Business Review xxx (2014) xxxxxx
voluntary employee turnover theories, job embeddedness theories

as well as domestic sales theories. However, the particular role,
intrinsic and extrinsic demands, and contextual determinants that
confront the export sales manager mean that these theories, at
best, only partially explain the high turnover rates that these
personnel are victim of.
Employee turnover presents many challenges for rms from the
high replacement costs, demanding training requirements, loss of
learning and experience effects, reduced morale among the
remaining organizational members and the critical nancial
performance consequences (Reiche, 2008). Established business
relationships may be undermined, and valuable knowledge
regarding certain export markets, as well as personal networks,
may be lost. In addition, the export sales organizations will incur a
signicant replacement cost, whereas it may take many years
before sales performance reaches previous levels in the affected
markets. Thus, export sales organizations can benet greatly by
minimizing turnover rates among these employees. Surprisingly,
the relevant literature devotes limited research on the factors the
inuence the role perceptions, job-related attitudes and behavior,
work outcomes, and turnover rates of export sales personnel.
The present study aims to enhance the existing body of
knowledge regarding the determinants of export success by
focusing on the role of the export sales manager. A review of
the pertinent export marketing literature indicates that the vast
majority of previous studies on export performance have examined the impact of export marketing strategy elements, rmspecic factors and external/environmental inuences (e.g.,
Cavusgil & Zou, 1994; Leonidou, Katsikeas, & Samiee, 2002;
Morgan, Kaleka, & Katsikeas, 2004; Murray, Gao, & Kotabe, 2011).
Furthermore, within the rich research stream that examines the
performance outcomes of export marketing strategy, attention has
focused on the export strategy planning process and the planned
export marketing strategy, whereas the implementation of
planned export marketing strategy has been largely ignored
(Morgan, Katsikeas, & Vorhies, 2012). We propose that export sales
managers play a crucially important role in the implementation of
export marketing strategies in specic foreign markets, as they are
commonly the only company personnel that have direct personal
contacts with current and prospective foreign customers/distributors to promote exports and negotiate sales deals. As a result of
their personal visits to foreign markets, export sales managers
acquire valuable knowledge and rst-hand experience regarding
foreign customers needs and preferences and local market
conditions, while at the same time have the opportunity to
develop successful business relationships. In addition, export sales
managers can directly assess export customers response to the
strategic actions enacted by a rms export marketing plan.
Therefore, export sales managers can signicantly enhance an
exporting rms external implementation effectiveness, as dened by
Morgan, Katsikeas, and Vorhies (2012, p. 273).
When an experienced export sales manager quits his/her job,
the export organization loses a valuable resource that is extremely
difcult, costly, and time consuming to replace, whereas in the
meantime established protable relationships with specic export
customers may deteriorate. To safeguard the long-term performance of exporting operations exporting organizations must adopt
appropriate management practices that will enable them to retain
their most competent export sales managers on a long-term basis.
This study aims to ll a notable gap in the extant export marketing
literature by offering a comprehensive conceptual framework that
explains how export sales managers form intentions to quit the
export sales organization. The ndings of this study provide useful
insights to export executives in terms of the action they can take in
order to predict quitting intentions and take appropriate measures
in order to reduce this undesirable outcome.
Our study makes also an important contribution to sales

management theory by demonstrating that organizational structure and management control systems are distinct but complementary mechanisms that enable the export sales organization to
direct the attitudes and behavior of export sales personnel toward
desirable outcomes. Previous research, conducted predominantly
in a domestic sales context, has focused on sales management
control systems which are commonly implemented at the level of
eld sales managers who are responsible for monitoring, directing,
evaluating, and rewarding eld salespeople (Anderson & Oliver,
1987; Challagalla & Shervani, 1996). In contrast, less research
attention has been devoted on the role of organizational structure,
which is an alternative form of control, usually determined at
higher levels in the organizational hierarchy, but which can still be
very effective in directing, inuencing, and shaping the attitudes
and behavior of salespeople (Jaworski, 1988). We posit that the
role of organizational structure is increasingly important in the
case of export sales organizations. The personal interaction
between export sales managers (travelling abroad) and their
immediate supervisors (based in the home-market) is seriously
constrained as a result of the large geographical distances involved,
and therefore traditional management control approaches become
less effective. Organizational structure can eliminate this aw by
setting the decision-making boundaries for export sales managers,
clarifying their lines of responsibility and authority, and determining the nature of their relationship with their immediate
supervisors.
Finally, we expand the existing sales management literature by
providing empirical evidence regarding the chain-of-effects among
key organizational, supervisory-related, and psychological factors
that ultimately lead to the reduction of export sales managers
intentions to leave the organization. Our study also theorizes, and
empirically conrms, that within the export sales context, it is
imperative to consider factors which reect the inherent
complexities in international business activities, or enable
managers to deal with this complexity. In this respect, we examine
the moderating effects of psychic distance and export sales
managers experience.
The rest of the article is organized as follows. The next section
presents the conceptual framework of the study and develops
research hypotheses. Section three presents the research methodology, and section ve reports the results of statistical analysis. The
article concludes with a discussion of the key ndings, managerial
implications, limitations and directions for future research.
2. Conceptual framework and research hypotheses
Given the ubiquity of empirical studies devoted to employee
turnover in the literature, it is incongruous to suggest that still
little insight is available into why people quit their jobs (Lee,
Gerhart, Weller, & Trevor, 2008). Turnover refers to the actual
number of employees that leave an organization within a specied
time period (Jaramillo, Mulki, & Boles, 2013). A high turnover rate
has signicant negative consequences for the rm that include
increased costs and reduced effectiveness and efciency in serving
existing customers and prospects (Sager, Varadarajan, & Futrell,
1988). Although salespeople may leave the organization for
various reasons, most research attention has been devoted on
identifying the factors that drive them to voluntarily resign from
their job. Mobley (1977) proposes a model of the turnover process
that includes various psychological and behavioral stages that
precede an employees nal decision to quit. A key proposition of
this model is that a salesperson withdraws from the job and starts
exhibiting dysfunctional behavior (e.g., absenteeism, passive job
behavior) well before actually quitting the job. Moreover, as
Mobleys model suggests, job attitudes are more directly related
G Model

to withdrawal cognitions associated with the decisions to leave

and only indirectly related to the actual turnover behavior
(Mowday, Koberg, & McArthur, 1984, p.80). A number of empirical
studies nd that propensity to leave/intention to quit is the
stronger predictor of actual turnover behavior (Johnston, Varadarajan, Futrell, & Sager, 1987; Sager et al., 1988).
The present study aims to expand the pertinent literature on
salespeoples turnover to the context of export sales managers. As
previously discussed, the negative consequences associated with
export sales managers turnover are even more severe. Domestic
sales organizations are often able to ll vacant sales positions in
particular sales territories through transfers from nearby territories. Although redesigning sales territories can still have devastating effects on existing customersalesperson relationships, at least
replacements will still operate within a familiar business
environment. However, export sales organizations face a
completely different situation. Finding a competent person to ll
a vacant sales position in a specic export market may be
extremely difcult, as valuable knowledge regarding environmental and market conditions, business practices, and relationship
norms that prevail in foreign markets accumulates after many
years of personal experience and repeat visits to each market.
Therefore, retaining competent and experienced export sales
managers should be an important priority for export sales
organizations. Academic research can assist export sales organizations to reduce turnover rates by investigating the factors that
drive export sales managers to form turnover intentions, and this
was the primary motive for the execution of this study.
Given the lack of relevant studies in an export marketing
context, we reviewed the mainstream sales management literature
for insights regarding the key antecedents of salespeoples
propensity to leave a sales organization. Our objective was not
only to identify the immediate precursors of intentions to leave,
but also to adopt a backward induction process in order to
establish a sequence of effects that ultimately leads to export sales
managers propensity to leave. We also tried to develop a concise
conceptual model, capable of explaining a signicant percentage of
the variance of propensity to leave, but without adding unnecessary complexity by including variables whose contribution in
explaining additional variance is negligible.
Drawing on the extant sales management literature, and
integrating insights from our pre-study interviews, we developed
the conceptual model outlined in Fig. 1. In particular, we propose

