Commitment or Control

7/28/2019 Commitment or Control

1/43

1

Commitment or control?

Human resource management practices in female and

male-led businesses

Ingrid Verheul

Centre for Advanced Small Business EconomicsErasmus University Rotterdam / Faculty of Economics

P.O. Box 17383000 DR Rotterdam

The NetherlandsTel. +31 10 [email protected]

and

EIM Business and Policy ResearchP.O. Box 7001

2701 AA Zoetermeer The Netherlands

Tel. +31 79 [email protected]

Version: January 2003

Keywords: gender, entrepreneurship, human resource management, commitment, control

Acknowledgement:The author would like to thank Paul Boselie, Jan de Kok, Gavin Reid, Heleen Stigter, Roy Thurik,Lorraine Uhlaner and Claartje Vinkenburg for their helpful comments. Ingrid Verheul acknowledgesfinancial support by the Fund Schiedam Vlaardingen e.o. and the Trust Fund Rotterdam.

Abstract: This paper investigates the extent to which HRM differs between female- and male-led businesses. A Control-Commitment Continuum consisting of several HRM dimensions is proposed. Totest to what extent HRM systems and specific practices in female- and male-led businesses differ withrespect to commitment-orientation, use is made of a panel of approximately 2,000 Dutch entrepreneurs.Contrary to what is generally believed, we find that HRM in female-led firms is more control-orientedthan in male-led firms.


2/43

2

1 Introduction

Research in the field of human resource management (HRM) has demonstrated that the shaping of

HRM practices depends upon factors, such as the sector in which activities are undertaken (Mowday,

1998; Ram, 1999; Curran et al., 1993), business strategy (Schuler and Jackson, 1987; Lengnick-Hall

and Legnick-Hall, 1988; Youndt et al., 1996) and firm size (De Kok and Uhlaner, 2001; Ram, 1999).

Small firms usually face impediments to structuring HRM practices because they lack the time, money

and employees to formalize these issues (Hornsby and Kuratko, 1990; Deshpande and Golhar, 1994;

Marlow and Patton, 1993; Jackson et al., 1989; De Kok and Uhlaner, 2001). In many small businesses

functional areas, such as finance, marketing and production, seem to have precedence over HRM

(McEvoy, 1984).

There is also variation within the small business sector. As Nooteboom (1993, p. 287) argues: It is

always a delicate matter to make general statements about SME, because its diversity may be its most

important characteristic. This diversity counts especially when comparing the organization of the

businesses of female and male entrepreneurs. Because women and men have a different profile, e.g.,

they differ regarding educational and experiential background, risk attitude and time invested in the

business (Verheul and Thurik, 2001), they may be expected to differ regarding the management of their

business, and human resource management in particular. Although researchers in the field of

entrepreneurship have shown that female entrepreneurs differ from their male counterparts in areas,

such as motivation for start-up (Buttner and Moore, 1997), performance (Kalleberg and Leicht, 1991;

Gundry and Welsch, 2001) and finance (Carter and Rosa, 1998; Verheul and Thurik, 2001) and their

number has increased in the last decades few studies have focused upon gender differences in small

business management (Brush, 1992; Carter, 1993). Even fewer studies have attempted to shed light on


3/43

3

gender differences in HRM (Verheul et al., 2002). A study by Verheul et al. (2002), using data of 28

real estate agents, shows that gender differences in HRM indeed exist.

Several types of studies can be distinguished when comparing management styles of women and men.

The management literature generally provides inconclusive evidence regarding the question whether

women and men are different managers or leaders. In the scientific management literature (Gilligan,

1982; Ely, 1994; Grant, 1988) as well as in the popular management literature (Helgesen, 1990;

Rosener, 1990, 1995; Loden, 1985) it is argued that women and men adopt different management

styles. Others claim that, when controlling for the context within which women and men run their

business, the ways in which they behave are fairly similar (Dobbins and Platz, 1986). A different, but

related, discussion surrounds the question whether a distinction can be made between masculine and

feminine management styles, where both women and men can adopt masculine or feminine styles

(Van Engen, 2001). Most research investigating gendered management styles focuses on large

businesses.

The present study explores gender effects in HRM in small businesses. We argue that gender can have

both a direct and an indirect effect on HRM (Figure 1). In addition to gender, other factors can

influence the shaping of HRM practices, including firm size, sector, goals and strategy and firm age. In

Figure 1 these factors are referred to as the business profile. We take these factors into account in the

present study, i.e., the business profile is controlled for, because gender can be expected to influence

the factors making up the business profile (Verheul and Thurik, 2001).

Figure 1: Gender and HRM

Business profile

Gender

HRM

Business profile

Gender

HRM


4/43

4

Several perspectives have been used to compare HRM practices, including the extent to which they

contribute to increased firm performance (Guest, 1997; Huselid, 1995; Huselid et al., 1997; Ichniowski

et al., 1997; Koch and McGrath, 1996; Paauwe, 1998) and level of sophistication (Arthur and Hendry,

1990; Deshpande and Golhar, 1994; Duberley and Walley, 1995; Hornsby and Kuratko, 1990; Koch

and McGrath, 1996; Marlow and Patton, 1993). As noted by Boselie (2002, p. 40) there are also

different approaches to studying commitment and control, including the focus on structure (Mintzberg,

1979; Verburg, 1998; Paauwe, 1989), culture (Legge, 1995), strategy (Walton, 1985) and HRM

(Arthur, 1994).

Following Boselie (2002), and making use of the work of Beer et al. (1984), Walton (1985) and Arthur

(1992; 1994), in the present study a distinction is made between those HRM practices that focus upon

enhancing employee commitment and those practices that increase control of the owner-manager over

employees and the production process. These two aspects of HRM practices are considered the

extremes on a continuum, where HRM practices tend to be either more commitment- or more control-

oriented. The research question in this study is whether HRM practices in female and male-led

businesses differ on the Control-Commitment Continuum. Hypotheses are formulated on the influence

of gender on both the HRM system and a range of HRM dimensions. These hypotheses are tested using

SME Policy Panel data of EIM Business and Policy Research in Zoetermeer in the Netherlands. Every

four months approximately 2,000 Dutch entrepreneurs participate in this SME Policy Panel which is

used for both cross-sectional and longitudinal research.

The set-up of the present paper is as follows. Section 2 focuses upon the Control-Commitment

Continuum as first proposed by Walton (1985) and put in the context of HRM in studies by Arthur


5/43

5

(1994), Godard (1998) and Boselie (2002). Several HRM dimensions on the Control-Commitment

Continuum are specified, which are operationalized in the empirical study. In Section 3 attention is

paid to the influence of gender and other factors on HRM. Hypotheses are formulated on the influence

of gender and the business profile on both the HRM system and a range of HRM dimensions. In

Section 4 data and methodological issues are discussed. Making use of factor analysis HRM variables

are constructed that are used for further analyses. In Section 5 correlations between dependent and

independent variables in the study are investigated. The hypotheses are tested, making use of

regression analysis techniques, and the results are discussed. Section 6 concludes, summarizing and

discussing the most important findings and giving suggestions for further research. In addition, the

limitations of the study are discussed.

2 The Commitment-Control Continuum

2.1 Commitment versus Control HRM Systems

The distinction between commitment and control can be traced back to McGregors (1960) Theory X

and Y, referring to the tension between the instrumental rationality of bureaucratic systems and the

affective needs of employees or the need to achieve both control and consent of employees to maintain

or improve performance (Legge, 1995) 1. Other classifications that show some resemblance to the

control and commitment dichotomy are autocratic versus democratic decision-making (Lewin and

Lippitt, 1938), task-oriented versus interpersonal oriented styles (Bales, 1950; Blake and Mouton,

1964), transactional versus transformational leadership (Bass et al., 1996), direct control versusresponsible autonomy (Friedman, 1977), Tannenbaum and Smiths (1958) continuum (tell-sell-consult-

join) and the categories of employer control (fraternalism-paternalism-benevolent autocracy-sweat

shop) proposed by Goss (1991). These management modes either emphasize maintenance of tasks


6/43

6

through direct forms of control or nurturing of interpersonal relationships through indirect or self-

control of employees (Van Engen, 2001).

Walton (1985) explicitly proposes the distinction between commitment and control strategies and this

distinction is further elaborated in the context of HRM by other authors (Guest, 1987; Arthur, 1992,

1994; Legge, 1995; Godard, 1998). Commitment and control are two distinct ways in which employee

behaviors and attitudes can be influenced (Arthur, 1994). Given the assumption that HRM consists of a

series of internally consistent HRM practices, which combine into a specific HRM system, it can be

argued that HRM systems are either control- or commitment-oriented. Control HRM systems are

characterized by a division of work into small, fixed jobs for which individuals can be held

accountable, and direct control with managers supervising rather than facilitating employees (Walton,

1985). This type of HRM system aims at reducing direct labor costs, or improve efficiency, by

enforcing employee compliance with specified rules and procedures (Walton, 1985; Eisenhardt, 1985).

