Gender diversity advantage at middle management:Implications for high performance work systemimprovement and organizational performance

và thông tin bổ ích giúp sinh viên tham khảo, ôn luyện và phục vụ nhu cầu học tập của mình cụ thể là có định hướng, ôn tập, nắm vững kiến thức môn học và làm bài tốt trong những bài kiểm tra, bài tiểu luận, bài tập kết thúc học phần, từ đó học tập tốt và có kết quả cao cũng như có thể vận dụng tốt những kiến thức mình đã học.

6 3 lượt tải Tải xuống

Chia sẻ Báo cáo tài liệu

O R I G I N A L A R T I C L E

Gender diversity advantage at middle management:

Implications for high performance work system

improvement and organizational performance

Min-Kyu Joo

| Jeong-Yeon Lee

| Dejun Tony Kong

| Phillip M. Jolly

Discipline of Organisational Studies,

University of Sydney Business School,

University of Sydney, Sydney, Australia

Department of Management, School of

Business, Seoul National University, Seoul,

South Korea

Division of Organizational Leadership and

Information Analytics, Leeds School of

Business, University of Colorado, Boulder,

Colorado, USA

Hospitality Management College of Health

and Human Development, Pennsylvania State

University, University Park, Pennsylvania, USA

Correspondence

Min-Kyu Joo, Discipline of Organisational

Studies, University of Sydney Business School,

University of Sydney, Sydney, Australia.

Email: minkyu.joo@sydney.edu.au

Abstract

Research on women in leadership positions has largely focused on the board or top

management team (TMT) leadership level, showing that increased female representa-

tion at these levels can benefit organizational performance. However, the strategic

implications of female representation at middle management have been largely

neglected. The current study addresses this issue in relation to High Performance

Work System (HPWS) improvement and organizational performance (profitability).

By analyzing the multi-wave (2009, 2011, and 2013) Workplace Panel Survey (WPS)

data collected from 1101 organizations in South Korea, we found a gender diversity

advantage at middle management; that is, a higher level of gender diversity of middle

management translated into a higher level of organizational performance due to

HPWS improvement. However, this advantage appeared only in the presence of a

high level of subordinates' gender diversity. Our findings have important implications

for gender diversity and strategic human resource management.

K E Y W O R D S

gender diversity, high performance work system, leadership, middle management, organizational

performance

Companies in the top quartile for executive teams' gender diversity

are 21% more likely to financially outperform their counterparts in the

bottom diversity quartile (Hunt et al., 2018 2015). Post and Byron ( )

found that female board representation is positively related to

(a) organizational accounting-based performance in all countries and

(b) organizational market performance in countries with greater gen-

der parity (i.e., more equal access to resources and opportunities for

women), presumably because female board representation benefits

monitoring and strategy involvement. Recent research has found that

when women join top management teams, organizations may shift

from a buying to a building strategic approach (Post et al., 2022).

These are but a few recent studies that have investigated the effects

of female executive leadership. Given that women are severely under-

represented at the top of the corporate ladder—only 6% of C-suite

executives (chairman, president, CEO, COO, etc.) are women (Eagly &

Carli, 2007)—it is not surprising that research has largely focused on

the strategic implications of female representation at the top leader-

ship level (e.g., Leslie et al., 2017 2016; Tinsley et al., ).

Yet compared to what we know about organizational outcomes

and female representation in top leadership (board or TMT), our knowl-

edge of female representation at middle management, and its implica-

tions for organizational performance, is rather limited. This gap is

glaring, given the importance of middle management to the develop-

ment and implementation of organizational strategy. Middle managers,

including but not limited to general line managers, functional line man-

agers, and team or project-based executives (Wooldridge et al., 2008),

are “coordinator(s) between daily activities of the units and the strate-

gic activities of the hierarchy” (Floyd & Wooldridge, 1994, p. 48), and

are closer to daily operations but are often defined by their “lack (of)

the formal role authority held by their seniors to act strategically”

Jeong-Yeon Lee and Dejun Tony Kong are equal second authors.

DOI: 10.1002/hrm.22159

1099050x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.22159 by RMIT UNIVERSITY LIBRARY, Wiley Online Library on [05/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

(Rouleau & Balogun, 2011, p. 954). Indeed, Floyd and Wooldridge

(1997) noted that middle managers play crucial strategic roles when

they “mediate, negotiate, and interpret connections between the orga-

nization's institutional (strategic) and technical (operational) levels”

(p. 466). Moreover, middle managers also play key roles in developing

and motivating employees (Kehoe & Han, 2020).

Therefore, middle managers can have a positive influence on

the strategic operations and positioning of their organization

(e.g., improved decision making, strategy/system development, and

implementation), thereby improving organizational performance

(Floyd & Wooldridge, 1992 1994 1997 1990, , ; Wooldridge & Floyd, ).

Though still underrepresented at middle management levels, women

have made greater gains at this level than in the upper echelons of

organizations (Einarsdottir et al., 2018). Thus, it is important to shed

greater light on the impacts that female representation at lower levels

of management may have on organizations. We introduce the term

gender diversity advantage at middle management to indicate the

potential enhancement of organizational performance from increased

gender diversity of middle management.

Specifically, we argue that

increased gender diversity at middle management can enhance orga-

nizational performance through strengthened HPWS implementation.

However, simply increasing gender diversity of middle management is

not beneficial for organizational performance; its advantage can be

realized only when subordinates are gender diverse.