that export sales managers propensity to leave is directly
inuenced by their level of job satisfaction. In turn, job stress, as
determined by role ambiguity and role conict, is the main
antecedent of export sales managers job satisfaction. Based on our
conceptualization, export sales organizations can reduce export
sales managers role stress by employing appropriate control
mechanisms, at two different levels. At the lower level, the export
sales management control system focuses on the working
relationship between an export sales manager and his/her
immediate supervisor. In this study, we focus on sales management behavior control, which involves the activities which
supervisors undertake in order to inuence the behavior of export
sales managers and provide appropriate encouragement, guidance
and support (Anderson & Oliver, 1987; Challagalla & Shervani,
1996). At a higher level, the top management of an export sales
organization designs an appropriate organizational structure that
determines the job duties and responsibilities of export sales
managers and their immediate supervisors, the delegation of
decision making authority at different hierarchical levels, and the
rules, policies and procedures that organizational members should
follow in performing their job. According to Jaworski (1988, p. 27)
although structure is not part of the traditional management
control system, it should be considered as an additional form of
control, insofar as it directs, inuences, and shapes individual and
group behavior. Furthermore, as a higher-level control mechanism,
organizational structure inuences the employment of an effective
sales management control system.
Our conceptual model includes also two moderating variables,
psychic distance and export sales managers experience, which
capture to some extent the unique circumstances faced by export
sales organizations. Psychic distance is dened as the perceived
differences between the home country and the foreign country in
terms of economic development, climatic conditions, lifestyles,
consumer preferences, language, education, and cultural values
(Sousa & Bradley, 2006). The internationalization process literature
suggests that rms are more familiar with psychologically close
countries, and therefore can manage better their operation in these
countries than in more distant ones (Johanson & Vahlne, 1977).
Therefore, export sales managers should be more familiar with
business practices and market conditions in psychologically close
countries, and have a thorough knowledge and understanding of
Psychic Distance
Formalization
Export sales
manager
experience
Role
Ambiguity
Export Sales
Management
Control
Job
Satisfaction
Propensity to
Leave
Role
Conflict
Centralization
Psychic Distance
Fig. 1. A conceptual model of organizational structure, export sales management control, job satisfaction, and propensity to leave.
G Model

4
export customers characteristics, preferences, requirements, and

purchasing habits. Therefore, we expect that control mechanisms
(i.e., organizational structure and sales management control) will
have a stronger impact on export sales managers role stress in
more distant export markets. Similarly, experienced export sales
managers are more capable in coping with the uncertainty and
complexity inherent in international business activities. Consequently, an export sales management control system is likely to be
more effective in reducing the role stress of less experienced export
sales managers. We develop the rational for each of the
hypothesized relationships in the following paragraphs.
2.1. Organizational structure and role stress
During the past four decades, empirical research examined the
negative inuence of role stress on important sales job outcomes
like job satisfaction, organizational commitment and job performance (Netemeyer, Brashear-Alejandro, & Boles, 2004). By
contrast, the organizational factors that can potentially contribute
towards reducing role stress received much less research attention.
Role stress consists of role ambiguity and role conict (Johnson &
Sohi, 2014). Role ambiguity is the extent to which an individual is
uncertain about the expectations of role partners, whereas role
conict refers to the perceived incompatibility among role
expectations and demands (e.g., from supervisors, customers,
policies) (Rizzo, House, & Lirtzman, 1970). Thus, in order to
effectively handle feelings of role stress among their export sales
managers, export sales organizations must nd ways to reduce the
uncertainty surrounding their job, and assist them in handling
conicting requests and demands. Organizational structure can
serve as an effective control mechanism and contribute signicantly towards achieving these objectives.
The structural constructs that have attracted most research
attention in the relevant literature include formalization and
centralization (Chi & Sun, 2013). Formalization is the degree to
which rules dene roles, authority relations, communications,
norms and sanctions and procedures (Hall, Johnson, & Haas, 1967).
As implied in this denition, formalization provides guidance to
problem solving, creates less confusion, lowers divergent interpretations to similar activities, reduces double standards, claries
role expectations, and shifts the risk and attribution of failure to
management (Auh & Menguc, 2007; Michaels, Cron, Dubinsky, &
Joachimsthaler, 1988). Therefore, provided that the content of
formal rules and procedures is appropriately determined (Jaworski
& Kohli, 1993), formalization can potentially reduce perceptions of
role ambiguity.
The relationship between formalization and role conict is
more controversial. In particular, some authors suggest that
formalization increases role conict by reducing the exibility
available to salespeople to deal with conicting demands and
preventing them from adapting to idiosyncratic circumstances
(Agarwal, 1993). Although these claims have some merit, they are
based on the assumption that formal standards and policies are
very strict and prevent salespeople from exercising any discretion
when negotiating with customers. However, our pre-study interviews revealed that the main objective of exporting rms when
developing formal standards and policies is to determine the
boundaries within which export sales managers can operate.
Formalization makes export sales managers less vulnerable to the
pressures and unreasonable demands of foreign customers, and
enables them to devote their time and selling efforts on the most
promising customers and prospects. Formal rules and procedures
can be viewed as a guiding framework that serves as a point of
reference for export sales managers when dealing with foreign
customers. Therefore, formalization is expected to reduce perceptions of role conict. Limited empirical evidence also supports a
negative relationship between formalization and role stressors

(Michaels et al., 1988).
H1. Formalization relates negatively to role ambiguity
H2. Formalization relates negatively to role conict
In a highly centralized export sales organization, export sales
managers have restricted decision-making authority. However, the
boundary-spanning nature of their job often forces them to make
decisions and resolve conicts between foreign customer expectations and demands and company offerings (Sohi, 1996).
Furthermore, greater centralization prevents export sales managers from participating in determining company policies and
standards whereas their input, thoughts and ideas are excluded
from the decision making process (Auh & Menguc, 2007). In such a
working environment, they are less likely to understand the
expectations of their supervisors or devote effort on dening their
role requirements (Sohi, 1996). Moreover, they may feel that
company policies and standards do not reect market realities
overseas and impose unreasonable constraints on their behavior
(Walker, Churchill, & Ford, 1975). Similarly, centralization constrains export sales managers exibility and reduces their ability to
resolve conicting demands between customers expectations and
company standards, increasing perceptions of role conict. Based
on the above discussion, we posit the following:
H3. Centralization relates positively to role ambiguity
H4. Centralization relates positively to role conict
2.2. Organizational structure and export sales management behavior
control
Previous research has devoted limited research attention on the
factors that either guide the choice of a particular control system or
determine its effective implementation (Baldauf, Cravens, & Piercy,
2005). The existing typologies of sales management control
(Anderson & Oliver, 1987; Challagalla & Shervani, 1996) emphasize
the working relationship between a supervisor and his/her
subordinates. However, they do not explicitly examine the factors
that determine the particular type of control system that an
organization adopts or the hierarchical level at which this decision
is made.
To be effective, a particular sales management control system
must be centrally developed and implemented consistently
throughout an export sales organization. In particular, organization-wide policies, rules, and directives must dene the nature of
the working relationship between export sales managers and their
immediate supervisors. Upper management should specify the
preferred form, frequency and content of information exchange
between the two parties (e.g., informal meetings versus formal
reports). Moreover, the responsibilities of supervisors in terms of
providing constructive feedback, guidance and support to export
sales managers must be formally specied. Otherwise, the
effectiveness of the control system will depend solely on the
discretion and eagerness of individual supervisors. Based on this
discussion, this study proposes that an export sales organization is
more likely to develop and implement an effective sales
management control system under a formalized and centralized
organizational structure.
H5. Formalization relates positively to sales management behavior control
H6. Centralization relates positively to sales management behavior control
G Model

2.3. Export sales management behavior control and role stressors

Behavior control systems concentrate on developing and
maintaining relationships and encourage long-term results rather
than short-term outcomes (Baldauf, Cravens, & Grant, 2002).
Behavior control systems emphasize managerial coaching and
mentoring designed and implemented to develop subordinate
skills and capabilities (Mallin, Asree, Koh, & Hu, 2010). Challagalla
and Shervani (1996) make a distinction between two types of
behavioral control, namely activity control and capability control.
Activity control concentrates on the routine activities that
employees are expected to perform on a regular basis, the
monitoring of actual behavior and the administering of rewards
and punishments on the basis of the performance of specied
activities. Capability control emphasizes the development of
individual skills and abilities, and represents an attempt to
inuence performance by ensuring that employees possess the
set of skills and abilities that enable good performance (Challagalla
& Shervani, 1996, p. 90).
The effective implementation of a behavior-based control
system requires the close cooperation and richer communication
between export sales managers and their immediate supervisors.
Supervisors are induced to commit time and effort in monitoring
and providing feedback to export sales managers, as well as in
assessing their skills and capabilities and providing guidance and
support for improvement. Through this interaction, export sales
managers acquire a greater knowledge and understanding of
their supervisors expectations, and recognize what activities and
behaviors are consistent with the export sales strategy and are
likely to lead to desired end-results (Challagalla & Shervani,
1996). Moreover, under a behavior-based control system, export
sales managers can work closely with their supervisors and rely
on their guidance and support in order to handle effectively
customers demands and requests that may be incompatible with
company policies and standards. Therefore, behavior control is
expected to lower export sales managers role ambiguity and
role conict. Available empirical evidence also supports a
negative relationship between behavior control and role stress
(Challagalla & Shervani, 1996; Jaworski, Stathakopoulos, &
Krishan, 1993).
H7. Sales management behavior control relates negatively to role
ambiguity
H8. Sales management behavior control relates negatively to role
conict
2.4. Role stressors and job satisfaction
During the past four decades numerous studies have examined
the effects of job satisfaction on important work outcomes like
organizational commitment, job performance, and turnover
intentions (Netemeyer, Johnston, & Burton, 1990; Netemeyer,
Brashear-Alejandro, & Boles, 2004; Singh, Verbeke, & Rhoads,
1996). Although the ndings of these studies have not always been
consistent (particularly regarding the causal direction between job
satisfaction and other work outcomes) conventional wisdom
suggests that high levels of job satisfaction are desirable. The
extant literature has also devoted signicant research attention on
the antecedents of job satisfaction. In particular, previous research
has studied extensively the effects of role ambiguity and role
conict (Netemeyer et al., 1990). Theoretically, role stress leads to
tensions and pressures due to job requirements, including the
psychological and physical consequences of such pressures, and
therefore reduces job satisfaction (Fry, Futrell, Parasuraman, &
Chmielewski, 1986). Furthermore, empirical research indicates
that both role ambiguity and role conict have a signicant