In contrast, commitment HRM systems are characterized by managers who facilitate rather than

supervise., i.e., there is indirect control. This type of HRM system aims at establishing (psychological)

links between organizational and personal goals. Commitment here is seen as an individuals bond with

an organization, i.e., attitudinal (affective) commitment. Establishing a link between employee

commitment and firm performance, through behavioral commitment, is not within the scope of the

present paper.

Boselie (2002, p. 40) argues that in most cases the issue of control versus commitment is

dominated by ethical considerations, the authors treating commitment as deirable and control as

unwanted. In addition, according to Boselie (2002), commitment fits in with the values of a modern

1 For a detailed discussion of the origin of the distinction between control and commitment we refer to Boselie (2002).


7/43

7

(Western) society 2. Using a normative approach, Guest (1987, p. 513) argues that: committed

employees will be more satisfied, more productive and more adaptable. In this study we take a

descriptive approach rather than a normative approach as we attempt to catagorize HRM practices and

do not pass judgement on which approach has precedence over the other.

2.2 Dimensions of the Commitment-Control Continuum

In the present paper a distinction is made between HRM practices that can be labeled approximately as

control- or commitment-oriented. As the basis for this study we use the dimensions of the traditional

versus high-commitment work system as proposed in Beer et al. (1984). Following Walton (1985),

Arthur (1994) and Boselie (2002) these two work systems are labeled control and commitment HRM

systems, respectively. According to Arthur (1994) control HRM systems focus on cost reduction, or

improvement of efficiency, whereas commitment HRM systems emphasize employee development and

trust. The HRM systems are bundles of coherent HRM practices, characterizing the strategic HRM

approach (Boselie, 2002). Because HRM practices usually do not add up to a coherent system

(Duberley and Walley, 1995; Legge, 1995; De Kok et al., 2002), in the present study we distinguish

between a range of dimensions of the HRM system.

Table 1 presents the distinction between control and commitment HRM systems we use as the basis for

our study. It is a continuum where the HRM dimensions (e.g., job scope, task assignment, etc.) differ

with respect to their commitment-orientation. The selection of HRM dimensions presented in Table 1

is a combination of the dimensions as presented in Beer et al. (1984) and Arthur (1992; 1994). We also

include the distinction between formal and informal HRM systems, representing the degree to which

2 The Western world is also changing. Audretsch and Thurik (2000; 2001) refer to a change from a managed to anentrepreneurial economy. Such a shift is likely to bring about a change in the nature of the employment relationship (Klein,2001). The modern economy is characterized by an increasing importance of small firms. Commitment may play a differentrole here than in large firms. Moreover, Meyer et al. (1998) argue that commitment is not viewed the same way it was when


8/43

8

procedures and regulations are formalized. The higher the degree of formalization, the higher the

degree of (direct) control over employees and the production process.

Most dimensions in Table 1 can clearly be divided into a control and commitment side. For instance,

indirect supervision is a commitment feature, whereas direct supervision is a control feature. However,

explicitly paying attention to the learning process of employees can enhance both commitment

employees are involved and willing to make efforts for the organization and control learning is a

tool for successfully pursuing cost reduction (Boselie, 2002).

Table 1: Commitment-Control Continuum

HRM Dimension Commitment ControlJob scope Broadly defined jobs Narrowly defined jobsTask assignment Task differentiation Fixed tasksSupervision Indirect DirectOrganizational structure Informal Formal

Learning Structured learning (explicit) Learning-by-doing (implicit)Training General SpecificEmployee role Team member IndividualInformation sharing Global (firm) information Local (task) informationStatus Not important ImportantEmployee participation High LowDecentralization High LowSocial activities Important UnimportantBasis of payment Skills mastered Job contentSource: adapted from Beer et al. (1984) with input from Arthur (1992; 1994) and Boselie (2002).

employees could spend their entire career with a single company. Some argue that commitment is losing its importance inthe modern economy (Baruch, 1998), others suggest that commitment is more important than ever (Mowday, 1998).


9/43

9

3 Determinants of the Commitment-Orientation of HRM

Management styles tend to be contextual. They vary with environmental characteristics, e.g., stability

versus uncertainty, as well as organizational features, such as firm size, industry or sector, business

strategy and firm age or stage of development (i.e., the business profile). The present study assumes an

interaction of gender, the business profile and HRM (see Figure 1). In the empirical analysis we will

control for the business profile characteristics to single out the direct gender effect 3. The influence of

gender and each of the business profile characteristics on the extent to which HRM systems are

commitment- or control-oriented is discussed and hypothesized below.

HRM systems are comprised of several HRM dimensions (see Table 1). Because these HRM

dimensions are close to the practice of HRM, in the present paper we will use the terms practices and

dimensions interchangeably. According to the strategic HRM perspective, the HRM practices within a

system should have a similar focus. However, in most organizations HRM practices do not add up to a

coherent package deriving from a long-term coherent management strategy (Duberley and Walley,

1995, p. 905). Hence, in addition to testing for a gender effect on the general focus of the HRM system,

we test for gender effects on eight of the 13 HRM dimensions as outlined in Table 1. These eight

dimensions are chosen because they are covered by the data set at our disposal (see Section 4). For the

remaining dimensions no data are available. Accordingly, nine hypotheses are formulated on the

influence of gender on the commitment-orientation of HRM, one referring to the commitment-

orientation of the HRM system and eight referring to the commitment-orientation of the separate HRM

practices within the system. Although the main focus is to investigate the influence of gender on HRM,

hypotheses are also formulated on the influence of each of the controls (i.e., business profile

3 The direct gender effect equals the effect of gender on HRM when controlled for the business profile. For a more detaileddiscussion of direct and indirect gender effects we refer to Verheul and Thurik (2001).


10/43

10

characteristics) on the commitment-orientation of the HRM system as this benefits the discussion of the

results.

3.1 Gender Differences in HRM

Gender and the HRM system

Many authors refer to more instrumental (transactional), task-oriented, autocratic styles, explicitly as

masculine leadership styles, and to interpersonally oriented, charismatic (transformational) and

democratic styles as feminine leadership styles. Whereas the masculine style often refers to a

leadership style that emphasizes maintenance of tasks, the feminine style is based on nurturing of

interpersonal relationships (Van Engen, 2001) 4.

A management style is referred to as participative or democratic if employees are consulted and are

able to participate in decision-making. If elements of consultation and delegation of decisions are not

present, a management style is referred to as autocratic (Lewin and Lippitt, 1938). A management style

is transactional when job performance is viewed as a series of transactions with employees where they

are motivated by rewards and punishments, and where the leader derives his/her power by charisma.

Instead, a transformational leadership style focuses upon getting subordinates to transform their self-

interest into the interest of the group through concern for a broader goal, i.e., motivation by inclusion,

and leader power is based on position (Bass, 1985). An interpersonally oriented leadership style

includes behavior such as supporting employees, being available, explaining procedures and looking

out for their welfare, whereas a task-oriented leadership style consists of behavior such as having

employees follow rules and procedures, maintaining high performance standards and explicitly

formulating work roles and tasks (Bales, 1950; Blake and Mouton, 1964). Rosener (1990) argues that

4 It should be born in mind that this is stereotyping and that the dichotomies of leadership styles do not necessarily coincidewith biological sex.


11/43

11

the female leadership style goes beyond the transformational and participative style, to being an

interactive style, with women positively interacting with their employees, encouraging participation

and sharing power and information. In addition to these leadership dichotomies, sometimes another

style, laissez-faire, is added indicating an absence of leadership (White and Lippitt, 1960).

Although the bulk of the management and entrepreneurship literature indicates that women tend to

engage in the more feminine leadership styles (Rozier and Hersh-Cochran, 1996; Rosener, 1990;

Chaganti, 1986; Eagly and Johnson, 1990; Kabacoff, 1998; Yammarino et al., 1997; Bass et al., 1996),

contradicting evidence was presented by Sadler and Hofstede (1976) arguing that both men and women

prefer a consult style and Mukthar (2002) who finds that female owner-managers are more autocratic.

This latter finding may be explained by the fact that because women tend to combine work and

household responsibilties, their ambitions are different from those of men. They are less likely to grow

beyond a certain treshold size, beyond which they can no longer control their activities and combine

responsibilities 5. Based on the bulk of the literature arguing that women are likely to practice

feminine leadership styles, and because these participative, transformational or people-based styles

bear a close resemblance to the commitment-oriented HRM system, we expect that HRM systems in

female-led businesses are commitment-oriented rather than control-oriented. Despite some

counterintuitive findings the following hypothesis is formulated:

H1: HRM systems in female-led businesses are more commitment-oriented [ COMMITM ] than those

in male-led businesses

Gender and HRM dimensions

In addition to the general focus of the HRM system in female-led businesses (as hypothesized in H1),

the influence of gender on the commitment-orientation of the separate HRM dimensions will be


12/43

12

discussed to create a better insight into the composition and coherency of HRM systems in female- and

male-led businesses. Hypotheses are formulated using a selection of the dimensions of the

Commitment-Control Continuum as represented in Table 1 6. Hypotheses are based on the literature

and, where no theoretical evidence is available, we assume that HRM practices will be in line with the

general commitment-orientation of HRM systems in female-led businesses, as hypothesized in H1.