In the current study, drawing from research suggesting that com-

prehensive HR systems, known as high performance work systems

(HPWS), are positively associated with organizational performance

(Jiang et al., 2012), we argue that gender diversity of middle manage-

ment is indirectly and positively related to organizational performance

via the improvement or development of HPWS. HPWS improvement

refers to the strengthening of existing HPWS through the implemen-

tation of additional high-value HR practices, which results in positive

changes in the functions of HPWS and subsequent improvement in

organizational capabilities essential for performance in dynamic envi-

ronments (Jiang et al., 2012 2017; Roberson et al., ).

Specifically, we argue that an increased female representation at

middle management can facilitate the generation of information and

knowledge around employee needs and desires, leading to more dif-

ferentiated solutions (Wiersema & Bantel, 1992). Given middle man-

agers' important role in managing strategic and operational activities,

gender diversity of middle management has the potential to facilitate

organizational performance indirectly through HPWS improvement,

contingent upon nonmanagerial employees' gender diversity. HPWS

is a comprehensive HR system intended to enhance organizational

performance and is arguably among the most effective means that

middle managers can use to achieve strategic and operational goals.

By considering (a) HPWS improvement as an intermediary mechanism

and (b) nonmanagerial employees' gender diversity as a boundary con-

dition, we help delineate how gender diversity of middle management

can indirectly enhance organizational performance. We term this key

benefit a “gender diversity advantage at middle management.”

We integrate research on gender diversity and strategic human

resources (HR) for our theorizing in the country of South Korea. We

expect a gender diversity advantage at middle management to be

strong in this collectivistic culture characterized by significant gender

inequality. In addition, the potential confounding effect of racial diver-

sity would not be of great concern in our study using archival data

from South Korea, given the country's racial homogeneity, allowing us

to focus more on gender effects.

Our study makes several contributions to the gender diversity

and strategic HRM literatures. First, the current study sheds critical

light on research on gender diversity. Previous research has focused

on how gender diversity at the top or bottom levels of the organiza-

tion influences organizational performance (e.g., Carter et al., 2010;

Gonzalez & DeNisi, 2009). Although middle managers, who handle

the day-to-day operations of their organization, play central roles in

enhancing organizational performance, we have very little knowledge

about how and when gender diversity of middle management can

enhance organizational performance. Our study provides some evi-

dence regarding the critical role that gender diversity of middle man-

agement plays in enhancing organizational performance via HPWS

implementation. Second, very little work has looked at the anteced-

ents of HPWS systems or influences on their adoption and implemen-

tation, and those that have focused largely on external or top-down

drivers (Kim et al., 2021). Our work provides insights into the impor-

tant ways in which the demographic characteristics of middle manage-

ment, namely gender diversity, can impact the possible development

and implementation of HR policies and practices that can contribute

to an overall HPWS and organizational performance, contributing to

the gender diversity literature. Third, we attempt to clarify how and

when gender diversity of middle management can be a competitive

advantage for organizations and provide a new mechanism through

which a gender diversity advantage at middle management for HPWS

improvement and organizational performance is realized. Lastly, this

study contributes significantly to the understanding of context in stra-

tegic HRM research (i.e., South Korea) by exploring this phenomenon

using a large sample of South Korean panel data and identifying a con-

textual factor (subordinate gender diversity) that may serve as a

boundary condition of a gender diversity advantage at middle

management.

1 | THEO RE TI CA L FO UN DA TI ON S A ND

HY PO TH ES ES

1.1 | Strategic role of middle management

Strategy is often considered to be solely the domain of top manage-

ment. However, from the perspective of strategy as processual

(Whittington, 1993), stakeholders outside senior management, such

as those in middle management, play a critical role in realizing strate-

gic change because “strategy is a mix of deliberate intent and emer-

gent actions and decisions, as a result of which formulation and

implementation of strategy are intertwined rather than being separate

processes” (Currie & Procter, 2001, p. 55). Therefore, middle man-

agers are “not just passive recipients, but also active interpreters,

2 JOO .ET AL

mediators, and intermediaries in implementing and reforming strategy

and policy” (Xie et al., 2013, p. 3481). In other words, middle managers

are involved in organizing, directing, synthesizing, and controlling imple-

mentation of strategic plans and thus have a substantial influence on

strategic development processes (Floyd & Wooldridge, 2000; Huy,

2001; Kehoe & Han, 2020; Rouleau & Balogun, 2011; Wooldridge

et al., 2008).

We focus specifically on the key linking role that middle managers

may play in connecting HR strategies and employees (Currie &

Procter, 2001). HR strategies are considered to be “third order' strate-

gies…flowing from, but also upwardly influencing…wider corporate

strategies” (Purcell, 2000, p. 62). Middle managers can play a signifi-

cant role in HR strategy (Kehoe & Han, 2020), not only in terms of

downward implementation, but also through active identification of

beneficial practices and policies and the promotion of these policies

to organizational decision makers. In identifying, implementing (on a

local level), and providing resources to test emergent ideas, middle

managers engage in championing new ideas and alternatives, which is

one of the primary pathways for middle managers' strategic involve-

ment (Floyd & Wooldridge, 1992).

The main purpose of championing new ideas and alternatives is

to adjust and improve the current content of strategy to benefit orga-

nizational functioning (Floyd & Lane, 2000 2005 2008; Mantere, , ).

From an HR perspective, middle managers can serve as organizational

champions, persistently and persuasively advocating for entrepreneur-

ial or innovative options, initiatives, and strategies to senior manage-

ment and HR management. They are implementers of organizational

strategy and observers of its effectiveness (Raes et al., 2011), and act“

as an initial screen” (Floyd & Wooldridge, 1994, p. 50) of HPWS

implementation in day-to-day operations as well as subordinates'

reactions to HPWS. If HPWS implementation is ineffective or subordi-

nates have adverse reactions to HPWS, middle managers can select

from a wide range of business opportunities, novel procedural pro-

posals, and administrative changes suggested at operating levels

(Floyd & Wooldridge, 1994) and provide this information to strategic

decision makers (Raes et al., 2011). Because middle managers “(live) in

the organizational space between strategy and operations,” they “are

uniquely qualified to make such judgments” (Floyd & Wooldridge, 1994,

p. 50) and decide to which issues senior managers and HR managers

should pay attention (Wooldridge et al., 2008).