negative effect on job satisfaction (Fry et al., 1986).
H9. Role ambiguity relates negatively to job satisfaction
H10. Role conict relates negatively to job satisfaction
2.5. Job satisfaction and propensity to leave
Propensity to leave is dened as the subjective estimation of an
individual regarding the possibility to quit the organization in the
near future (Mowday et al., 1984). This construct represents a
behavioral intention that precedes actual salesperson turnover
(Brown & Peterson, 1993). Conceptually, the contention that
intention to leave predicts actual turnover is consistent with the
Fishbeins (1967) model of attitudes, intentions, and behavior.
Previous research provides consistent evidence that propensity to
leave is the most effective predictor of salespeople turnover
(Brown & Peterson, 1993).
A low turnover rate among salespeople is desirable because it
reduces signicantly replacement costs and keeps knowledgeable
and experienced personnel in the sales organization (Good, Page, &
Young, 1996). Understanding the underlying causes of turnover
can provide insights into how to control the phenomenon (Good
et al., 1996). Therefore, intention to leave is a key construct
examined in models of salespeoples role perceptions, job
attitudes, behavior, and performance (Johnston, Parasuraman,
Futrell, & Black, 1990; Netemeyer et al., 1990, 2004; Sager, 1994;
Singh et al., 1996). The results of these studies indicate that job
satisfaction is a primary determinant of turnover intentions, and a
direct negative relationship exists between the two constructs.
Thus, satised salespeople are less likely to leave the organization.
H11. Job satisfaction relates negatively to propensity to leave
2.6. Moderating effects
2.6.1. Psychic distance and formalization
The concept of distance has attracted interest to international
business and marketing scholars who aim to explain variations in
international business and marketing strategies and operations
across foreign markets (Prime, Obadia, & Vida, 2009, p. 184). Psychic
distance includes factors that can prevent the ow of information
from and to a foreign market (Johanson & Vahlne, 1977). Examples
include differences in language, education, business practices,
culture and industrial development. Due to the geographical
distances involved and the inherent complexities associated with
international operations (Piercy, Low, & Cravens, 2011), which are
more evident in psychologically distant markets, the prescription of
clear rules, policies and guidelines through a highly formalized
organizational structure is essential in order to reduce export sales
managers perceptions of role conict and ambiguity. On the other
hand, export sales managers are more capable and knowledgeable
to deal with job requirements and customer demands in
psychologically close export markets. Therefore:
H12. The inuence of formalization on export sales managers role
ambiguity is stronger in psychologically distant markets
H13. The inuence of formalization on export sales managers role
conict is stronger in psychologically distant markets
2.6.2. Psychic distance and centralization
In a centralized organization, export sales managers have
limited decision making authority. By contrast, export executives
must initiate or approve all major decisions. Still, export executives
G Model

6
are based in the companys headquarters and travel much less

frequently than export sales managers. Therefore, their exposure
to the environmental and market conditions that prevail in export
markets is rather limited. Moreover, they may lack a thorough
understanding of export customers needs, preferences, and
purchasing habits. Under these circumstances, decisions made
by export executives are unlikely to reect market realities in
export countries. In addition, export sales managers will experience higher levels of role ambiguity and role conict since they
will have limited participation in decision making, constrained
exibility to respond effectively to the idiosyncrasies of export
markets, and incapability to resolve conicting demands between
customers expectations and company standards. However, both
export executives and export sales managers have more accurate
knowledge with respect to export markets that are perceived as
psychologically close. Consequently, export executives are better
equipped to make decisions regarding these markets, whereas
export sales managers are able to carry out more effectively their job
duties and responsibilities, despite the constrains imposed by
centralization. Therefore, this study expects that the negative
inuence of centralization on export sales managers role perceptions will be stronger in psychologically more distant markets.
H14. The inuence of centralization on export sales manager role
ambiguity is stronger in psychologically distant markets
H15. The inuence of centralization on export sales managers role
conict is stronger in psychologically distant markets
2.6.3. Export sales managers experience
Experience is an important variable that moderates relationships among salespeoples perceptions, attitudes and behavior
(Cron, Dubinsky, & Michaels, 1988). Compared to their experienced
counterparts, inexperienced salespeople are more sensitive and
react strongly to work related factors such as role stressors (Russ &
McNeilly, 1995). Therefore, sales managers must devote greater
attention and provide more extensive guidance and support to
inexperienced salespeople. Although the authors argue that
behavior control is an effective mechanism for reducing stress
among all export sales managers, they expect that an individuals
level of job experience moderates this inuence. In particular,
experienced export sales managers are in a better position to deal
with problems and difculties that may arise when performing
selling activities in export markets. Moreover, they are more
familiar with company policies and standards, and accustomed
with organizational culture and top managements leadership
style. Hence, although experienced export sales managers also
benet by the direction and support they receive from their
immediate supervisors, their less experienced colleagues are in a
much greater need for such assistance. Therefore, this study
expects that the contribution of behavior-based control towards
reducing role ambiguity and role conict will be greater among
less experienced export sales managers.
H16. The inuence of behavior control on role ambiguity is stronger for less experienced export sales managers
H17. The inuence of behavior control on role conict is stronger
for less experienced export sales managers
3. Research methodology
3.1. Sample and data collection procedures
This study tests the proposed conceptual model and hypotheses
using data collected from small- and medium-sized exporters in
the United Kingdom. The sampling frame for this study is drawn
from the Dun and Bradstreet commercial database of Key British
Enterprises. To be eligible for inclusion in the study sample, a rm
must meet the following thresholds: exporting at least for three
years; employ up to 500 full-time employees; and, operate in one
of the following industrial sectors: textile mill products; apparel
and other nished goods made from fabric and other similar
materials; chemical and allied products; rubber and miscellaneous
plastics products; machinery except electrical; and electrical and
electronic machinery, equipment and supplies. The selection of a
multi-industry sampling design is necessary in order to enhance
the generalizability of the research ndings. A random sample of
1000 eligible rms is selected for inclusion in this study.
This study employs a mail survey methodology for data
collection, following the guidelines provided by Dillman (2000).
A mail package is sent to the Head of Exporting Operations to
each of the 1000 sample rms, along with a cover letter that
explains the purpose and objectives of the study, and a postagepaid return envelope. Recipients are instructed to pass on the
survey packet to one individual that fullls the export sales
manager role in their rm. The initial mailing and two further
waves produce 160 usable responses, for a response rate of 16%.
However, 191 of the rms that were initially included in the study
sample are dropped, mainly because of incorrect conduct details or
termination of their exporting operations. Thus, the effective
response rate is 19.8% (160 of the eligible 809 rms). The
demographic characteristics of the sample are presented in
Appendix.
3.2. Response issues
Following Armstrong and Overton (1977) guidelines, the data are
checked for non-response bias by comparing early and late
respondents on all study constructs as well as on key company
characteristics including number of full-time employees, export
experience, export sales, and number of export markets. Nonresponse bias checks also involve comparing a random sample of
75 non-respondents with respondents in terms of the aforementioned company characteristics. The non-signicant t-test results
indicate that non-response bias is not a major problem in this study.
The key informant in this study is the export sales manager. This
study employs certain steps in order to ensure the quality of
responses. First, the cover letter explains clearly and in details the
job activities and responsibilities of an export sales manager and
instructs the recipients of the mailing package that the questionnaire must be lled by a person in the exporting rm that performs
this role. Second, it stresses that the study is conducted under the
guidance of an academic institution for purely academic research
purposes. Third, participants are promised anonymity and
condentiality of responses, and are offered a copy of the study
results. Fourth, this study follows the widely accepted practices for
questionnaire design and survey administration. Further, a post
hoc test for key informant quality is performed. In particular, the
last section of the questionnaire includes four statements that
assess respondents: (1) knowledge regarding the rms exporting
activities; (2) involvement with the rms exporting activities; (3)
responsibility for the rms exporting activities; and (4) condence
in answering the questions of the survey instruments. The mean
scores for these statements range between 6.01 and 6.40 (on a
seven-point scale anchored by Low and High) indicating that
potential bias attributable to the key-informant is negligible.
This study employs the Harmans one-factor test (Podsakoff &
Organ, 1986) to test for common method bias (CMB). First, this
analysis involves a principal components analysis of all constructs
examined in this study. The two second-order constructs, namely
export sales management control system and job satisfaction, are
G Model