Several studies find that female managers are more likely to let employees participate in decision-

making (Cromie and Birley, 1991; Neider, 1987; Stanford et al., 1995; Jago and Vroom, 1982).

Verheul et al. (2002) argue that although both men and women let employees participate in decision-

making, the degree to which employees are able to contribute is dependent upon the gender of the

entrepreneur. Although male entrepreneurs let employees participate in decision-making, they usually

make the final decisions themselves. Female entrepreneurs are assumed to be more likely to involve

employees throughout the decision-making process. According to Jago and Vroom (1982, p. 781):

Women managers may be more likely to recognize the need for commitment to decisions by others

and this may cause them to take appropriate measures to obtain that commitment in the decision-

making process. Because the existing literature indicates that employee participation is higher in

female-led businesses we expect to find the following:

H1.1: In female-led businesses there will be a higher degree of employee participation [ PARTICIP ]

than in male-led businesses

Brush (1992) describes the role of women as coordinating relationships rather than ordering people

around. Women leaders tend to focus on relationships instead of hierarchy (Buttner, 2001; Brush,

5 These counterintuitive findings may be explained by the fact that most of the studies on gender differences in managementin the field of entrepreneurship are qualitative and make use of samples that consist only of females.6 We make a selection of the Commitment-Control Continuum dimensions as we do not have information on all dimensionsin the data set. We make use of the study by Verheul et al. (2002) investigating gender differences in HRM in Dutch realestate brokerage as the basis for formulating the hypotheses as, to our knowledge this is the only study focusing upongender differences in HRM in small businesses.


13/43

13

1992; Belenky et al., 1986; Fischer and Gleijm, 1992; Stanford et al., 1995). Accordingly, Rosener

(1990) assumes a high degree of decentralization and delegation of decision-making power in

businesses headed by women. Stanford et al. (1995) view women entrepreneurs as more open to

criticism and accessible for employees. They foster relationships with employees based on mutual trust

and respect. Because it is suggested that women tend to structure their business in a non-hierarchical

manner, the following hypothesis is formulated:

H1.2: In female-led businesses there will be a higher degree of decentralization [ DECENTR ] than in

male-led businesses

Because female-led businesses are said to focus upon relationships instead of hierarchy, it is likely that

control mechanisms are structured accordingly. Indeed, Verheul et al. (2002) suggest that whilst male

entrepreneurs directly control the production process where failures or clumsiness are corrected

within the process, female entrepreneurs are more likely to make use of indirect ways of controlling

employees motivating them by encouraging commitment to the companys goals and scheduled

meetings. Such evidence leads us to believe that control mechanisms in female-led businesses tend to

be more indirect than in male-led businesses. Accordingly, the following hypothesis is formulated:

H1.3: In female-led businesses supervision is relatively indirectly structured [ INDIRECT ] as

compared to male-led businesses

Several studies suggest that businesses of women use more informal practices (Brush, 1992; Cuba et

al., 1983; Hisrich and Brush, 1987; Chaganti and Parasuraman, 1996; Rosener, 1990). The assumed

non-hierarchical structure of women-led businesses seems to give way to an emphasis on informal,

non-systematic structuring. Mukthar (2002) argues that given that female owner-managers are more

informal, and thus more flexible using little or no documented procedures, they rely on intuition for

decision making within the business. Because of this evidence and the recent argument of Mukthar


14/43

14

(2002) we expect that female-led businesses are structured more informally than male-led businesses,

leading to the following hypothesis:

H1.4: Female-led businesses have a relatively informal structure [ INFORMAL ] as compared to male-

led businesses

The alleged non-hierarchical and informal environment of female-led businesses is likely to have

consequences for the way in which jobs and tasks are structured, e.g., enhancing employee motivation,

and in turn commitment. In accordance with the assumption that HRM systems in female-led

businesses are more commitment-oriented (see Hypothesis H1), we expect jobs to be more broadly

defined and with a high degree of task differentiation, leading to the following hypotheses on the scope

and contents of jobs/tasks in female- and male-led businesses:

H1.5: In female-led businesses jobs are more broadly defined [ BROADJOB ] than in male-led

businesses

H1.6: In female-led businesses tasks are more differentiated [ TASKDIFF ] than in male-led businesses

Regarding gender differences in the degree to which attention is paid to training and development of

employees, the evidence there is (Verheul et al., 2002), suggests that in both female- and male-led

businesses attention is paid to the learning process of employees. However, in Verheul et al. (2002) it is

also suggested that female entrepreneurs are more likely to oblige their personnel to engage in training

and development than male entrepreneurs, which may be an indication of their educational demands.

Moreover, the focus of the training may differ in female- and male-led businesses. For instance, female

entrepreneurs may be more likely to pay attention to management training as the bulk of women in

management positions do not have the advantage of experience in management positions and they tend

to rely more on their employees (Cromie and Birley, 1991) or they may be expected to pay more


15/43

15

attention to social development of employees as they are thought to value collective action (Jago and

Vroom, 1982; Gibson, 1995) for which social skills are important. Because these forms of training

have a more general character, it may be that women focus more on general training. Although it is

difficult to a priori formulate hypotheses on the degree and focus of training and development in

female- and male-led businesses, the following hypotheses are proposed, based on scant evidence,

logical reasoning and intuition:

H1.7: In female-led businesses more explicit attention is paid to the learning process of employees

[ LEARN ] than in male-led businesses

H1.8: In female-led businesses there is a higher degree of general training [ TRAINGEN ] than in male-

led businesses

3.2 Business Profile and HRM

Firm size

Several studies show that firm size influences the shaping of HRM practices (Hornsby and Kuratko,

1990; Deshpande and Golhar, 1994; Marlow and Patton, 1993; Jackson et al., 1989; De Kok and

Uhlaner, 2001). The bulk of these studies argues that small businesses often spend less time on

developing and formalizing HRM practices. In addition, small-scale activities enable a more flexible,

informal and personal style with direct communication between employees and the owner-manager.

This may facilitate delegation and a high autonomy of employees in decision-making (Ram, 1999) 7.

Hence, firm size is likely to have implications for the degree of control or commitment in HRM

systems. Golhar and Deshpande (1997) and Deshpande and Golhar (1994) find that both large and

small (manufacturing) firms rank open communication, training of new employees, and employee

7 However, as tasks in a small business are often less specialized this may also limit the extent to which delegation isnecessary (Mukthar, 2002).


16/43

16

participation initiatives among the most important HRM practices. Goffee and Scase (1995) argue that

adaptiveness and job variety are distinctive characteristics of small firms. They indicate that work in

small businesses is broadly defined and employees have a high degree of discretion and responsibility.

In addition, there may be little need for direct management control as work is continuously adapted to

customer preferences.

Commitment seems particularly important in small businesses as employees are often allrounders

who are difficult to replace as they possess firm-specific (tacit) knowledge. Also, it is relatively costly

for small businesses to find new employees in the market because of the absence of economies of scale

(Nooteboom, 1993). Moreover, small firms usually have no specialized staff for different functional

areas, such as finance, marketing and personnel, and jobs are broadly defined, comprising of different

tasks. Being involved in a broad range of activities may increase the involvement of employees with

the organization.

Although there are many arguments favoring a commitment-oriented strategy in small businesses, it

has to be noted that owner-managers of small businesses often are reluctant to give up control over

their business. They want to be entrepreneur instead of manager. As Goffee and Scase (1995, p. 41/2)

note: To shift . to a management style that requires the ability to trust staff in a hands-off manner so

that systems of delegation can be established requires a fundamental change in proprietal attitude and

competence . Moreover, owner-managers of small businesses do not have to give up control over the

business as the firm is sufficiently small to participate in day-to-day operations and directly control

production. However, in general, we would expect HRM systems in small businesses to be more

commitment-oriented than those in large businesses. This leads to the following hypothesis:

H2: HRM systems in smaller businesses are more commitment-oriented than those in larger

businesses [control = firmsize ]


17/43

17

Sector: service versus non-service firms

Businesses in different sectors may be characterized by different employment cultures (Curran et al.,

1993). Moreover, Curran et al. (1993) show that also within different types of service firms there is

variation in employment culture. Accordingly, employee commitment may be more important in

certain business environments than in others. Commitment is more likely to be a goal of HRM in the

service sector than in other sectors because commited employees are important for customer loyalty

and satisfaction (Heskett et al., 1997; Hall, 1993; Maister, 1997). In service firms the relationship

between customers and employees is the key to the production process (Heskett et al., 1997, p. 98).