As alluded to above, middle managers play two primary strategic

roles: (1) collection of information regarding both the effectiveness of

the implemented strategy and potential suggestions and alternatives

to existing strategy; and (2) identification, evaluation, and presentation

of new ideas and alternatives. Accordingly, we argue that two mana-

gerial capabilities should be particularly important for successful

middle managers' effectiveness of influencing strategic HR decisions.

First, creating an environment where information regarding employee

needs flows freely up to middle managers from subordinates is instru-

mental to information gathering (Kehoe & Han, 2020), because such

an environment will make subordinates more willing to provide

insights and ideas on what employees believe would be valuable

from an HR practice perspective. Second, considering that middle

managers' impact on HR strategy is a function of their ability to influ-

ence decision makers and exchange information with them in a pro-

ductive way (Raes et al., 2011), middle managers must be able to

persuasively present and advocate for recommendations based on

collected information. Given the extant evidence that female middle

managers are more likely to engage in participative/democratic lead-

ership and collaborative negotiation, we propose that the presence of

higher gender diversity of middle management is likely to lead to

greater improvement in HPWS.

1.2 | Gender diversity of middle management and

HPWS improvement

In order to effectively champion new ideas and alternatives regarding

HPWS, middle managers should pay attention to the needs of internal

human capital and provide suggestions for systems that will appropri-

ately address these needs. Middle managers should also be proficient

at selling HPWS-related issues to top executives to obtain buy-in and

necessary financial resources (Dutton et al., 1997; Dutton &

Ashford, 1993). Finally, they also must have an ability to encourage

suggestions from employees and a strong willingness to implement

HPWS, helping their organization meet its employees' needs. In light

of these assertions, we identify a gender diversity advantage at middle

management in that such diversity contributes to HPWS improve-

ment; this is because of women's participative/democratic leadership

in the process of gathering feedback and suggestions from subordi-

nates and their collaborative negotiation in the process of solving

problems with senior managers and HR managers. Said differently,

gender-diverse middle management (i.e., an increased proportion of

female middle managers) should lead subordinates to speak up about

issues (Eagly & Johnson, 1990) and should also possess a greater abil-

ity to integrate diverse sets of information in ways that might lead to

innovative HR policy and practice solutions and then advocate for

those solutions (Eagly et al., 2003 1990; Eagly & Johnson, ). Our argu-

ments to this effect are similar to those arguing for the benefits of

gender diversity in new venture teams in the entrepreneurship litera-

ture (see Dai et al., 2019).

First, well-documented differences in information processing

between males and females should lead to greater collection and integra-

tion of suggestions from subordinates by middle management that is

more gender diverse. Women are likely to pay more attention to subtle

information (Darley & Smith, 1995), are less likely to adopt problem-

solving strategies that conform to existing beliefs or heuristics (Chung &

Monroe, 1998 1998), retain more detailed information (Seidlitz & Ed, ),

and have better recall after encountering new information (Erngrund

et al., 1996) than men. All of these findings support women's tendency

to manage subordinates in a more participatory and democratic manner,

which promotes employee voice and input (Book, 2000; Greene

et al., 2003), thus providing the information necessary to understand the

needs of employees with respect to HR policies and practices.

Second, women tend to engage in greater levels of relational

information processing than men, meaning that they focus more on

JOO ET AL. 3

the relationships between sets of individual pieces of information

(Chung & Monroe, 1998), such as diverse suggestions from multiple

subordinates as to HR policy preferences and needs. Such information

processing increases the likelihood that gender-diverse middle manage-

ment will be able to craft innovative and norm-challenging HR policies

and practices. However, this is only the first step to HPWS improve-

ment. Middle managers must also persuade their senior management

to adopt new HR practices in order to accommodate the needs of

employees (Ashford & Detert, 2015; Mirabeau & Maguire, 2014). Nota-

bly, despite its potential benefits at the operational level, improving

HPWS is not cost free to organizations, and therefore, finding a solu-

tion that benefits all parties is important for the success of such negoti-

ation (cf. Raes et al., 2011); it requires collaboration and problem

solving among all negotiation parties (Kong et al., 2014; Walton &

McKersie, 1965).

Middle management's gender diversity can facilitate exploration

of creative ideas and collaboration with senior managers and HR man-

agers, as women have been found to focus more on problem solving,

conflict resolution, and finding a mutually beneficial solution during

negotiation (Brahnam et al., 2005 2013; Henderson et al., ; Walters

et al., 1998). Taken together, these findings support the argument that

gender-diverse middle management can help senior managers and HR

managers formulate better HPWS strategy by directing attention

toward strategic issues in regard to HPWS improvement and advocat-

ing for resources for HPWS improvement. Hence, gender diversity of

middle management can be a crucial driver of successful HPWS

improvement (Gilbert et al., 2015).