represented in this analysis by the summated scales of their

underlying rst-order factors, whereas all other constructs by their
original measurement items. The unrotated solution results in
seven factors with eigenvalues greater then 1.0, accounting for 71%
of the variance. No general factor emerges, whereas the rst factor
explains only 25% of the variance. Second, a CFA approach to
Harmans one-factor test is used in which the summated scales of
the 15 rst-order factors included in the structural model are
restricted to load on a single factor. The results obtained from this
analysis indicate a poor t: x2(90) = 662.64, p < .001; CFI = .45,
NNFI = .36, and RMSEA = .20. Although the Harmans single-factor
test has several limitations, these results provide an indication that
common method bias does not pose a serious problem in our
investigation.
In order to provide a secondary analysis, the study tests for CMB
using the marker-variable approach proposed by Lindell and
Whitney (2001). Because the questionnaire is not designed a priori
to include a marker variable, this analysis uses the second smallest
correlation among the study constructs as an estimate of the
marker variable (Malhotra, Kim, & Patil, 2006). Based on this
estimate (rM = .004, correlation between activity punishment and
propensity to leave) the CMB-adjusted correlation between all
constructs and its t-statistic is computed using the following
formulas (Lindell & Whitney, 2001; Malhotra et al., 2006):
rA
r u r M
;
1 r M
r
A
t a=2;n3 p
1rA 2 =n3
where rA = CMB-adjusted correlation; ru = original correlation;

rM = marker variable. This procedure makes no difference to the
statistical signicance of any correlation. Thus, though CMB cannot
be completely ruled out, the results of this test indicate that such
bias does not explain the results.
3.3. Measurement
The authors employ multi-item reective scales to measure
study constructs. All measures are based on well-established scales
that are available in the empirical literature. The items for each of
the study measures and their sources are presented in Table 1. The
authors develop a preliminary version of the survey and then focus
their attention on establishing measures content validity for the
research study in hand. The rst step in this process involves
personal interviews with export executives and export sales
managers in 10 exporting rms. The average interview lasts for
approximately 2 h. During these interviews, the authors discuss
extensively the conceptual model and obtain managers comments
on the list of measurement items that they compose on the basis of
the literature review. Certain adaptations in the wording of certain
items are performed in order to emphasize the export sales
context. Second, the survey is administered to ve academic
researchers who are familiar with the areas of sales management
and export marketing and serve as expert judges to appraise its
face validity. The nal version of the questionnaire is extensively
pre-tested with export sales managers and no particular problems
are identied with respect to the measures, the clarity of questions,
response formats, or the length of the questionnaire.
4. Analysis and results
4.1. Measure validation
This study assesses the psychometric properties of the measures
using structural equation modeling procedures. Following Anderson
and Gerbings (1988) guidelines, this study estimates the measurement model separately and prior to the simultaneous estimation of
the measurement and structural model. The assessment of the
measurement model involves performing conrmatory factor
analysis (CFA) using the EQS statistical package (Bentler, 1995).
Because of the large number of constructs and measurement items
included in this study, this procedure estimates three CFA submodels. This approach enables the achievement of a higher ratio of
sample size to the number of free parameters to be estimated, which
should be at least 5:1 (Bentler & Chou, 1987). In each model the
analysis is performed using the elliptical reweighted least square
procedure, which provides unbiased parameter estimates for both
normal and nonnormal data. Moreover, each measurement item is
restricted to load only on its a priori specied factor.
The rst CFA model estimates formalization and centralization
as rst-order factors, and export sales management control as a
second-order factor comprising six dimensions: activity information, activity rewards, activity punishment, capability information,
capability rewards, and capability punishment. The chi-square
statistic of this model is signicant (x2(315) = 648.93, p < .05), as it
might be expected, given this statistics sensitivity to sample size
(Bagozzi & Yi, 1988). However, all other t indices suggest a good
t to the data (comparative t index [CFI] = .96, nonnormed t
index [NNFI] = .95, and root mean square error of approximation
[RMSEA] = .08). The second CFA model estimates job satisfaction as
a second-order construct comprising four dimensions: satisfaction
with overall job, satisfaction with fellow workers, satisfaction with
supervision, and satisfaction with customers. The goodness-of-t
indices for this model indicate a close t to the data
(x2(100) = 135.69, p < .05; comparative t index [CFI] = .98, nonnormed t index [NNFI] = .98, and root mean square error of
approximation [RMSEA] = .05). Finally, the third CFA model
contains ve rst-order constructs, namely role conict, role
ambiguity, propensity to leave, psychic distance and export sales
managers experience. The t indices obtained from this analysis
indicate a good t to the data (x2(179) = 232.27, p < .05; comparative t index [CFI] = .97, nonnormed t index [NNFI] = .97, and root
mean square error of approximation [RMSEA] = .04).
Table 1 presents the standardized loadings and t-values
obtained from the estimation of the three measurement submodels. All rst-order and second-order factor loadings are large
and signicant, indicating convergent validity. To test for
discriminant validity, this study follows the procedures recommended by Anderson and Gerbing (1988) and Fornell and Larcker
(1981). Discriminant validity is evident for all possible pairs of
constructs. To assess construct reliability this study estimates
Cronbachs alpha coefcient, composite reliability score, and
average variance extracted. As Table 1 shows, all constructs have
alpha values and composite reliabilities scores that exceed .7.
Moreover, the average variance extracted of all constructs is equal
to or greater than .5, satisfying the recommended thresholds
(Bagozzi & Yi, 1988; Fornell & Larcker, 1981). Table 2 presents the
correlation matrix and descriptive statistics for the study
constructs.
4.2. Structural model estimation
This study tests the research hypotheses by estimating the
structural model presented in Fig. 1. To attain an adequate ratio of
sample size to the number of free parameters, the two secondorder constructs, export sales management control and job
satisfaction, are aggregated to have six and four indicators
respectively, by averaging the items of their corresponding rstorder factors. All other constructs, are represented in the structural
model by their measurement items. Table 3 presents t statistics,
standardized coefcients and associated t-values for the structural
G Model