Mowday (1998, p. 398) shows that employee commitment to the organization may result in greater

positive returns in the service sector than in manufacturing. In the same fashion Ram (1999, p. 15/6)

argues that The loyalty and commitment of key workers in professional service firms is secured by

developing a person-centred culture characterised by high-trust work relations . Hence, it may

be argued that HRM systems are more commitment-oriented in service than in non-service firms. The

following hypothesis is formulated:

H3: HRM systems in service businesses are more commitment-oriented than those in non-service

businesses [control = service ]

Business strategy

It has been argued that business strategy influences the type of leadership or, in general the shaping of

HRM practices (Schuler and Jackson, 1987; Lengnick-Hall and Lengnick-Hall, 1988; Youndt et al.,

1996). Boselie (2002) argues that Waltons (1985) distinction between control and commitment

strategies appears to be inspired by Porters (1980, 1985) distinction between the strategies cost

reduction, focus and differentiation. According to Guthrie et al. (2002) firms adopting a differentiation

strategy also aim for high involvement work practices.


18/43

18

Youndt et al. (1996) make a distinction between three strategies, including cost, quality and flexibility.

Each of these strategies will have important implications for the shaping of HRM systems (Youndt et

al., 1996). The most efficient approach to HRM for firms minimizing costs is to adopt a command-and-

control style where emphasis is placed on efficiently managing low-skilled, manual workforce. For

firms pursuing a quality strategy the determinant of organizational competitiveness may be the

intellectual capital of the firm. In these firms there is a transition from manual labor, where

responsibilities are limited to the physical execution of work, to knowledge work with broader

responsibilities. In these firms the aim is to develop human-capital enhancing HRM systems with a

focus on training and development of employees. In the same fashion, firms pursuing flexibility

strategies require human-capital-enhancing HRM systems focusing on skill acquisition and

development in an effort to facilitate adaptability and responsiveness. This leads to the following

hypotheses regarding the influence of low-cost or low-price, quality and focus 8 strategies on the

commitment-orientation of HRM systems.

H4: HRM systems in firms focusing on minimizing costs, aiming at low prices, are less

commitment-oriented than those in firms that do not pursue this type of strategy [control =

lowprice ]

H5: HRM systems in firms pursuing a quality strategy are more commitment-oriented than those in

firms that do not pursue this type of strategy [control = quality ]

H6: HRM systems in firms pursuing a focus strategy are more commitment-oriented than those in

firms that do not pursue this type of strategy [control = focus ]

8 We hypothesize the influence of a focus strategy (instead of a flexibility strategy) on the shaping of HRM as we argue thatthese two tend to go hand in hand. Firms that produce tailor-made (i.e., individualized) products and services and,accordingly, are said to pursue a focus strategy, need flexibility in their organizational structure, enabling them to instantlyreact to whimsical consumer needs. Focus here refers to differentiation focus where special needs of buyers in certainsegments are served (Porter, 1985).


19/43

19

In addition to the focus of business strategy, the extent to which a firm pursues growth may influence

the shaping of HRM practices. It is argued that the pursuit of a growth strategy is related to more

formal and professionally developed HRM practices (Thakur, 1999; Matthews and Scott, 1995). As a

formal structure is expected to be characteristic of a control-oriented HRM system (see Table 1), the

following hypothesis is formulated:

H7: HRM systems in firms pursuing a growth strategy are less commitment-oriented than those in

firms that do not pursue such a strategy [control = growth ]

Firm age or stage of development

Several scholars have argued that as firms move through various stages of development, differing

problems must be addressed, resulting in the need for different management skills, priorities, and

structural configurations (Greiner, 1972; Churchill and Lewis, 1983; Kazanjian, 1988; Kimberly and

Miles, 1980; Smith, Mitchell and Summer, 1985). Accordingly, firm age may be of influence on the

shaping of HRM systems. Commitment in HRM systems may be more important in the first stages of

business development when the business is still small and struggles to stay alive. When the business

grows, more employees need to be recruited and management control becomes more important. Since

younger firms tend to be in an earlier stage of development than older firms, firm age may be of

influence on the shaping of HRM systems. The following hypothesis is formulated to capture the effect

of firm age on the commitment-orientation of the HRM system:

H8: HRM systems in younger firms are more commitment-oriented than those in older firms

[control = firmage ]

Time invested in the business


20/43

20

Although there is no literature to support the relationship between time invested in the business and the

employment relationship, it may be argued that whether someone works full-time or part-time in the

business influences the shaping of HRM practices. Commitment may be more important in businesses

where the entrepreneur, or owner-manager, is not always present to control the production process.

Time invested in the business may be related to the degree to which employees work independently on

their jobs. In addition, and related to the previous arguments, the degree to which team building is

important may be dependent upon the course of the work week of the entrepreneur.

H9: HRM systems in firms where the entrepreneur invests many hours are less commitment-

oriented than those in firms where the entrepreneur invests few hours [control = hours ]

4 Methodology

4.1 Data Collection and Sample Characteristics

To test the hypotheses, use is made of a panel of EIM Business and Policy Research in Zoetermeer.

Every four months approximately 2,000 Dutch entrepreneurs participate in this so-called SME Policy

Panel. The SME Policy Panel is used for both cross-sectional and longitudinal research and registers

two types of data. The first type concerns basic information about both the business and its owner, i.e.,

the entrepreneur. These data are renewed every year because of their short-term character. This basic

information is collected using a questionnaire consisting of a fixed set of questions. The second type of

information relates to more specific information regarding performance, attitudes and behaviors of the

Dutch SMEs. This information is collected through telephone interviews. Three times a year panel

members are approached to participate in the interviews where two or three specific (often policy-

related) topics are discussed.


21/43

21

For the present study use is made of a selection of questions from the SME Policy Panel concerning

both human resource management issues and other information concerning the business and the

entrepreneur. Independent variables include gender of the entrepreneur (i.e., the owner-director or

director of the business), (logarithm of) firm size, firm age, hours invested in the business, whether a

business is a service business and business strategy (low-price, quality, focus and growth). See Table 2

for a description of the independent variables used in this study. The independent variables (i.e., gender

and business profile) are selected from earlier panel rounds than the dependent variable(s) (i.e.,

human resource management) to ensure an adequate direction of the relationship between human

resource management and the explanatory variables in the empirical analysis.

Table 2: Description of Independent Variables

Variable Description Measurement N Mean Std. dev.Gender Is the entrepreneur female or male? Dummy variable: female = 1 and male = 0. 3431 0.12 0.33Firmsize Logaritm of number of people

employed in the firmMin. value = 0 and max. value = 7.87. 2359 2.65 1.47

Firmage Number of years the firm has been inexistence

Response categories: 1=0-2 years, 2=3-5 years,3=6-10 years, 4= more than 10 years.

2404 3.45 0.87

Hours Number of hours per week investedin the business

Response categories: 1=1-20 hours, 2=21-40 hours,3=41-60 hours, 4= more than 60 hours.

1491 3.11 0.64

Service Is the firm located in the servicesector? Dummy variable: services = 1 and non-services =0. 2063 0.44 0.50

Lowprice To what extent adopts the business alow-price strategy?

Response categories: 1=none, 2= limited extent,3=some extent, 4=large extent, 5=very large extent.

2135 2.63 1.15

Quality To what extent adopts the business ahigh-quality strategy?

Response categories: 1=none, 2=limited extent,3=some extent, 4=large extent, 5=very large extent.

2256 4.30 0.83

Focus To what extent adopts the business a(differentiation) focus strategy?


2151 3.67 1.17

Growth To what extent adopts the business agrowth strategy?


2368 2.26 0.70

The average firm size in the sample, as measured by the number of people employed, is 34.73 9. The

number of people employed includes the owner(s), manager(s), working family members, fulltime and

part-time employees as well as helpers or assistants. More than 60 percent of the firms in the analysis

has less than 25 employees, 35 percent has between 26 and 100 employees, and 4 percent has more


22/43

22

than 100 employees. Since we expect that the effect of increasing size on human resource management

diminishes, we represent firm size by the logarithm of the number of employees, rather than taking into

account the absolute number of employees (see Table 2). Six firms have no employees; these firms are

excluded from the analysis.

4.2 Measurement of HRM

Of the 15 dimensions of the Control-Commitment Continuum presented in Table 1, eight dimensions,

covered by the available data, are selected in the empirical study 10. Commitment dimensions are

constructed using multiple items from the SME Policy Panel questionnaire. Factor analysis is used to

determine which items belong to which dimension. The results of the factor analysis are presented in

Table 3.

The first factor appears to consist of items pertaining to the dimensions of informal structure and

learning . For the purpose of the present study, both dimensions are included separately in the analysis.