In sum, since more gender-diverse middle management tends to

focus on development, learning, individual/communal goals, and

voices (Paustian-Underdahl et al., 2014), with lateral and upward sup-

port, they likely better serve as champions who gather information

from subordinates about what issues require new ideas and alterna-

tives that help improve the existing HPWS, and negotiate collabora-

tively with senior managers and HR managers, thereby facilitating

HPWS improvement. In addition, HPWS can provide a context where

female middle managers find a good fit with their cooperative man-

agement style and soft skills (Kato & Kodama, 2017). Therefore, we

predict a gender diversity advantage at middle management, such that

as more women are represented there, the greater HPWS improve-

ment will be. As such, we propose the following hypothesis:

Hypothesis H1. Gender diversity of middle management

in an organization is positively related to the level of HPWS

improvement.

1.3 | Gender, management positions, and HPWS

in South Korea

Gender differences may play more critical roles in implementing HR

practices in male-dominant cultures (Ciancetta & Roch, 2021; Hofstede

et al., 2010; Williams & O'Reilly, 1998). South Korea is a relatively

unfriendly place for working women in terms of the glass-ceiling index,

which includes various measures such as pay, childcare costs, and per-

centage of senior positions (Economist, 2018). In South Korea, female

employees' contributions to their organization have been less valued or

appreciated than those of their male counterparts, which results in gen-

der inequality in promotion to top management (Cooper-Jackson, 2001)

and a lack of women in upper management positions (Easton, 2015).

Several critical factors constitute barriers to women's career advance-

ment in South Korea, such as Confucianism, traditional norms, and gen-

der stereotypes and roles (e.g., about married women with children)

(Ibarra et al., 2013). Because women are burdened by both work and

family demands and responsibilities, they feel the need to put extra

effort toward standing out in the workplace via their work performance

(Kim et al., 2020) while resolving the conflict between the roles of

breadwinners and homemakers/caregivers (Cho et al., 2019). In addi-

tion, many women in South Korea have limited networking and leader-

ship development opportunities, which make them struggle in rising to

management positions (Cho et al., 2017; Rowley et al., 2016).

We focus on HPWS as an important mediator between gender

diversity of middle management and organizational performance in

South Korea because Bae and Lawler (2000) pointed out four reasons

why HPWS could be important to implement in a collectivistic coun-

try such as South Korea. First of all, South Korea's collectivism facili-

tates cooperation, loyalty, and harmony, which are aligned well with

HPWS (Lee & Johnson, 1998). Second, recent Korean culture can be

characterized as a composite of Asian and Western values. That is,

Korean organizational culture shows that individualism and group har-

mony are both equally strong. After the financial crisis in 1997, many

organizations operated their organizations with dynamic collectivism,

reflecting multidimensional and paradoxical subcomponents: in-group

harmony, optimistic progressivism, and the hierarchical principle

(Cho & Yoon, 2001). Dynamic collectivism is defined as a mixture of

harmony and change, face-saving and aggressiveness, and emotional

community and impersonal achievement (Cho et al., 2014).

The third reason is globalization (Bae et al., 2007). In order to

achieve better performance under globalization, organizations should

be more flexible to adapt to uncertainty in their environment. After

the 1997 financial crisis, the Korean Chaebols—large conglomerates—

westernized their HRM systems (e.g., from seniority- to performance-

based compensation) and increased employment flexibility (e.g., from

high job security to more part-time or contingent job positions). Chae-

bols started emphasizing teamwork, employee empowerment, voice,

and merit-based hiring, incorporating many characteristics of HPWS

(Bae et al., 2007). Many South Korean organizations emphasize inno-

vation, market culture, and risk-taking and implement HPWS (Rhee

et al., 2016). Relative to Western culture, collectivistic organizations

are more likely to encourage people to collaborate to better coordi-

nate and integrate organizations' skills and resources (Lee &

Miller, 1999). Knowledge exploitation and exploration through knowl-

edge transfer within the organization can be accomplished by such

collaboration, which is aligned well with the core function of HPWS

(Miller & Shamsie, 1996).

Finally, because HPWS involves continuous training, it fits well

with the fact that South Korean people value education and

4 JOO .ET AL

development as a means to be competitive (Bae & Lawler, 2000). The

above reasons explain the transformation of South Korean cultures,

leading to more acceptance and willingness to work under HPWS

(e.g., selective staffing, comprehensive training, outcome-based com-

pensation, employee empowerment, participation in a decision making

process, etc.), and pointing to the need to understand the increasing

role of HPWS in South Korean organizations.

1.4 | Subordinate gender diversity as a critical

contingency

As previously discussed, national culture plays important role in imple-

menting HR practices and perceiving leadership (Schuler, 2013;

Stoeberl et al., 1998). We also argue that the national context of our

data collection (South Korea) may influence the degree to which inter-

nal context (subordinate gender diversity) can shape the relationship

between gender diversity of middle management and HPWS improve-

ment. Employees who are in individualistic countries may have an

independent view, while employees who are in collectivistic countries

such as South Korea, Japan, and China emphasize an interdependent

view stressing connectedness and relationships (e.g., Guanxi; Chen

et al., 2013) as well as high vertical collectivism—high collectivism and

high power distance (Singelis et al., 1995). For example, organizations in

Japan still maintained traditional East Asian HR system (e.g., seniority-

based compensation and promotion, long-term employment; Bae

et al., 2010). Traditional HR management in East Asian countries relied

more on informal personal relationships and people's judgment than on

official and objective criteria and regulations (Lam et al., 2002; Xian

et al., 2019). Confucian values and the Chinese family business model

based on traditional authority are pervasive in South Korea as well

(Claessens et al., 2000).

Traditional South Korean HR management has prevented

employees from becoming involved in decision making processes as

information was passed down from senior management (Deyo, 1989;

Frenkel & Lee, 2010) and emphasized employee loyalty and long

working hours. In these contexts, individuals perceived typical male

leadership styles as more appropriate than female leadership

(Triandis, 1995); a majority of organizations did not have any women

on their board in those counties (Low et al., 2015; The Japan

Times, 2021 2020; The Korea Herald, ). Despite high levels of technol-

ogy and economic development, traditional social and cultural norms

are still alive, influencing beliefs about gender roles in those countries

(Brooks, 2006).