Table 1
Measurement scales, conrmatory factor analysis results, and reliabilities.
Constructs and measurement items
Standardised loadingsa
Formalization (a = .81; CR = .82; AVE = .61) (Aiken & Hage, 1966, 1968) (seven-point scale, anchored by Strongly Disagree
and Strongly Agree)
How things are done here is left up to the employee doing the work (R)
Employees here are left to do as they please (R)
Most people here make up their own rules in the job (R)
.63 (7.78)
.92 (11.81)
.77 (9.73)
Centralization (a = .88; CR = .89; AVE = .72) (Aiken & Hage, 1966, 1968) (seven-point scale, anchored by Strongly Disagree
and Strongly Agree)
Even small matters have to be referred to someone higher up for a nal answer
I have to ask my immediate manager before I do almost anything
Any decision I make has to have my immediate managers approval
.75 (10.16)
.98 (14.71)
.80 (11.01)
Export sales management control (Second-order factor) (Challagalla & Shervani, 1996) (Items measured on a seven-point scale,
anchored by Strongly Disagree and Strongly Agree)
A. Activity Information (a = .92; CR = .92; AVE = .70)
My immediate manager informs me about the export sales activities I am expected to perform
My immediate manager monitors my export sales activities
My immediate manager informs me on whether I meet his/her expectations in the venture market
My immediate manager tells me to adjust my export sales activities, if he/she feels that there is need to do so
My immediate manager evaluates my export sales activities
B. Activity rewards (a = .93; CR = .93; AVE = .87)
If I perform export sales activities well my immediate manager will commend me
I would be recognized by my immediate manager if he/she was pleased with how well I perform export sales activities
C. Activity punishment (a = .86; CR = .87; AVE = .68)
I would receive an informal warning if my immediate manager was not pleased with how I perform export sales activities
I would receive a formal reprimand if my immediate manager was unhappy with how I perform export sales activities
I would be put on probation if my immediate manager was unhappy with how I perform specied export sales activities
D. Capability information (a = .93; CR = .93; AVE = .74)
My immediate manager has standards by which my selling skills are evaluated
My immediate manager periodically evaluates the selling skills I use to accomplish a task
My immediate manager provides guidance on ways to improve selling skills and capabilities
My immediate manager evaluates how I make sales presentations and communicate with foreign customers
My immediate manager assists by suggesting why using a particular sales approach may be useful
E. Capability rewards (a = .81; CR = .80; AVE = .58)
Assignment of better territories or accounts depends on how good my selling skills are
I am commended if I improve my selling skills
Promotion opportunities depend on how good are my selling skills and capabilities
F. Capability punishment (a = .93; CR = .91; AVE = .76)
I receive an informal warning if my immediate manager is not pleased with my selling skills and capabilities
I receive a formal reprimand if my immediate manager is not pleased with my selling skills and capabilities
I am put on probation if my immediate manager is not happy with my selling skills and capabilities
.77 (7.67)
.73b
.85 (10.05)
.89 (10.62)
.85 (10.09)
.85 (10.12)
.68 (7.68)
.91b
.95 (13.53)
.63 (6.67)
.81b
.93 (11.70)
.72 (9.40)
.83 (8.89)
.79b
.87 (11.89)
.91 (12.79)
.85 (11.60)
.89 (12.28)
.83 (7.61)
.71b
.85 (8.67)
.71 (7.65)
.64 (7.47)
.89b
.89 (19.95)
.84 (14.51)
Role conict (a = .78; CR = .79; AVE = .50) (Rizzo et al., 1970) (seven-point scale, anchored by Never and Always)
I work with two or more groups who operate quite differently
I receive incompatible requests from two or more people
I do things that are apt to be accepted by one person and not accepted by another
I receive an assignment without adequate resources and material to execute it
Role ambiguity (a = .90; CR = .90; AVE = .64) (Rizzo et al., 1970) (seven-point scale, anchored by Never and Always)
I feel certain about how much authority I have (R)
I have clear, planned goals and objectives for my job (R)
I know what my responsibilities are (R)
I know exactly what is expected of me (R)
Explanation is clear of what has to be done (R)
Job satisfaction (second-order construct) (Churchill, Ford, & Walker, 1974) (seven-point scale, anchored by Strongly Disagree
and Strongly Agree)
A. Satisfaction with overall job (a = .93; CR = .93; AVE = .78)
My work gives me a sense of accomplishment
My work is exciting
My work is satisfying
I am really doing something worthwhile in my job
B. Satisfaction with Fellow Workers (a = .85; CR = .87; AVE = .63)
The people I work with are very friendly
The people I work with help each other out when someone falls behind or gets in a tight spot
My fellow workers are pleasant
My fellow workers are selsh (R)
C. Satisfaction with supervision (a = .90; CR = .90; AVE = .70)
My immediate manager is up to date
My immediate manager has always been fair in dealings with me
My immediate manager gives us credit and praise for work well done
My immediate manager lives up to his/her promises
D. Satisfaction with customers (a = 0.87; CR = 0.87; AVE = 0.63)
My customers live up to their promisesb
My customers are trustworthy
My customers are loyal
My customers are understanding
.62
.84
.75
.57
(7.16)
(10.35)
(9.07)
(6.52)
.76
.75
.87
.87
.74
(9.71)
(9.58)
(11.80)
(11.91)
(9.33)
.71 (6.48)
.87b
.88 (13.69)
.93 (15.13)
.86 (13.13)
.67 (6.02)
.86b
.84 (11.80)
.90 (12.97)
.51 (5.99)
.60 (5.15)
.71b
.82 (8.91)
.95 (9.98)
.85 (9.20)
.52 (4.43)
.72b
.84 (8.78)
.83 (8.66)
.78 (8.25)
G Model

Table 1 (Continued )
Constructs and measurement items
Standardised loadingsa
Propensity to leave (a = .91; CR = .89; AVE = .73) (Bluedorn, 1982)

(seven-point scale, anchored by Very Low and Very High)
I am going to quit my job during the next three months
I am going to quit my job during the next six months
I am going to quit my job during the next year
.79 (10.19)
.90 (12.18)
.88 (11.76)
Psychic distance (a = .89; CR = .89; AVE = .63) (Klein & Roth, 1990) (seven-point scale, anchored by Very Different and Very Similar)
Culture (traditions, values, language, etc)
Accepted business practices
Economic environment
Legal system
Communications infrastructure
.71
.84
.89
.74
.75
(8.88)
(11.26)
(12.24)
(9.44)
(9.51)
Export sales managers experience (a = .88; CR = .85; AVE = .58) (Teas, 1983; Walker et al., 1975)
The time you have been with your company
The time you have been with the specic export venture/line
The time you have been with the export venture market
The time you have been with the companys export department
.79
.68
.65
.91
(9.98)
(8.26)
(7.73)
(12.00)
Notes: R denotes reverse items.

a
t-Values are in parentheses.
b
Fixed item.
model. Fit statistics (x2(339) = 626.69, p < .05; comparative t

index [CFI] = .92, nonnormed t index [NNFI] = .91, and root mean
square error of approximation [RMSEA] = .07) suggest a good t to
the data. Moreover, with one exception, the structural paths are
signicant and in the expected direction, supporting the research
hypotheses. The job satisfaction variance explained by this model
is 33% and for propensity to leave this coefcient of determination
is 54%.
More specically, in support of H1 and H2, formalization results
in lower levels of export sales managers role ambiguity (H1:
b = .24, t = 2.43) and role conict (H2: b = .31, t = 2.80). On
the other hand, centralization relates positively to both role
ambiguity (H3: b = .19, t = 1.98) and role conict (H4: b = .22,
t = 2.09). Both formalization (H5: b = .20, t = 2.11) and centralization (H6: b = .36, t = 3.76) facilitate the implementation of an
export sales management behavioral control system. In turn,
export sales management control results in lower levels of export
sales managers role ambiguity (b = .21, t = 1.97). However,
contrary to expectation, export sales management control has no
inuence on role conict (H8: b = .01, t = .10). Role ambiguity
(H9: b = .43, t = 4.21) and role conict (H10: b = .33, t = 3.12)

have a signicant negative impact on job satisfaction. Finally, as
expected, job satisfaction has a strong negative association with
export sales managers propensity to leave the organization (H11:
b = .74, t = 5.95).
4.3. Moderating effects
This study tests for moderating effects using multiple group
analyses, splitting the initial sample into two sub-samples based
on the median value of each individual moderator. With regard to
psychic distance, a median split separates rms into high distance
(n1 = 81) and low distance (n2 = 77) export markets. Similarly,
export sales managers are grouped into more experienced (n1 = 79)
and less experienced (n2 = 79). For each moderator, this procedure
estimates an equal model, in which all paths of the structural
model are set equal across the two groups, and a free model in
which all paths are constrained to be equal across the two groups,
except for the path that is potentially affected by the moderator
variable. A signicant decrease in chi-square from the equal model
Table 2
Descriptive statistics and correlation matrix.
1. Formalization
2. Centralization
3. Activity information
4. Activity rewards
5. Activity punishment
6. Capability information
7. Capability rewards
8. Capability punishment
9. Role ambiguity
10. Role conict
11. Satisfaction with overall job
12. Satisfaction with fellow workers
13. Satisfaction with supervision
14. Satisfaction with customers
15. Propensity to leave
16. Psychic distance
17. Export sales manager experience
Mean Score
Standard deviation
.08
.20
.20
.06
.24
.15
.02
.29
.31
.04
.06
.11
.11
.09
.08
.14
5.05
1.35
.13
.01
.42
.19
.27
.40
.15
.22
.14
.39
.18
.08
.11
.09
.19
2.13
1.34
.60
.42
.73
.50
.39
.17
.04
.04
.03
.22
.12
.04
.03
.00
4.55
1.65
.36
.51
.52
.28
.31
.07
.15
.01
.39
.01
.26
.07
.12
4.69
1.69
.41
.45
.77
.02
.20
.01
.14
.07
.03
.00
.09
.17
3.08
1.61
10
11
12
13
14
.56
.15
.00
.06
.10
.17
.04
.15
.04
.01
3.71
1.58
.01
.13
.01
.15
.03
.05
.01
.07
.12
2.77
1.58
.39
.37
.37
.38
.09
.34
.02
.11
2.55
1.33
.29
.32
.40
.14
.35
.17
.18
3.45
1.33
.38
.41
.38
.52
.12
.17
5.47
1.15
.34
.26
.34
.04
.25
5.45
1.02
.20
.36
.01
.06
5.06
1.30
.10
.01
.08
4.66
1.10
15
16
17
.03
3.67
1.47
9.55
7.23
.58
.53
.17
.01
.05
.03
.32
.06
.03
.01
.01
3.85
1.64
.02
.14
1.70
1.48
G Model

10
Table 3
Standardised path coefcients and t-values for the structural model.
Hypothesised paths
Expected sign
H1 Formalization ! role ambiguity