Factor two clearly shows the decentralization dimension. Because the fourth item employees control

their own work does not contribute to any of the other factors and it can be viewed as a form of

decentralization, it is included in the decentralization dimension. Factors three and four clearly show

the general training and broadly defined jobs dimensions, respectively. Factor five is made up of

employee participation items. Because the third item employees are involved in decision-making does

not contribute to any of the other factors it is included in the employee participation dimension. Factors

six and seven show the indirect supervision and task differentiation dimension. Eigenvalues range from

2.72 of the first factor to 1.30 for the seventh (last) factor. The factors in Table 3 correspond with the

dimensions in Table 1 surprisingly well. Hence, it can be argued that the dimensions as identified by

9 The standard deviation is 77.02.10 Initially, nine commitment dimensions were selected. However, in the factor analysis the dimension of the importance of collective, i.e., team-based, action, dropped out. That is why we decided to proceed with eight variables.


23/43

23

Beer et al. (1984), Arthur (1992; 1994) and Boselie (2002), are relevant and valuable for studying

HRM issues.

In addition to the factor analysis, Cronbachs Alpha is computed to verify whether, given a certain

dimension, the selected items form a coherent group. Cronbachs Alpha amounted to 0.53 for the

participation items, 0.66 for the decentralization items, 0.67 for the indirect supervision items, 0.69 for

the informal structure items, 0.45 for the broadly defined job items, 0.31 for the task differentiation

items, 0.58 for the learning items and 0.72 for the general training items. Hence, in three of the eight

cases Cronbachs Alpha is relatively small ( < 0.6). In general, it seems that the coherence between

the items is sufficient but limited. Including separate items, instead of constructed dimensions, in the

analysis did not yield any additional information.

Following the dimensions in Table 3, eight commitment variables are constructed as an unweighted

average of the underlying items. These commitment variables are presented in Table 4 and include

employee participation ( PARTICIP ), decentralization ( DECENTR ), indirect supervision ( INDIRECT ),

informal structure (INFORMAL), broadly defined jobs (BROADJOB), task differentation

(TASKDIFF ), explicit attention paid to learning ( LEARN ) and general training ( TRAINGEN ). Also, a

general commitment variable ( COMMITM ) is constructed as an unweighted average of the eight

specific commitment variables.


24/43

24

Table 3: Factor Analysis Matrix (Principal Component Analysis, Varimax Rotated)

FactorsDimensions and items 1 2 3 4 5 6 7

Participation

1: Employees involved in recruitment/selection 0.20 0.812: Employees involved in employee assessment 0.863: Employees are involved in decision-making 0.43 0.31 0.20 0.26 -0.16

Decentralization 1: Employees determine their own decisions a 0.82 0.142: Employees make their own decisions a 0.84 0.133: Employees determine their work pace 0.68 0.204: Employees control their own work -0.12 0.36 -0.37 -0.20 0.34

Indirect supervision 1. Employees work independently 0.18 0.822: Employees fulfill their tasks without directsupervision

0.29 0.77

Informal structure 1: There are no written rules/procedures -0.58 -0.18 0.13 0.112: Consultation does not occur via fixed rules -0.57 -0.17 0.35 0.153: Jobs/tasks (contents) are not written down -0.71 0.26 0.15

Broadly defined jobs1: Employees each do not have specific tasks 0.532: Order of tasks is not determined in advance 0.28 0.60 0.143: Outcomes are not specified in advance -0.34 0.56 0.224: Employees jobs are interchangeable 0.55 -0.15Task differentiation1: Work is diverse 0.12 0.14 0.59

2: Employees have multiple tasks 0.76 Learning 1: Employees are provided with feedback 0.52 0.19 -0.11 0.322: Explicit attention for employee learning 0.59 0.13 0.173: Number of employees with training 0.64 0.17 0.28General training 1: Management training 0.30 0.64 0.192: Social and individual development training 0.18 0.853. Team building training 0.83 -0.11Eigenvalue (factor) 2.72 2.33 1.96 1.67 1.65 1.52 1.30

N=833

Note 1: all underlying items are questions with three response categories: 1 = to a limited extent, 2 = to some extent, 3= to a large extent.

Note 2: only factor loadings 0.1 are presented. Factor loadings 0.5 are highlighted in bold. Items with factor loadings in bold are included in the construction of the commitment variables.a The distinction between these two items is not entire clear. The first item may refer to decision-making at a higher hierarchical level where employees do not only make their own decisions, but also determine what kind of decisionsthey can make themselves. The inclusion of both items in the analysis is justified by their similar factor loadings.


25/43

25

Table 4: Description of Commitment Variables

Variable Description Measurement a N Mean Std. dev.PARTICIP Degree to which employees can

influence strategic decision-making,surpassing their immediate tasks

Unweighted average of three items,three response categories.

1483 1.74 0.48

DECENTR Degree to which employees are ableto fulfill their tasks autonomously

Unweighted average of four items,three response categories.

1064 2.28 0.53

INDIRECT Degree to which supervision isindirectly structured

Unweighted average of two items,three response categories.

2097 2.60 0.58

INFORMAL Degree to which the business isinformally structured


1087 1.79 0.63

BROADJOB Degree to which jobs are broadlydefined

Unweighted average of four items,three response categories.

1472 1.82 0.46

TASKDIFF Degree to which tasks aredifferentiated

Unweighted average of two items,three response categories.

1479 2.48 0.46

LEARN Degree to which explicit attention is paid to the learning of employees


943 2.50 0.48

TRAINGEN Degree to which training is gerenal Unweighted average of four items,three response categories.

1475 2.02 0.59

COMMITM Degree to which HRM systems arecommitment-oriented

Unweighted average of the eightspecific commitment HRMvariables.

833 2.16 0.21

a Response categories are the following: 1 = to a limited extent, 2 = to some extent, 3 = to a large extent. See Table 3 for details on construction of the commitment variables.

In an additional test, two types of commitment are included: type A commitment (COMMITA) i.e.,

employee involvement and competence development constructed as an unweighted average of the

commitment variables: employee participation (PARTICIP), decentralization (DECENTR), attention

paid to learning (LEARN) and general training (TRAINGEN), and type B commitment (COMMITB)

i.e., formalization of procedures and work structure constructed as an unweigthed average of the

commitment variables: indirect supervision (INDIRECT), informal structure (INFORMAL), broadly

defined jobs (BROADJOB) and task differentiation (TASKDIFF). The rationale for distinguishing

between these two commitment types is dealt with in the results section.


26/43

26

5 Results

In this section the relationship between the different independent and dependent variables is

investigated using Pearson Product-Moment correlation coefficients. Subsequently, the relationships

between the business profile characteristics, gender and the commitment variables (in Table 4) are

investigated through a series of regression analyses on the specific commitment variables and the

general commitment variable.

5.1 Correlation Analysis

Table 5 presents the Pearson correlations between all variables in the sample. Reviewing the

correlations between the independent variables, it is striking to see that gender correlates with all

except one of the other independent variables. From a bilateral perspective, women seem to have

smaller and younger businesses, invest less time in the business, have a service rather than a non-

service business and are less likely to pursue low price, focus and growth strategies (although these

correlations are relatively small). Most of these gender differences are in line with the literature. The

finding that gender is correlated with the other independent variables indicates the likelihood of gender

(indirectly ) influencing the extent to which HRM practices are commitment-oriented through the

business profile (see Figure 1). The high correlation of firm size with firm age ( r =0.35, p


27/43

27

Table 5: Pearson Correlation between All Variables in the Sample

1 2 3 4 5 6 7 8 9 10 11 12 13

. gender 1

. firmsize -0.26 *** 1

. firmage -0.12 *** 0.35 *** 1

. hours -0.16 *** -0.05 -0.08 *** 1

. service 0.14 *** -0.17 *** -0.07 *** -0.07 ** 1

. lowprice -0.05 ** 0.11 *** -0.15 -0.001 -0.05 ** 1

. focus -0.07 *** 0.10 *** 0.02 0.04 0.003 0.08 *** 1

. quality -0.02 0.06 *** 0.02 0.03 -0.02 0.10 *** 0.36 *** 1

. growth -0.08 *** 0.19 *** -0.10 *** 0.04 -0.01 0.11 *** 0.13 *** 0.10 *** 10. PARTICIP -0.09 *** 0.42 *** 0.10 *** -0.14 *** 0.04 0.01 0.07 ** 0.03 0.11 *** 11 . DECENTR -0.06* -0.02 -0.02 -0.09 *** 0.13 *** -0.05 0.13 *** 0.05 0.04 0.12 *** 12. INDIRECT -0.01 -0.10 *** -0.01 -0.02 0.09 *** -0.04 0.06 *** 0.04 ** -0.003 -0.04 0.32 *** 1