While employees in countries where individualism is highly valued

seek opportunities and incentives for themselves, collectivism would

lead employees to reorient their goals toward collective benefits.

According to Schuler (2013), in countries where collectivism is strong,

such as East Asian countries, HR practices that facilitate information

sharing may be less prevalent. High power distance also prevents the

drive to seek personal benefits and would be well aligned with auto-

cratic leadership (Dorfman et al., 2012). In high power distance cultures,

employees may agree with the strategic choices of management

because they are more likely to have and accept unequal power distri-

butions within social institutions and for organizational representatives

to invoke legitimate authority (Hofstede, 1984). Managers can flaunt

authority more and enjoy greater hierarchical inequality in the higher

power distance context, which thereby enhances the possibility that

managers' strategic choices concerning low levels of HPWS will be

accepted.

Set against this context, we argue that the effect of gender diver-

sity of middle management on HPWS improvement is bounded by

gender diversity of subordinates for two reasons. First, while gender-

diverse middle management tends to have more engaging leadership

and collaborative negotiation skills than gender-homogenous middle

management, the former faces challenges due to gender role incon-

gruity (i.e., a mismatch between their gender and middle manager role)

in South Korea. Whether female middle managers can effectively play

the champion role that facilitates HPWS improvement depends on

the internal environment; one important aspect of the internal envi-

ronment is organizational demography. We argue that gender diver-

sity of subordinates is a demography-related boundary condition.

South Korea ranked dead last in terms of gender egalitarianism in

House et al.'s (2004) study; the culture values a masculine leadership

style, and thus, female middle managers in South Korea likely face

prejudice and discrimination from their male subordinates (cf. Eagly &

Carli, 2007 2002; Eagly & Karau, ).

For example, female middle managers, who are not common in

South Korea, may be viewed by male subordinates as less competent

or effective and thus be less accepted by male subordinates. In a

male-dominant internal environment, female middle managers may be

compelled to adjust their leadership behaviors to more dominant and

assertive ones to match their male subordinates' preferences for mas-

culine leadership styles. By doing so, gender-diverse middle manage-

ment loses its advantage in participative/democratic leadership and

collaborative negotiation. As a result, the benefit of gender diversity

of middle management for HPWS improvement is less likely to hap-

pen. By contrast, in an internal environment characterized by a higher

level of subordinate gender diversity, female middle managers are not

as threatened by negative gender stereotypes or role incongruity

(Paustian-Underdahl et al., 2014). Therefore, gender diversity of mid-

dle management can translate into more widespread participative/

democratic leadership and collaborative negotiation, facilitating a fer-

tile environment for HPWS improvement.

Second, the new ideas and alternatives that gender-diverse mid-

dle management champions for HPWS improvement can come from

subordinates. Female and male subordinates have different life and

work experiences. Due to such differences, they likely have different

views on how the existing HPWS is working and how it should work.

Gender diversity, despite its potential for engendering conflict, can be

a pivotal source of novel and useful ideas (Egan, 2005; Jackson

et al., 2003). Stated differently, gender diversity is a source of creative

ideas and alternatives in that male and female employees have a

greater range of viewpoints and lived experiences (Mateos de Cabo

et al., 2012) from which to draw to create suggestions for changes

and/or improvements in HR policies and practices. With these ideas

JOO ET AL. 5

and alternatives, gender-diverse middle management is in a better

position to collect information used to understand the issues and

potential solutions based on HPWS improvement.

Taken together, the above arguments lead us to hypothesize that

gender diversity of subordinates strengthens the link between gender

diversity of middle management and the level of HPWS improvement.

Hypothesis H2. Subordinate gender diversity modifies

the relationship between gender diversity of middle man-

agement in an organization and the level of HPWS

improvement, such that this relationship is stronger when

subordinate gender diversity is higher.

1.5 | HPWS improvement and organizational

performance

HPWS, as a unique, inimitable, causally ambiguous, and synergetic

mechanism, can deliver organizations a sustainable competitive

advantage (Lado & Wilson, 1994 1994). Wright et al. ( ) argued that

individual HR practices are not a competitive advantage as HR

practices can easily be copied by competitors. However, it is difficult

to imitate bundles of HR practices due to the interdependence of

multiple practices and their synergistic effect (Lado & Wilson, 1994).

Characteristics of HPWS bundles, such as unique historical paths, can

make them difficult for competitors to copy (Becker & Huselid, 2006).

Indeed, Saridakis et al. (2017) found that HPWS, as a bundle of multi-

ple HR practices, facilitated organizational performance to a greater

extent than individual HR practices, and this effect appeared for both

financial performance (e.g., accounting returns) and operational

performance (e.g., workforce productivity). The positive implication of

HPWS for organizational performance has been demonstrated using

various indicators of organizational performance (e.g., Huselid, 1995;

Jiang et al., 2012 2011 2017; Messersmith et al., ; Saridakis et al., ) in

various countries, such as China, Japan, South Korea, Spain, Canada,

the U.K., and the U.S. (Choi, 2014 1995; Huselid, ; Prieto &

Santana, 2012 2014 2007; Rabl et al., ; Sun et al., ; Takeuchi

et al., 2007).