H2 Formalization ! role conict
H3 Centralization ! role ambiguity
H4 Centralization ! role conict
H5 Formalization ! sales management behavior control
H6 Centralization ! sales management behavior control
H7 Sales management behavior control ! role ambiguity
H8 Sales management behavior control ! role conict
H9 Role ambiguity ! job satisfaction
H10 Role conict ! job satisfaction
H11 Job satisfaction ! propensity to leave

+
+
+
+

Standardised coefcient
t-Value
*
.24
.31
.19
.22
.20
.36
.21
.01
.43
.33
.74
2.43
2.80**
1.98*
2.09*
2.11*
3.76**
1.97*
.10
4.21**
3.12**
5.95**
Hypothesis test
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Not supported
Supported
Supported
Supported
Note: t statistics for structural model: x2339 = 626.69, p < .01; x2/d.f = 1.85; CFI = .92; NNFI = .91; RMSEA = .07.
**
p < .01.
*
p < .05.
to a model in which one relationship is set free implies that the

moderator variable has a signicant inuence on that relationship.
Table 4 presents the results obtained from this analysis.
Concerning psychic distance, the results indicate that the
negative inuence of formalization on role ambiguity is stronger
when export sales managers operate in more distant export
markets (high-distance group: b = .44, t = 3.17; low-distance
group: b = .02, t = .19; Dx2(1) = 10.74, p < .05). The positive impact
of centralization on role conict is also stronger in more distant
markets (high-distance group: b = .36, t = 2.58; low-distance
group: b = .05, t = .38; Dx2(1) = 6.57, p < .05). These ndings
provide support for H12 and H15. By contrast, psychic distance
does not moderate the relationship between formalization and role
conict (high-distance group: b = .40, t = 2.73; low-distance
group: b = .16, t = 1.16; Dx2(1) = 2.47, p > .05) or the relationship between centralization and role ambiguity (high-distance
group: b = .22, t = 1.79; low-distance group: b = .12, t = .95;
Dx2(1) = 1.17, p > .05). Even though the estimated coefcients
for these associations are stronger in more distant markets, chisquare differences are not statistically signicant. Therefore, no
support is provided for H13 and H14. With respect to export sales
managers experience, the ndings indicate that the export sales
management behavior control system is more effective in reducing
perceptions of role ambiguity among less experienced export sales
managers (more experienced managers groups: b = .13, t = .96;
less experienced managers group: b = .39, t = 2.83; Dx2
(1) = 4.58, p < .05). These results provide support for H16. On
the other hand, the inuence of the export sales management
control system on role conict is insignicant in both the more
experienced and less experienced groups (more experienced

managers group: b = .09, t = .64; less experienced managers group:
b = .21, t = 1.50; Dx2(1) = 4.45, p < .05). Although the ndings
indicate a stronger effect within the low experience group, no
conclusion can be derived for H17.
5. Discussion and implications
In response to the high turbulence that characterizes the global
business environment, exporting rms pay increasing attention to
the development of strong, trust-based, working relationships
with their foreign customers (Barnes, Leonidou, Siu, & Leonidou,
2010), which can serve as an enduring source of competitive
advantage (Zhang, Cavusgil, & Roath, 2003). The main responsibility for identifying and exploiting business opportunities in
overseas markets, and managing relationships with export
customers lies with export sales managers. Therefore, export sales
managers are a valuable asset for their rms. However, these
managers confront many challenges and opportunities in their
day-to-day job activities. The requisite variety that arises from
dealing with a diverse range of customers, requiring customized
solutions, in an array of foreign markets, present conicting
challenges that can take their toll. Although there is prima facie an
allure to this role, these demands are often reected in poor job
satisfaction ratings and high job turnover. In the sales management
literature, there are several studies that examine the determinants
of salespeoples propensity to leave their organization but this is
the rst to examine this phenomenon in the context of export sales
management.
Table 4
Results of moderation analysis.
Moderating effects of psychic distance
Main effect
Hypothesized moderating effect
High psychic
distance
Low psychic
distance
Dx2 (Dd.f. = 1)
Formalization ! role ambiguity
H12: effect is stronger in more distant export markets
b = .44
b = .02
t = 3.17
b = .40
t = 2.73
b = .22
t = 1.79
b = .36
t = 2.58
t = .19
b = .16
t = 1.16
b = .12
t = .95
b = .05
t = .38
10.74
(p < .05)
2.47
(p > .05)
1.17
(p > .05)
6.57
(p < .05)
More experienced
managers
b = .13
t = .96
b = .09
t = .64
Less experienced
managers
b = .39
t = 2.83
b = .21
t = 1.50
Formalization ! role conict
Centralization ! role ambiguity
Centralization ! role conict
Moderating effects of export sales managers experience

Main effect
Hypothesized moderating effect
Sales management control ! role ambiguity
H16: effect is stronger among less experienced managers
Sales management control ! role conict
H17: effect is stronger among less experienced managers
Dx2 (Dd.f. = 1)
4.58
(p < .05)
4.45
(p < .05)
G Model

Against the backdrop of the salespersons propensity to leave,

this study is primarily interested in the control aspect of export sales
management. Given the nature of the salespersons role and their
characteristic exibility, control becomes a focal sales management
activity. Whilst researchers know a great deal about control and
how it inuences the attitudes and behavior of workers (Anderson &
Oliver, 1987; Jaworski, 1988) it is important to recognize that
control mechanisms can be established at different hierarchical
levels within an organization and particularly how top management
develop control mechanisms that dene an appropriate framework
for the operation of the sales organization, thereby determining the
specic activities to be performed by individual members. In line
with Jaworskis (1988) theoretical assumption, this study identies
organizational structure as such a control mechanism.
The study ndings reveal the favorable impact of formalization
and the unfavorable impact of centralization upon both role
ambiguity and role conict. Local market adaptation and the sales
functions ability to customize their engagement activity are
central to export strategy (Cavusgil & Zou, 1994). When operating
under conditions of centralization, this structural constraint
handicaps the export sales managers ability to adapt according
to the necessary conditions and role conict and role ambiguity
become inevitable consequences. Centralization effectively
becomes a straightjacket and removes the discretion that is
required for the export sales manager to confront challenges in vivo
and address appropriate response to customers in export markets.
In contrast, formalization provides a framework and offers rules
and procedures that govern role behavior. Although standardized,
these rules encourage specic action responses and reduce
confusion and judgmental reasoning. These benets mean that
role clarity and role ambiguity are less apparent under conditions
of high formalization because many of the sources of these role
stressors have been managed out by standardized practices.
As forms of hierarchical control, the study nds that both
formalization and centralization relate positively to the export sales
management behavior control system. In designing and administering a behavior-based control system, managers need to collate, map,
and audit important export sales activities and capabilities. The
ndings show that the administrative mechanisms that therefore
underlie this type of control system are traits of formalization for
codication and centralization for uniformity. Interestingly, the
ndings also reveal how the export sales management behavior
control system reduces role ambiguity by improving the clarity of
export sales managers expectations because it focuses attention
onto the controlled activities and capabilities. But this nding was not
mirrored for export sales managers perceptions of role conict upon
which behavior-based control has no inuence.
Although seemingly paradoxical, the working life of the export
sales manager benets from formalization while simultaneously
suffers from centralization. The implication of this is clear
exporting rms need to improve standardization of work routines
and operating methods without correspondingly centralizing
decision making. The challenge of this organizational solution
provides tangible benets for the outreach activities of the export
sales manager. By formally delineating clearly the routines and
practices required whilst also ensuring that centralized processes
do not hinder their day-to-day responsibilities, export sales
managers will benet in their work roles (Steward, Walker, Hutt,
& Kumar, 2010). The role stressors of role ambiguity and role
conict deleteriously affect export sales managers job satisfaction.
It is not surprising that when these managers suffer an array of
demands, each at variance with another, their job satisfaction
declines. Correspondingly, this study nds that export sales
managers propensity to leave is a likely consequence of this
chain of effects. This, in turn, can set in train a number of
consequences for management attention and so auditing the
11
extent of role ambiguity and role conict by encouraging feedback