3. INFORMAL 0.01 -0.43 *** -0.06 ** 0.02 0.02 -0.05 -0.07 ** -0.08 ** -0.14 *** -0.23 *** -0.01 0.05 1

4. BROADJOB 0.04 -0.18 *** -0.07 ** 0.02 0.004 -0.05 -0.02 -0.07 ** -0.04 -0.09 *** 0.03 0.11 *** 0.3

5. TASKDIFF -0.01 -0.13 *** -0.09 *** 0.04 0.08 ** -0.02 0.06 0.03 0.02 -0.03 0.12 *** 0.10 *** 0.0

6. LEARN -0.03 0.45 *** 0.04 -0.15 *** 0.14 *** 0.02 0.08 ** 0.01 0.14 *** 0.42 *** 0.16 *** 0.02 -0.43

7. TRAINGEN -0.09 *** 0.26 *** 0.03 0.02 0.06 ** 0.06 ** 0.03 0.01 0.16 *** 0.17 *** 0.04 -0.02 -0.318. COMMITM -0.09 *** 0.02 -0.05 -0.09 *** 0.15 *** -0.04 0.11 *** 0.01 0.09 *** 0.35 *** 0.58 *** 0.53 *** 0.22

* Coefficient is significant at the 0.10-level (2-tailed); ** Coefficient is significant at the 0.05-level (2-tailed); *** Coefficient is significant at the 0.01-level (2-t


28/43

28

Investigating the correlations between the commitment variables in Table 5, we see that, in general,

there is a moderate degree of correlation between the specific commitment variables. Most strongly

associated commitment variables include the relationships between informal structure

( INFORMAL ) and attention paid to learning ( LEARN ) (r =-0.43, p


29/43

29

Table 6: Regression Explaining Dimensions on the Control-Commitment Continuum (-values are represented

COMMITMENTVARIABLES

constant gender firmsize firmage hours service lowprice quality focus gro

PARTICIP 1.16*** -0.12** 0.16*** 0.01 -0.09*** 0.06** -0.01 0.02 0.03** 0.07DECENTR 2.37*** -0.24*** -0.0002 -0.02 -0.08** 0.14*** -0.04* 0.02 0.04* 0.02INDIRECT 2.66*** -0.19** -0.08*** 0.03 -0.02 0.05 -0.009 0.03 0.006 0.00INFORMAL 2.69*** -0.12 -0.26*** 0.03 0.02 -0.01 -0.009 0.0004 -0.01 -0.0BROADJOB 2.25*** 0.008 -0.07*** -0.02 -0.02 -0.01 -0.01 -0.002 0.01 -0.0TASKDIFF 2.50*** -0.03 -0.04** -0.02 0.01 0.07** -0.003 0.005 0.03 0.02LEARN 1.94*** 0.03 0.22*** -0.04 -0.07** 0.16*** -0.009 -0.02 0.04** 0.07TRAINGEN 1.16*** -0.05 0.13*** -0.0007 0.04 0.11*** 0.02 -0.002 0.004 0.09

COMMITM 2.15*** -0.08** 0.007 -0.009 -0.03** 0.06*** -0.008 -0.0003 0.02** 0.03

COMMITA 1.80*** -0.08* 0.13*** -0.02 -0.06*** 0.10*** -0.007 -0.02 0.03*** 0.07COMMITB 2.51*** -0.09* -0.11*** 0.004 0.0006 0.02 -0.01 0.02 0.004 -0.0*Coefficient is significant at the 0.10-level (2-tailed); ** Coefficient is significant at the 0.05-level (2-tailed); *** Coefficient is significant at the 0.0


30/43

30

Gender and HRM

Regarding gender effects on the specific commitment variables, we see in Table 6 that six of the

eight gender effects are negative, of which three are significantly negative. None are significantly

positive. On the whole, the effect of gender on the commitment-orientation of the HRM system is

negative. This can be seen also from the negative effect of gender on the general commitment

variable ( COMMITM ). Hence, hypothesis H1 is rejected: contrary to what is generally believed we

find that female entrepreneurs are less likely to make use of commitment in structuring their HRM

systems. Of the hypotheses on the HRM dimensions, operationalized by the specific commitment

variables, H1.1 through H1.3 are rejected. As compared to male-led businesses, in female-led

businesses there is a lower degree of employee participation, a higher degree of centralization and

more direct supervision of employees. In addition, no evidence is found for hypotheses H1.4

through H1.8. Female- and male-led businesses do not differ regarding formalization of the

organizational structure, job scope, task assignment, the attention paid to learning and general

training.

Business profile and HRM

In addition to gender, several of the business profile factors (see Figure 1) appear to influence the

commitment-orientation of HRM practices, including firm size ( firmsize ), time invested in the

business ( hours ), whether a business a service business ( service ), focus and growth strategies ( focus

and growth ).

Firm size ( firmsize ) has different effects on the commitment-orientation of HRM. Of the eight size

effects on the specific commitment variables, four are significantly negative and three are

significantly positive. As compared to larger firms, smaller businesses are characterized by a higher

degree of direct supervision of employees, a lower degree of formalization, more broadly defined

jobs and more task differentiation. These contrary effects produce an overall size effect on the


31/43

31

commitment-orientation of the HRM system ( COMMITM ) that is not significant. Hypothesis H2 is

not supported.

Service businesses ( service ) have a higher score on the commitment-orientation of the HRM

system. Hypothesis H3 is supported. With respect to the specific commitment variables, we see that

a service (rather than a non-service) business is characterized by higher degrees of employment

participation, decentralization, task differentiation, and more attention for learning and general

training.

Adopting a focus strategy ( focus ) has a positive impact on the degree to which the HRM system is

commitment-oriented. Hypothesis H6 is supported. Regarding the HRM dimensions, businesses

pursuing a focus strategy have a higher degree of employee participation and decentralization as

well as more explicit attention for learning. In addition, adopting a growth strategy ( growth ) also

positively influences the commitment-orientation of the HRM system. This was not foreseen.

Hypothesis H7 is rejected. Of the four significant growth effects on the specific commitment

variables, three are positive and one is negative. Businesses that pursue growth are characterized by

a high degree of employee participation, explicit attention for learning and general training and a

relative formal organizational structure.

In general, time invested in the business ( hours ) appears to negatively affect the commitment-

orientation of the HRM system. More time invested in the business entails a HRM system that is

less commitment-oriented. Hypothesis H9 is supported. Regarding the HRM dimensions, it can be

said that the more time invested in the business leads to less employee participation, more

decentralization and less attention paid to learning.

The development stage of a firm ( firmage ) as well as the pursuit of a low-price ( lowprice ) or quality

strategy ( quality ) do not influence the commitment-orientation of the HRM system in a systematic


32/43

32

fashion, although pursuing a low-price strategy appears to be accompanied by a higher degree of

centralization. Hypotheses H4, H5 and H8 are not supported.

To summarize, gender and time invested in the business negatively influence the commitment-

orientation of the HRM system, whereas a business in the service sector and pursuing a focus or

growth strategy positively influences the degree of commitment. When split up in effects on

specific commitment variables, firm size does influence commitment in HRM. However, because of

the contrary effects, the overall size-effect on the commitment-orientation of the HRM system

(COMMITM ) is cancelled out.

Because gender is significantly correlated with time invested in the business, service business and

pursuing a focus and growth strategy (see Table 5), and these factors in turn influence the

degree of commitment of the HRM system, it would be expected that, in addition to the direct

gender effect, gender influences the degree of commitment-orientation indirectly through the

business profile (see Figure 1). However, Table A-1 in the Appendix 11 shows that leaving out

either gender or the controls does not produce any disturbing effects 12. Hence, there is no evidence

of an indirect gender effect. Only for indirect supervision ( INDIRECT ) and general training

(TRAINGEN ) the gender effect is not similar when comparing regression results including all

variables and gender only. When controlled for the business profile factors in the first row, the

gender effect appears for indirect supervision ( INDIRECT ) and dissapears for general training

(TRAINGEN ).

Two types of commitment

Observing the specific commitment variables in Table 6 a distinction can be made between two

groups of commitment practices: those that focus upon employee involvement and competence

11 Table A-3 in the Appendix presents the results of regression analyses per commitment dimension, including allexplanatory variables in the first row, gender only in the second row and the business profile factors only in the thirdrow.


33/43

33

development (i.e., PARTICIP , DECENTR , LEARN , TRAINGEN ) and those emphasizing work

structure and procedures at different organizational levels (i.e., INDIRECT , INFORMAL ,

BROADJOB , TASKDIFF )13 . Additional regression analyses are performed to test for gender effects

on these two different groups of commitment practices, referred to as type A and B commitment

practices ( COMMITA and COMMITB ), respectively. Type A commitment is comprised of the

commitment dimensions: employee participation ( PARTICIP ), decentralization ( DECENTR ),

attention paid to learning ( LEARN ) and general training ( TRAINGEN ). Type B commitment is made

up of the commitment dimensions: indirect supervision ( INDIRECT ), informal structure

( INFORMAL ), broadly defined jobs ( BROADJOB ) and task differentiation ( TASKDIFF )14.

For both types of commitment, we find a similar negative gender effect (at a 10% significance

level). This could be an indication of the control-orientation of women regarding human resource

management practices in general. Firm size has a different influence on the two commitment types,

positively influencing type A commitment and negatively influencing type B commitment. Larger

businesses have higher employee involvement and pay more attention to competence development.