Following previous research (e.g., Jiang et al., 2012; Lepak

et al., 2006), we argue that HPWS has ability-, motivation-, and

opportunity-enhancing (i.e., AMO-enhancing) functions, satisfying

employees' psychological needs for competence, relatedness, and

autonomy (Ryan & Deci, 2000), and thus can help improve organiza-

tional performance. HPWS refers to “organizational actions or pro-

cesses and job characteristics that focus on attracting, developing,

and motivating employees and providing opportunities to contribute”

(Boon et al., 2019 2006, p. 2518). Lepak et al. ( ) conceptualized

systems of HR practices as falling into three distinct dimensions—skill-,

motivation-, and opportunity-enhancing HR practices. Stated

differently, HPWS provides three components to foster employee

performance: ability, motivation, and opportunity (Jiang et al., 2012;

Lepak et al., 2006; Messersmith et al., 2011). Practices such as stan-

dardized and well-developed recruitment practices, rigorous selection,

and comprehensive training opportunities are intended to help

employees acquire and develop their work-related knowledge, skills,

and competencies. Practices such as developmental feedback, competi-

tive compensation, performance-based incentives, high-quality bene-

fits, clear promotion and career development opportunities, and job

security are designed to enhance employees' work motivation. Finally,

practices such as team-based work, employee involvement systems,

flexible job designs, and pathways for information sharing provide

employees with the opportunity to use skills and motivation in ways

that achieve organizational objectives (Boon et al., 2019).

As HPWS is improved, employees will have stronger ability,

stronger motivation, and more opportunities to use their ability and

motivation for their work (Huselid, 1995 2000; Ramsay et al., ). Thus,

employees can more efficaciously participate in collaborative decision

making processes and work with others in delivering high-quality

output, thereby increasing their work efficiency (Boon et al., 2019)

and ultimately facilitating organizational performance.

Hypothesis H3. The level of HPWS improvement is posi-

tively related to the level of organizational performance.

Autonomous strategic behaviors are defined as activities that

create substantial internal variations by diverging from or substituting

for existing strategic plans (Burgelman, 1991 2018; Mirabeau et al., ).

For reasons explained above, we argue that gender diversity of

middle management may create significant content-based variance

(i.e., implemented HR practices) between organizations by actively

facilitating autonomous strategic behaviors such as championing new

ideas and initiatives related to programs and practices that constitute

HPWS (Noda & Bower, 1996). As a result, gender diversity of middle

management represents a potential source of heterogeneity among

organizations, which is the primary driver of “organizations' ability to

derive rents from complex HR systems such as HPWS” (Chadwick &

Flinchbaugh, 2021, p. 35), and thus creates a gender diversity advan-

tage at middle management. HPWS is arguably among the most effec-

tive means for middle managers' achievement of strategic goals, as

HPWS fosters human capital and employee motivation, which are two

important drivers of organizational performance (Jiang et al., 2012;

Messersmith et al., 2011) (see Figure 1 for our conceptual model).

Hypothesis H4. The level of HPWS improvement medi-

ates the interactive effect of gender diversity of middle

management and subordinate gender diversity in an orga-

nization on the level of organizational performance.

2 | ME TH OD

2.1 Sample|

To test our hypotheses, we used the Workplace Panel Survey (WPS)

data

collected in 2009, 2011, and 2013. This survey was conducted

6 JOO .ET AL

by the Korea Labor Institute, a government-funded policy research

body, in tandem with the South Korean Ministry of Labor. Approxi-

mately 1700 firms with 30 or more employees across the country

were randomly selected using stratified sampling and surveyed.

Organization-level data in 2009 and 2011 were available for 1737

and 1770 organizations, respectively, while organization-level data in

2013 were available for 1621 organizations.

Individual interviews (Blaise system-based Computer-Assisted

Personal Interviewing) were conducted to obtain information related

to human resources and labor relations. Mail and web surveys were

conducted to obtain data on employment status and financial status.

Organizations in the forestry, mining, fishing, and agriculture sectors,

as well as public service organizations, were excluded from the sam-

pling frame. The organization-level survey was completed by the HR

manager and a labor union representative of each organization. The

data of the endogeneity control variables were collected in 2007

(Time 0), those of the predictor and control variables in 2009 (Time 1);

those of the mediator (i.e., HPWS improvement) in 2011 (Time 2), and

those of organizational performance (profitability) in 2013 (Time 3).

We matched the organization-level panel data across the 3 years

using unique identification numbers. The resulting final sample com-

prised 1101 organizations that provided suitable and usable data for

analysis. After we matched identified numbers in each year, we used

1066 organizations to test Hypotheses H1 and H2. 621 organizations

were used to test Hypotheses H3 and H4. The response rates for

original samples were 70.3% (2009), 62.4% (2011), and 54.9% (2013),

respectively.

2.2 | Measures of key variables

Gender diversity of middle management (proportion of female middle

managers) (Time 1). Female middle managers included female team-

or project-based executives, deputy general managers, and depart-

ment managers. We aggregated them to derive the total number of

female middle managers in an organization. Then, we calculated gen-

der diversity of middle management in each organization by dividing

the total number of female middle managers by the total number of

middle managers (Kalev, 2009).

Gender diversity of subordinates (Time 1). Subordinates'gender

diversity in each organization was indicated by the Blau's index

(1977). We calculated this index by subtracting the sum of the respec-

tive squared proportions of male and female employees from 1; the

score ranged from 0 (most homogeneous) to 0.5 (most diverse)

(Gonzalez & Denisi, 2009 2004; Richard et al., ).