and open lines of communication will help to support more this
key boundary spanner (Piercy, Cravens, Lane, & Vorhies, 2006).
Moderation analysis results indicate that psychic distance
explains important distinctions in the proposed model. For
example, in more distant markets, formalization has a stronger
negative impact on export sales managers role ambiguity.
Similarly, the challenges that centralization imposes on export
sales managers role conict are also increasingly evident in more
distant markets. For exporting rms, these emphases imply a
greater need for formalization and lesser need for centralization in
improving, respectively, role ambiguity and role conict of their
export sales managers. By way of export sales managers
experience, this study nds a novel effect in that the export sales
management behavioral control system is more effective at
reducing perceptions of role ambiguity among less experienced
export sales managers. This means that the conformity that control
systems engender invokes role clarity at earlier stages of an export
sales managers career when learning and development opportunities are at their greatest. Therefore, appropriate training needs to
be established so that less experienced export sales managers
suffer less role ambiguity by better understanding the control
dashboard that is used to monitor their behavior.
This study has also important implications for theory development in the neglected areas of export sales management. In
particular, it highlights the critical role of export sales managers in
export success, and emphasizes the need to retain these personnel
in the rm on a long-term basis. The extant research on exporting
focuses on the export strategic planning process, and integrates
relevant theories like the resource-based view of the rm, and the
industrial organization theory, in an effort to explain the drivers of
superior export performance. Numerous studies on exporting
during the past four decades have contributed to the development
of a rich body of knowledge regarding export-related issues.
Surprisingly; however, these studies paid limited attention to the
fact that export sales managers are commonly the only rm
personnel that visit foreign markets to meet prospective buyers
and establish mutually benecial relationships with them.
Therefore, their contribution to export success is invaluable. To
advance further export-related theories, academic research must
devote more effort on the peoples aspect of exporting, and
examine appropriate managerial practices that can positively
inuence the job-related attitudes and behavior of export sales
managers, and direct their behavior towards achieving organizational objectives. The present study contributes to this emerging
stream of research by showing how control mechanisms at
different organizational levels contribute towards reducing export
sales managers intention to leave, which is a highly desirable
outcome with important long-term benets for export sales
organizations. Future studies must investigate other important
dimensions of export sales managers job, and identify the key
factors that either enhance or restrain export sales managers
performance.
6. Limitations and future research
This study has several limitations that should be addressed in
future research. First, the study accommodates for international
market variation in the research model by incorporating a measure
of psychic distance. Although the research model explains over half
of the variance about the mean of managers propensity to leave,
inevitably testing the extent to which competing explanations
offer insightful viewpoints is important. To this end, individual
level endogenous triggers such as managers perceived ease of
movement, receipt of unsolicited job offers, and work-family
conict should be considered while also rm level exogenous
G Model

12
effects such as stage of export sales growth, sales strategy success,

and export resources and capabilities may offer new insights into
widening the hitherto conservative explanada on quitting intentions and behavior.
Second, this study is cross-sectional in design and the
hypothesized associations are inferred as it was not possible to
determine cause-effect relationships. The results may be potentially vulnerable to opposing or bidirectional relationships in that,
for example, control systems may potentially determine patterns
of formalization and centralization. The arguments against such
reverse causality can be sourced to compelling theoretical
evidence and persuasive reasoning from extant studies. However,
because alternative explanations cannot be ruled out, both
longitudinal and experimental studies are suggested as a means
to overcome this limitation and demonstrate the causation that
this study infers.
Third, this study uses data collected from UK exporters.
Previous research has shown that the factors that inuence
salespeoples work attitudes, behaviors, and performance are
generally consisted across cultures (e.g., Baldauf et al., 2002; Piercy,
Low, & Cravens, 2004). Despite this evidence, however, the
generalizability of our ndings to other national contexts should
be further tested through replication studies in other developed as
well as developing countries.
Fourth, the ultimate dependent variable in our proposed
conceptual model is export sales managers propensity to leave
the export sales organization. Available empirical evidence
indicates that propensity to leave is a strong predictor of actual
turnover behavior (e.g., DeConinck, 2011; Griffeth, Hom, &
Gaertner, 2000; Johnston et al., 1990). On the other hand, however,
turnover intentions do not translate perfectly in predicting actual
turnover (DeConinck & Johnson, 2009). Thus, a fruitful avenue for
future research is to investigate the antecedents of actual export
sales managers turnover, and try to uncover the factors that
determine whether the formation of intentions to quit will
eventually lead to subsequent quitting behavior (Chandrashekaran, Mc Neilly, & Marinova, 2000).
Finally, the study examines the inuence of organizational
conditions on export sales managers job satisfaction. Previous
research has shown that the level of an employees intrinsic
motivation and other psychosocial conditions mediates this
relationship. Therefore, future research should expand the
nomological network of relationships considered in this study
so as to account for these inuences.
Acknowledgements
The authors would like to thank the editor and the two
anonymous IBR reviewers for their insightful comments and
constructive suggestions.
Appendix
Sample characteristics.
Number of full time employees

Number of staff fully/partly
involved in export activities
Firm export experience
(years in exporting)
Number of export countries
Age of responding
export sales managers
Respondents gender:
male 81%, female: 19%
Mean
Standard
deviation
Min
Max
132
10
115
20
16
3
500
110
25
16
80
28
44
29
11
3
24
110
67
References
Agarwal, S. (1993). Inuence of formalisation on role stress, organisational commitment, and work alienation of salespersons: A cross-national comparative study.
Journal of International Business Studies, 24, 715739.
Ahearne, M., Rapp, A., Hughes, D. E., & Jindal, R. (2010). Managing sales force product
perceptions and control systems in the success of new product introductions.
Journal of Marketing Research, 47, 764776.
Aiken, M., & Hage, J. (1966). Organisational alienation: A comparative analysis. American Sociological Review, 31, 497507.
Aiken, M., & Hage, J. (1968). Organisational independence and intra-organisational
structure. American Sociological Review, 33, 912930.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modelling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103, 411423.
Anderson, E., & Oliver, R. (1987). Perspectives on behaviour-based versus outcomebased sales force control systems. Journal of Marketing, 51, 7688.
Armstrong, S. J., & Overton, T. S. (1977). Estimating non-response bias in mail surveys.
Journal of Marketing Research, 14, 396402.
Auh, S., & Menguc, B. (2007). Performance implications of the direct and moderating
effects of centralisation and formalisation on customer orientation. Industrial
Marketing Management, 36, 10221034.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of
the Academy of Marketing Science, 16, 7494.
Baldauf, A., Cravens, D. W., & Grant, K. (2002). Consequences of sales management
control in eld sales organizations: A cross national perspective. International
Business Review, 11, 577609.
Baldauf, A., Cravens, D. W., & Piercy, N. F. (2005). Sales management control researchsynthesis and an agenda for future research. Journal of Personal Selling and Sales
Management, 25, 726.
Barnes, B. R., Leonidou, L. C., Siu, N. Y., & Leonidou, C. N. (2010). Opportunism as the
inhibiting trigger for developing long-term-oriented Western exporterHong Kong
importer relationships. Journal of International Marketing, 18, 3563.
Bentler, P. M., & Chou, C.-P. (1987). Practical issues in structural modelling. Sociological
Methods and Research, 16, 78117.
Bentler, P. M. (1995). EQS structural equations program manual. Multivariate software.
California: Encino.
Bluedorn, A. C. (1982). A unied model of turnover from organisations. Human
Relations, 35, 135153.
Brown, S. P., & Peterson, R. A. (1993). Antecedents and consequences of salesperson job
satisfaction: Meta analysis and assessment of causal effects. Journal of Marketing
Research, 30, 6377.
Cavusgil, S. T., & Zou, S. (1994). Marketing strategy-performance relationship: An
investigation of the empirical link in export market ventures. Journal of Marketing,
58, 121.
Challagalla, G. N., & Shervani, T. A. (1996). Dimensions and types of supervisory control:
Effects on salesperson performance and satisfaction. Journal of Marketing, 60,
89105.
Chandrashekaran, M., Mc Neilly, K., & Marinova, D. (2000). Fron uncertain intentions to
actual behavior: A threshold model of whether and when salespeople quit. Journal
of Marketing Research, 37, 463479.
Chi, T., & Sun, Y. (2013). Development of rm export market oriented behaviors:
Evidence from an emerging economy. International Business Review, 22, 339350.
Churchill, G. A., Ford, N. M., & Walker, O. C., Jr. (1974). Measuring the job satisfaction of
industrial salesmen. Journal of Marketing Research, 11, 254260.
Cron, W. L., Dubinsky, A. J., & Michaels, R. E. (1988). The inuence of career stages on
components of salesperson motivation. Journal of Marketing, 52, 7892.
DeConinck, J. B., & Johnson, J. T. (2009). The effects of perceived supervisor support,
perceived organizationla support, and organizational justice on turnover among
salespeople. Journla of Personal Selling and Sales Management, 29, 333350.
DeConinck, J. B. (2011). The effects of ethical climate on organizational identication,
supervisory trust and turnover among salespeople. Journal of Business Research, 64,
617624.
Dillman, D. A. (2000). Mail and internet surveys. New York, NY: Wiley.
Fishbein, M. (1967). Readings in attitude theory and measurement. Oxford: Wiley.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error: Algebra and statistics. Journal of
Marketing Research, 28, 3950.
Fry, L. W., Futrell, C. M., Parasuraman, A., & Chmielewski, M. A. (1986). An analysis of
alternative causal models of salesperson role perceptions and work-related attitudes. Journal of Marketing Research, 23, 15311563.
Good, L. K., Page, T. J., & Young, C. E. (1996). Assesssing hierarchical differences in jobrelated attitudes and turnover among retail managers. Journal of the Academy of
Marketing Science, 24, 148156.
Griffeth, R. W., Hom, P. W., & Gaertner, S. (2000). A meta-analysis of antecedents and
correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26, 463488.
Hall, R. H., Johnson, N. J., & Haas, E. J. (1967). Organisational size, complexity and
formalisation. American Sociological Review, 32, 903912.
Homburg, C., & Stock, R. M. (2004). The link between salespeoples job satisfaction and
customer satisfaction in a business-to-business context: a dyadic analysis. Journal
of the Academy of Marketing Science, 32, 144158.
Jaramillo, F., Mulki, P. J., & Boles, J. S. (2013). Bringing meaning to the sales job: The
effect of ethical climate and customer demandiness. Journal of Business Research,
66, 23012307.
Jaworski, B. J., & Kohli, A. K. (1993). Market orientation: Antecedents and consequences.
Journal of Marketing, 57, 5370.
G Model