Smaller firms have a broader job design and are relatively informally structured. The theoretical

underpinnings of the two scale effects, found in this study, deserves further attention. This is outside

the scope of the present paper.

As opposed to type B commitment, type A commitment is also influenced by other factors than

gender and firm size. The more hours invested in the business (hours), the lower commitment type

A. Hence, businesses where the entrepreneur is full-time involved are less likely to be characterized

by employee involvement and pay less attention to competence development. In addition,

businesses in the service sector and with a focus or growth strategy are characterized by a higher

commitment type A, i.e., more employee involvement and competence development.

12 Taking into account only those respondent cases for which all relevant data are available (i.e., where the N is equalfor the three regressions on the different commitment items), does not produce different results.13 These findings are data-driven.


34/43

34

6 Conclusion and Discussion

Making use of several HRM dimensions on the Commitment-Control Continuum (see Table 1), as

first proposed by Beer et al. (1984), the present study finds that gender of the entrepreneur

influences the degree to which HRM practices are commitment-oriented. Gender influences both

the commitment-orientation of several of the HRM dimensions (i.e., employee participation,

decentralization and supervision), as well as that of the aggregate HRM measures (i.e., HRM

system and commitment types A and B: employee involvement and competence development, and

work structure and procedures, respectively).

The effect of gender on the commitment-orientation of the HRM system is a direct effect, rather

than an indirect one, the latter working through business profile factors, such as firm size, sector,

business strategy, firm development stage and time invested in the business (see Figure 1). This

means that when the business profiles of female- and male-led businesses are similar, there still

remains a gender difference regarding the commitment-orientation of the HRM system. Although

women in the present study are more likely to have businesses in the service sector, invest less

hours in their business, are less likely to pursue a focus and a growth strategy, and these factors in

turn have a positive influence on the degree to which the HRM system are commitment-oriented,

the overall effect of gender remains negative.

Contrary to what is generally believed, the present study shows that HRM systems in businesses led

by women are more control-oriented and less commitment-oriented than those in male-led

businesses. Previous research has shown that women are more likely to let employees participate in

decision-making (e.g., Cromie and Birley, 1991; Stanford et al., 1995; Neider, 1987), focus upon

relationships instead of hierarchy (e.g., Buttner, 2001; Brush, 1992; Stanford et al., 1995) and have

a more indirect way of managing employees (e.g., Verheul et al. 2002). In this study, female-led

businesses are characterized by less employee participation, more centralization and direct instead

14 Type A and B commitment are computed as the mean of the four underlying commitment variables.


35/43

35

of indirect ways of controlling employees, as compared to male-led businesses. The counterintuitive

finding that women are more control-oriented than men may be related to gender differences in risk

taking propensity (e.g., Verheul and Thurik, 2001; Van Uxem and Bais, 1996). If women are less

willing to take risk than men, this may to some extent explain why they are less willing to involve

others in the decision-making process as relying upon others means giving up control. Practicing

direct control over others reduces uncertainty.

The finding that women are more control-oriented corresponds with the findings of Mukthar (2002,

p. 305/6), arguing that female owner-managers are more autocratic, less consultative, less willing

to allow employees make independent decisions and more reluctant to delegate authority to others.

This is also identified in a study by Piercy et al. (2001), showing that female sales managers use

higher levels of behavior control when they manage teams 15. Nevertheless, the results of the present

study should be interpreted with caution. For instance, there may be intermediating factors that are

not controlled for in the present study and that are associated with gender. For example, women

may be involved in specific types of businesses. Contingency control theory argues that

organizational structuring and type of control within a firm is dependent upon factors, such as type

of technology (e.g., routine versus non-routine) involved, firm size as well as environmental

uncertainty 16. Although the present study controls for firm size, it may be that the gender effect can

be ascribed to the fact that women are often less likely to be involved in high-tech businesses, and

with sectors with unstable environments, whereas these in turn may positively influence the

commitment-orientation in the organizational structure. A business in an uncertain environment

should maintain a flexible organizational structure to adequately adapt to changing market

circumstances. This flexibility is more likely to be feasible when a business focuses on commitment

in the structuring of HRM practices than when the focus is on control. To shed more light upon the

15 Sales management control strategy in this study has been defined, following Anderson and Oliver (1987), as theextent to which sales managers perform several monitoring, directing, evaluating and rewarding activities in carryingout their management responsibilities (see Piercy et al., 2001, p. 39/40).16 See Daft (1998, p. 354), largely based on Woodwards (1965) technological complexity scale.


36/43

36

actual, i.e., effective, gender effect on the structuring of HRM practices within firms, further

research should be conducted exploring these mediating effects of environmental and technological

complexity.

Further research should also focus on the influence of the other factors, such as firm size, on the

commitment-orientation of HRM. Interestingly, we did not find a firm size effect on the degree to

which the HRM system is commitment-oriented. The overall effect of firm size on the commitment-

orientation of the HRM system is cancelled out by reverse effects on the HRM dimensions. When

distinguishing between type A and type B commitment, we see that firm size positively effects type

A commitment (i.e., employee involvement and competence development) and negatively effects

type B commitment (i.e., formalization of procedures and work structure). Hence, the effect of firm

size should be further investigated, making a distinction between different types of HRM practices.

In the present study different HRM practices are added up to construct the aggregate measure of

HRM system . However, as noted in the theoretical section, in most firms HRM practices do not

form a coherent system. This is confirmed by the relatively low, and in some cases even negative,

correlations among the specific HRM variables. Hence, researchers should be made aware that the

use of aggregate measures of HRM practices may lead to misinterpretation of findings.

The present study is based upon the views of Beer et al. (1984), Walton (1985) and Arthur (1992,

1994), implicitly assuming that control and commitment are two sides of a single dimension.

However, it is important to investigate whether, indeed, commitment and control are two extremes

on one continuum (Boselie, 2002, p. 41). The study by Piercy et al. (2001) concludes that, next to

displaying a higher level of behavioral control, female sales managers also create more

organizational commitment in their teams. This may be an indication that control and commitment

can go hand in hand rather than be exclusive. Moreover, a distinction should be made between

different types of control and/or commitment. In the present study a distinction is made between

commitment types A and B, yielding different results. These commitment types are data driven and


37/43

37

could not logically be named or labeled, as both are comprised of fairly different components.

Although several scholars have proposed different types of control (e.g., Merchant, 1985; Harzing,

1999; Snell, 1992; Burton, 2001), further research is needed to investigate commitment types. In

addition, human resource managent systems may be classified according to different lines.

Although the distinction between a focus on control and commitment is a comprehensible one, it is

likely that in practice more diverse employment models can be identified. For instance, Burton

(2001) distinguishes between five employment models based on the structuring of three human

resource dimensions: attachment, coordination/control and selection.

The present study uses a sample of Dutch female and male entrepreneurs. Because it can be

expected that gender differences in leadership or management styles (Osland et al., 1998) or

leadership styles in general (Gibson, 1995) differ internationally, the results may not be generally

applicable. In addition, Hofstede (2001) finds that, as compared to other countries, the Netherlands

are characterized by a relatively low degree of masculinity. The relative feminine culture in the

Netherlands is likely to affect the extent to which women and men differ with respect to

management of their employees. Because in a feminine culture male behavior is likely to be more

similar to, than different from, female behavior, this may be an indication of the importance of the

gender differences found in the present study.


38/43

38

ReferencesAllen, N.J. and J.P. Meyer, 1990, The measurement and antecedents of affective, continuance and normative

commitment to the organization, Journal of Occupational Psychology 63 (1), 1-18.Anderson, E. and R.L. Oliver, 1987, Perspectives on behavior-based versus outcome-based salesforce

control systems, Journal of Marketing 51 (4), 76-88.Arthur, J.B., 1992, The link between business strategy and industrial relations systems in American steel

minimills, Industrial and Labor Relations Review 45 (3), 488-506.Arthur, J.B., 1994, Effects of human resource systems on manufacturing performance and turnover, Academy

of Management Journal 37 (3), 670-687.Arthur, M. and C. Hendry, 1990, Human resource management and the emergent strategy of small to

medium sized business units, International Journal of Human Resource Management 1 (3), 233-250.Audretsch, D.B. and A.R. Thurik, 2001, What is new about the new economy: sources of growth in the

managed and entrepreneurial economies, Industrial and Corporate Change 10 (1), 267-315.Audretsch, D.B. and A.R. Thurik, 2000, Capitalism and democracy in the 21st century: from the managed to

the entrepreneurial economy, Journal of Evolutionary Economics 10 (1), 17-34.Bales, R.F., 1950, Interaction Process Analysis: A Method for the Study of Small Groups, Reading, MA:

Addison-Wesley.Bartlett, K.R., 2001, The relationship between training and organizational commitment: a study in the health

care field, Human Resource Development Quarterly 12 (4), 335-352.Baruch, Y., 1998, The rise and fall of organizational commitment, Human Systems Management 17, 135-143.