B I

: : ¼ 1 

 

where, P = the proportion of subordinates in the organization (male

or female); i = the number of different categories represented in the

organization ( 2).=

HPWS improvement. We included HR practices to measure

HPWS in the current study, which are highly overlapped with those in

previous studies. According to the recent comprehensive review con-

ducted by Boon et al. (2019), HR practices most commonly used to

build HPWS are categorized into the following: training, participation/

autonomy, incentives, performance appraisal, recruiting, and selection,

which are typically identified as core HR practices for HPWS

(Posthuma et al., 2013). We selected five of these six HR practice cat-

egories for our HPWS construct (Boon et al., 2019; Lepak

et al., 2006). This HPWS composition was also similarly used across

countries (e.g., the United States, China, Spain, the United Kingdom,

Canada, and South Korea) (Rabl et al., 2014).

Specifically, to create an index of HPWS in each organization, we

used twelve HR practices that encompass the types of practices iden-

tified by prior studies (cf., Huselid, 1995 2012; Jiang et al., ): training,

recruiting, a regular and formal performance appraisal system, annual

salary, profit sharing, incentives, employee suggestion program, small

group activities, regular team meetings, job rotation, multi-functional

training, and autonomy. For all practices except autonomy, HR man-

agers responded to items (1 = yes, 0 = no), indicating whether or not

their respective organizations had implemented the respective HR

practices.

For autonomy, HR managers indicated the levels of autonomy in

their organizations with respect to task, speed, recruiting, and training,

ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item

was “the extent to which a major operating unit (e.g., team) of your

workplace has autonomy regarding how to work at the unit. We”

transformed these four scales into dummy codes (1–2 = 0, 3 1)–4 =

and then aggregated all of them to create an autonomy index. We

again transformed this value into dummy codes (0–2 = 0, 3 1),–4 =

Gender Diversity

of Subordinates

(T1)

High Performance

Work System

Improvement

(T2-T1)

Organizational

Performance

(T3)

Gender Diversity

of Middle

Management

(T1)

F I G U R E 1 Conceptual model

JOO ET AL. 7

with 1 representing the effective implementation of autonomy prac-

tices and 0 representing the ineffective implementation or lack of

autonomy practices.

Ten items were summed to measure training. These items were

corporate education, external group training, internet-based education

and training, support for meetings to learn work-related skills,

employee training on the technical guidance, technical guidance, and

employee training from the parent company/trust companies, paid or

unpaid day release training, tuition support for training institutions

(including school), financial aid/loans for educational institutions, and

participation in employee training programs during working hours.

Two items were used to measure recruiting: multi-media (e.g.,

newspaper, radio, and television) recruiting and internet-based recruiting

(1 = yes, 0 = no). We used four items to measure what types of incen-

tives were in place (1) group performance-based incentive, (2) company

performance-based incentive, (3) business unit performance-based incen-

tive, and (4) department or team performance-based incentive

(1 = yes, 0 = no).

We summed each item to create indices of training (10 items),

recruiting (2 items), and incentives (4 items). It is appropriate to sum

all indicators of HR practices because these practices' synergistic

effects on organizational performance are claimed to be realized from

their complementarity in bundles (Becker & Gerhart, 1996;

MacDuffie, 1995). To obtain the improvement of HPWS from Time

1 to Time 2, we aggregated all the practices and then subtracted

HPWS at Time 1 from HPWS at Time 2 (i.e., formative measure).

Organizational performance (Time 3). We focused on organiza-

tional profitability. Following prior research, we divided the logarithm

of net profit for 2013 (fiscal year) by the organizational size in 2013

(indicated by the logarithm of the total number of full-time

employees; Shaw et al., 2013).

2.3 | Measures of control variables

We controlled for several variables theoretically predictive of organi-

zational performance. The data of these control variables were avail-

able from the organization-level survey described above.

Industry differences. Industries affect organizational performance

(Datta et al., 2005). To account for the industry differences, we used

industry dummy variables created based on WPS-provided industry

codes.

Organizational size. The larger an organization, the more effica-

cious HR practices it may have (Arthur, 1994). Thus, an organization's

size could predict its performance. We operationalized the organiza-

tional size as the logarithm of the number of full-time employees.

Organizational age. More established organizations tend to be

less vulnerable to performance pressure (Hannan & Freeman, 1984).

Thus, we controlled for the age of an organization, which was calcu-

lated as the difference between 2009 and the founding year.

Foreign stake. An organization's foreign stake can predict its per-

formance (Douma et al., 2006). To measure an organization's foreign

stake, its HR managers indicated the extent to which the shares of the

organization were held by foreigners, using a percentage ranging from

0% (no foreign stake) to 100%.

Proportion of female executives. As noted earlier, the proportion

of female strategic leaders was found to predict organizational perfor-

mance (Post & Byron, 2015). Therefore, we controlled for the propor-

tion of female executives in each organization, calculated by dividing

the number of female executives by the total number of executives.

In order to address the concern about firm-specific unobservable

characteristics, we included a variable related to management salary

and a locality variable as the proxy of corporate culture in the models.

Regarding management salary, we included the total salary of a middle

manager before tax (first-year salary). Organizations in Seoul are more

likely to recruit and retain female employees (i.e., higher levels of gen-

der equality for all positions, including top management) and have

progressive and innovative corporate culture than organizations else-

where in South Korea. It is plausible that organizations in Seoul have a

more progressive corporate culture, and therefore, organizations in

Seoul are more likely to implement HPWS (Bae & Lawler, 2000) and

promote gender diversity to a greater extent than organizations else-

where. Corporate cultures that emphasize gender equality or devalue

discrimination toward women should be more conducive to overall

gender diversity and female employees' career advancement, helping

more female employees be promoted to management positions. For

example, most Chaebols where their culture is progressive and inno-

vative are in Seoul (dummy coded, 1 = Seoul, 0 = other areas). Our

correlation table supports our arguments that locality could be a proxy

for corporate culture. Being an organization in Seoul is positively cor-

related with gender diversity of middle management ( 0.11,r =

p < 0.01), HPWS level at Time 2 (r = 0.07, p < 0.05), and gender diver-

sity of subordinates (r = 0.10, p < 0.01) as stated.