Jaworski, B. J., Stathakopoulos, V., & Krishan, S. H. (1993). Control combinations in
marketing: Conceptual framework and empirical evidence. Journal of Marketing,
57, 5769.
Jaworski, B. J. (1988). Towards a theory of marketing control: Environmental context,
control types and consequences. Journal of Marketing, 52, 2339.
JDH Group (2011). Employee turnover cost calculator. Ohio: JDH Group.
Johanson, J., & Vahlne, J.-E. (1977). The internationalisation process of the rmA
model of knowledge development and increasing foreign market commitments.
Journal of International Business Studies, 8, 2332.
Johnson, J. S., & Sohi, R. S. (2014). The curvilinear and conditional effects of product line
breadth on salesperson performance, role stress and job satisfaction. Journal of the
Academy of Marketing Science, 42, 7189.
Johnston, M. W., Varadarajan, R. P., Futrell, C. M., & Sager, J. K. (1987). The relationship
between organisational commitment, job satisfaction and turnover among new
salespeople. Journal of Personal Selling & Sales Management, 7, 2938.
Johnston, M. W., Parasuraman, A., Futrell, C. M., & Black, W. C. (1990). A longitudinal
assessment of the impact of selected organisational inuences on salespeoples
organisational commitment during early employment. Journal of Marketing Research, 27, 333344.
Klein, S., & Roth, V. J. (1990). Determinants of export channel structure: The effects of
experience and psychic distance reconsidered. International Marketing Review, 7,
2738.
Lee, T.-H., Gerhart, B., Weller, I., & Trevor, C. O. (2008). Understanding voluntary
turnover: Path-specic job satisfaction effects and the importance of unsolicited
job offers. Academy of Management Journal, 51, 651671.
Leonidou, L. C., Katsikeas, C. S., & Samiee, S. (2002). Marketing strategy determinants of
export performance: A meta analysis. Journal of Business Research, 55, 5167.
Lindell, M. K., & Whitney, D. J. (2001). Accounting for common method variance in
cross-sectional designs. Journal of Applied Psychology, 86, 114121.
Malhotra, N. K., Kim, S. S., & Patil, A. (2006). Common method variance in IS research: A
comparison of alternative approaches and a reanalysis of past research. Management Science, 52, 18651883.
Mallin, M. L., Asree, S., Koh, A. C., & Hu, M. Y. (2010). Antecedents to managerial
trust and sales control in Malaysian sales force. International Business Review, 19,
292305.
Michaels, R. E., Cron, W. L., Dubinsky, A. J., & Joachimsthaler, E. A. (1988). Inuence of
formalisation on the organisational commitment and work alienation of salespeople and industrial buyers. Journal of Marketing Research, 25, 376383.
Mobley, W. H. (1977). Intermediate linkages in the relationship between job satisfaction and employee turnover. Journal of Applied Psychology, 62, 237240.
Morgan, N. A., Kaleka, A., & Katsikeas, C. S. (2004). Antecedents of export venture
performance: A theoretical model and empirical assessment. Journal of Marketing,
68, 90108.
Morgan, N. A., Katsikeas, C. S., & Vorhies, D. W. (2012). Export marketing strategy
implementation, export marketing capabilities and export venture performance.
Journal of the Academy of Marketing Science, 40, 271289.
Mowday, R. T., Koberg, C. S., & McArthur, A. W. (1984). The psychology of the
withdrawal process: A cross validational test of Mobleys intermediate linkages
model of turnover in two samples. Academy of Management Journal, 27, 7994.
Murray, J. Y., Gao, G. Y., & Kotabe, M. (2011). Market orientation and performance
of export ventures: The process through which marketing capabilities and
competitive advantages. Journal of the Academy of Marketing Science, 39, 252
269.
13
Netemeyer, R. G., Johnston, M. K., & Burton, S. (1990). Analysis of role conict and role
ambiguity in a structural equations framework. Journal of Applied Psychology, 75,
148157.
Netemeyer, R. G., Brashear-Alejandro, T., & Boles, J. S. (2004). A cross-national model of
job-related outcomes of work role and family role variables: A retail sales context.
Journal of the Academy of Marketing Science, 32, 4960.
Noble, C. H. (2008). The inuence of job security on eld sales manager satisfaction:
Exploring frontline tensions. Journal of Personal Selling & Sales Management, 28,
247261.
Piercy, N. F., Cravens, D. W., Lane, N., & Vorhies, D. W. (2006). Driving organisational
citizenship behaviours and sales person in-role behaviour performance: The role of
management control and perceived organisational support. Journal of the Academy
of Marketing Science, 34, 244262.
Piercy, N. F., Low, G. S., & Cravens, D. W. (2011). Country differences concerning sales
organisation and salesperson antecedents of sales unit effectiveness. Journal of
World Business, 46, 104115.
Piercy, N. F., Low, G. S., & Cravens, D. W. (2004). Consequences of sales managements
behavior- and compensation-based control strategies in developing countries.
Journal of International Marketing, 12, 3057.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organisational research:
Problems and prospects. Journal of Management, 12, 531544.
Prime, N., Obadia, C., & Vida, I. (2009). Psychic distance in exporter-importer relationships: A grounded theory approach. International Business Review, 18, 184198.
Reiche, S. B. (2008). The conguration of employee retention practices in multinational
corporations foreign subsidiaries. International Business Review, 17, 676687.
Rizzo, J. R., House, R. J., & Lirtzman, S. I. (1970). Role conict and ambiguity in complex
organisations. Administrative Science Quarterly, 15, 150163.
Russ, F. A., & McNeilly, K. M. (1995). Links among satisfaction, commitment and
turnover intentions: The moderating effect of experience, gender and performance. Journal of Business Research, 34, 5765.
Sager, J. K., Varadarajan, R. P., & Futrell, C. M. (1988). Understanding salesperson
turnover: A partial evaluation of Mobleys turnover process. Journal of Personal
Selling & Sales Management, 13, 2035.
Sager, J. K. (1994). A structural model depicting salespeoples job stress. Journal of the
Academy of Marketing Science, 22, 7484.
Singh, J., Verbeke, W., & Rhoads, G. K. (1996). Do organisational practices matter in role
processes?. A study of direct and moderating effects for marketing oriented
boundary spanners. Journal of Marketing, 60, 6986.
Sohi, R. S. (1996). The effects of environmental dynamism and heterogeneity on
salespeoples role perceptions, performance and job satisfaction. European Journal
of Marketing, 30, 4967.
Sousa, C. M. P., & Bradley, F. (2006). Cultural distance and psychic distance: Two peas in
a pod? Journal of International Marketing, 14, 4970.
Steward, M. D., Walker, B. A., Hutt, M. D., & Kumar, A. (2010). The coordination
strategies of high-performing salespeople: Internal working relationships that
drive success. Journal of the Academy of Marketing Science, 38, 550566.
Teas, K. R. (1983). Supervisory behaviour, role stress and the job satisfaction of
industrial salespeople. Journal of Marketing Research, 20, 8491.
Walker, O. C., Jr., Churchill, G. A., Jr., & Ford, N. M. (1975). Organisational determinants
of industrial salesmans role conict and ambiguity. Journal of Marketing, 39, 3239.
Zhang, C., Cavusgil, S. T., & Roath, A. S. (2003). Manufacturer governance of foreign
distributor relationships: Do relational norms exchange competitiveness in the
export market? Journal of International Business Studies, 34, 550556.

International Business Review Volume 24 Issue 3 2015 (Doi 10.1016/j.ibusrev.2014.08.009) Katsikea, Evangelia Theodosiou, Marios Morgan, Robert E. - Why People Quit - Exp

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

International Business Review Volume 24 Issue 3 2015 (Doi 10.1016/j.ibusrev.2014.08.009) Katsikea, Evangelia Theodosiou, Marios Morgan, Robert E. - Why People Quit - Exp

Transféré par

Droits d'auteur :

Formats disponibles

G Model

IBR-1140; No. of Pages 13

Contents lists available at ScienceDirect