Bass, B.M., 1985, Leadership and Performance Beyond Expectations, New York: Free Press.Bass, B.M., B.J. Avolio and L.E. Atwater, 1996, The transformational and transactional leadership of men

and women, Applied Psychology 45, 5-34.Beer, M., B. Spector, P.R. Lawrence, D.Q. Mills and R.E. Walton, 1984, Managing Human Assets , New

York: Free Press.Belenky, M.F., B.M. Clinchy, N.R. Goldberger and J.M. Tarule, 1986, Womens Way of Knowing , New

York: Basic Books.Blake, R.R. and J.S. Mouton, 1964, The Managerial Grid , Houston: Gulf Publishing Company.Boselie, P., 2002, Human Resource Management, Work Systems and Performance , Dissertation, Erasmus

University Rotterdam: Tinbergen Institute Research Series and Thela Thesis.

Brush, C.G., 1992, Research on women business owners: past trends, a new perspective and futuredirections, Entrepreneurship Theory and Practice 16 (4), 5-30.Burton, M.D., 2001, The company they keep: founders models for organizing new firms, in: Schoonhoven,

Claudia Bird and E. Romanelli (eds.), The Entrepreneurship Dynamic: Origins of Entrepreneurship and the Evolution of Industries , Stanford, CA: Stanford University Press.

Buttner, E.H., 2001, Examining female entrepreneurs management style: an application of a relationalframe, Journal of Business Ethics 29, 253-269.

Buttner, E.H. and D.P. Moore, 1997, Womens organizational exodus to entrepreneurship: self-reportedmotivations and correlates with success, Journal of Small Business Management 35 (1), 34-46.

Carter, S., 1993, Female business ownership. In: Allen, S. and C. Truman, (eds.), Women in Business. Perspectives on Women Entrepreneurs , London/New York: Routledge, 148-160.

Carter, S. and P. Rosa, 1998, The financing of male- and female-owned businesses, Entrepreneurship and Regional Development 10, 225-241.

Chaganti, R., 1986, Management in women-owned enterprises, Journal of Small Business Management 24(4), 18-29.

Chaganti, R. and S. Parasuraman, 1996, A study of the impacts of gender on business performance andmanagement patterns in small businesses, Entrepreneurship Theory and Practice Winter 1996, 73-75.

Churchill, N.C. and V.L. Lewis, 1983, The five stages of small business growth, Harvard Business Review May-June, 30-50.

Cromie, S. and S. Birley, 1991, Networking by female business owners in Northern Ireland, Journal of Business Venturing 7 (3), 237-251.

Cuba, R., Decenzo, D. and A. Anish, 1983, Management practices of successful female business owners, American Journal of Small Business 8, (2), p. 40-45.


39/43

39

Curran, J., Kitching, J., Abbott, B. and V. Mills, 1993, Employment and Employment Relations in the SmallService Sector Enterprise, Small Business Research Centre, Kingston Business School.

Daft, R.L., 1998, Organization: Theory and Design , Cincinnati Ohio: International Thomson Publishing.De Kok, J.M.P. and L.M. Uhlaner, 2001, Organization context and human resource management, Small

Business Economics 17, 273-291.De Kok, J.M.P., Uhlaner, L.M. and A.R. Thurik, 2002, Human resource management in small and medium-

sized firms, Strategic Study, Zoetermeer: EIM Business and Policy Research.Deshpande, S.P. and D.Y. Golhar, 1994, HRM practices in large and small manufacturing firms: A

comparative study, Journal of Small Business Management 32 (2), 49-56.Duberley, J.P. and P. Walley, 1995, Assessing the adoption of HRM by small and medium-sized

manufacturing organizations, International Journal of Human Resource Management 6 (4), 891-909.Eagly, A.H. and B.T. Johnson, 1990, Gender and leadership style: a meta-analysis, Psychological Bulletin

108, 233-256.Eisenhardt, K.M., 1985, Control: organizational and economic approaches, Management Science 31, 134-

149.Ely, R.J., 1994, The effects of organisational demographics and social identity on relationships among

professional women, Administrative Science Quarterly 39 (2), 203-238.Engen, M.L. van, 2001, Gender and Leadership: a Contextual Perspective , Tilburg University: Dissertation

Social and Behavioral Sciences.Fischer, M.L. and H. Gleijm, 1992, The gender gap in management: a challenging affair, Industrial and

Commercial Training 24 (4), 5-11.Friedman, A.L., 1977, Industry and Labour: Class Struggle at Work and Monopoly Capitalism, London:

MacMillan Press.Gibson, C.B., 1995, An investigation of gender differences in leadership across four countries, Journal of

International Business Studies 26 (2), 255-279.Gilligan, C., 1982, In a Different Voice: Psychological Theory and Womens Development, Cambridge,

MA: Harvard University Press.Godard, J., 1998, Workplace reforms, managerial objectives and managerial outcomes: the perceptions of

Canadian IR/HRM managers, International Journal of Human Resource Management 9 (1), 18-40.Goffee, R. and R. Scase, 1995, Corporate Realities: The Dynamics of Large and Small Organisations ,

London/Boston: International Thomson Business Press.Golhar, D.Y. and S.P. Deshpande, 1997, HRM practices of large and small Canadian manufacturing firms,

Journal of Small Business Management 35 (3), 30-38.Goss, D., 1991, Small Business and Society , London: Routledge.Grant, J., 1988, Women as managers: what can they offer to organizations?, Organizational Dynamics 16

(3), 56-63.Greiner, L.E., 1972, Evolution and revolution as organizations grow, Harvard Business Review July-August

50, 37-46.Guest, D.E., 1987, Human resource management and industrial relations, Journal of Management Studies 5

(24), 503-521.Guest, D.E., 1997, Human resource management: a review and research agenda, International Journal of

Human Resource Management 8 (3), 263-276.Gundry, L.K and H.P. Welsch, 2001, The ambitious entrepreneur: high growth strategies of women-owned

enterprises, Journal of Business Venturing 16, 453-470.Guthrie, J.P., C.S. Spell and R.O. Nyamori, 2002, Correlates and consequences of high involvement work

practices: the role of competitive strategy, International Journal of Human Resource Management 13 (1),183-197.Hall, R., 1993, A framework linking intangible resources and capabilities to sustainable competitive

advantage, Strategic Management Journal 14, 607-618.Harzing, A-W, K., 1999, Managing the Multinationals: an International Study of Control Mechanisms,

Cheltenham: Elgar.Helgesen, S., 1990, The Female Advantage: Womens Way of Leadership, New York: Doubleday.Heskett, J.L., W.E. Sasser and L.A. Schlesinger, 1997, The Service Profit Chain: How Leading Companies

Link Profit and Growth to Loyalty, Satisfaction and Value, New York: Free Press.


40/43

40

Hisrich, R.D and C.G. Brush, 1987, Women entrepreneurs: a longitudinal study. In: N.C. Churchill, J.A.Hornaday, B.A. Kirchhoff, O.J. Krasner and K.H. Vesper (eds.), Frontiers of Entrepreneurship Research ,Wellesley, M.A.: Babson College, 187-199.

Hofstede, G., 2001, Cultures Consequences: Comparing Values, Behaviors, Institutions, and OrganizationsAcross Nations, Thousand Oaks, CA: Sage.

Hornsby, J.S. and D.F. Kuratko, 1990, Human Resource Management in small business: Critical issues for the 1990's, Journal of Small Business Management 28 (3), 9-18.

Huselid, M.A., 1995, The impact of human resource management practices on turnover, productivity, andcorporate financial performance, Academy of Management Journal 38 (3), 635-672.

Huselid, M.A., S.E. Jackson and R.S. Schuler, 1997, Technical and strategic human resource managementeffectiveness as determinants of firm perfromance, Academy of Management Journal 4 (1), 171-188.

Ichniowski, C., C. Shaw and G. Prennushi, 1997, The effects of human resource management practices on productivity: a study of steel finishing lines, American Economic Review 87 (3), 291-313.

Jackson, S.E., Schuler, R.S. and J.C. Rivero, 1989, Organizational characteristics as predictors of personnel practices, Personnel Psychology 42, 727-786.

Jago, A.G. and V.H. Vroom, 1982, Sex differences in incidence and evaluation of participative leader behavior, Journal of Applied Psychology 67 (6), 776-783.

Kabacoff, R.I., 1998, Gender differences in organizational leadership: a large sample study, ManagementResearch Group, Portland, ME.

Kalleberg, A.L. and K.T. Leicht, 1991, Gender and organizational performance: determinants of small business survival and success, Academy of Management Journal 34 (1), 136-161.

Kazanjian, R.K., 1988, Relation of dominant problems to stages of growth in technology-based newventures, Academy of Management Journal 31 (2), 257-279.

Kimberly, J.R. and R.H. Miles, 1980, The Organizational Life Cycle , San Francisco: Jossey-Bass.Klein, H.J., 2001, Invited reaction: the relationship between training and organizational commitment a

study in the health care field, Human Resource Development Quarterly 12 (4

Documents

Commitment or Control