2.4 | Endogeneity control

We addressed the endogeneity of the relationship between gender

diversity of middle management (i.e., the proportion of female middle

managers) and HPWS improvement by providing the fixed effect esti-

mates on the effect of gender diversity of middle management on

HPWS improvement. We controlled for the possibility that gender

diversity of middle management is more likely to be related to HPWS

improvement. Following prior research (Tang et al., 2018), we first

regressed gender diversity of middle management at time t against a

set of antecedent variables measured 2 years prior, that is, year 2.t 

These instrumental variables included HR strategy—whether the pri-

mary goal of HR is to maximize savings on fixed labor costs and

whether HRM is based on individual performance (a five-point Likert

scale from 1 to 5). In order to account for the possibility that prior

birthrate in each organization's location might drive a higher level of

gender diversity of middle management in the areas in which each

organization operates, we included birthrate ratio (the ratio of boys

per one hundred girls) measured in 1990 (this year is the earliest one

we could obtain) by using the Korean government database (https://

kostat.go.kr/portal/eng/index.action).

In addition, we included other

8 JOO .ET AL

HR variables associated with work-life balance (i.e., selective work

hours and flexible work hours; yes = 1, no = 0) (Avery &

McKay, 2006 2017; Kato & Kodama, ) and family-friendly HR policies

(maternity leave, parental leave, workplace childcare facilities, support

for childcare cost, leaves for regular doctor visits during pregnancy,

stillbirth vacation, and infertility leave; yes = 1, no = 0). We aggre-

gated all scores related to family-friendly HR policies. We also

included favored recruitment processes for women (yes = 1, no 0).=

We then generated a predicted gender diversity of middle manage-

ment score and included it as an endogeneity control in the models.

2.5 Results|

Since all data have been conducted at an organizational level, multile-

vel analysis such as hierarchical linear modeling or generalized latent

and mixed modeling (GLLAMM) is not necessary for our study. We

believe that the PROCESS macro is a good tool to explore our

conditional mediation model (Hayes, 2013). Before analyzing the data,

we centered the components for the interaction term around their

respective means. Table 1 presents the descriptive statistics and cor-

relations. We included the control variables at Step 1. Next, we exam-

ined the main effect of gender diversity of middle management at

Step 2. At Step 3, we added the moderating variable of gender diver-

sity of subordinates and its associated two-way interaction term

(i.e., gender diversity of middle management  gender diversity of

subordinates). As shown in Table 2, at Step 1, some of the control var-

iables (i.e., management salary, organizational size, foreign stake,

HPWS at Time 1, and endogeneity control) reached (marginal) signifi-

cance, which accounted for 29% of the variance. Hypothesis H1 pre-

sents that gender diversity of middle management in an organization

is positively related to the level of HPWS improvement. At Step

2, gender diversity of middle management significantly predicted the

level of HPWS improvement (Model 2a: b = 1.95, 0.73,SD =

p < 0.01), which supported Hypothesis H1. Hypothesis H2 states

that subordinate gender diversity modifies the relationship between

TA B L E 1 Descriptive statistics and correlations

Variable M SD 1 2 3 4 5 6 7 8 9

1 Locality (proxy of

corporate culture)

0.52 0.50 ––

2 Management's salary 35.51 11.46 0.06* ––

3 Service 0.50 0.50 0.20** 0.04 ––

4 Finance 0.04 0.19 0.12** 0.20** 0.20** ––

5 Organizational age 24.92 17.47 0.06* 0.07* 0.02 0.01 ––

6 Organizational size 4.98 1.24 0.10** 0.28** 0.01 0.07* 0.27** ––

7 Foreign stake 4.21 15.99 0.01 0.09** 0.13** 0.06* 0.01 0.16** ––

8 Proportion of female

executives

0.05 0.15 0.02 0.06* 0.07* 0.03 0.02 0.11** 0.01 ––

9 GDMM (T1) 0.10 0.16 0.11** 0.02 0.31** 0.00 0.03 0.03 0.01 0.29** ––

10 HPWS (T1) 8.63 4.56 0.04 0.43** 0.02 0.11** 0.05 0.31** 0.12** 0.09** 0.03

11 HPWS (T2) 9.19 4.50 0.07* 0.32** 0.01 0.10** 0.07* 0.38** 0.19** 0.11** 0.04

12 HPWS improvement

(T2 T1)–

0.56 4.25 0.03 0.12** 0.01 0.01 0.02 0.07* 0.08* 0.01 0.08*

13 Gender diversity of

subordinates (T1)

0.31 0.16 0.10** 0.10** 0.03 0.15** 0.02 0.09** 0.00 0.07* 0.18**

14 Organizational

performance (T3)

1.63 0.55 0.11** 0.14** 0.07 0.06 0.09* 0.22** 0.10* 0.04 0.05

15 Endogeneity control 0.10 0.03 0.28** 0.11** 0.16** 0.13** 0.02 0.16** 0.13** 0.04 0.20**

Variable 10 11 12 13 14 15

10 HPWS (T1) ––

11 HPWS (T2) 0.56** ––

12 HPWS improvement (T2–T1) 0.48** 0.46** ––

13 Gender diversity of subordinates (T1) 0.14** 0.14** 0.00 ––

14 Organizational performance (T3) 0.09* 0.18** 0.10* 0.06 ––

15 Endogeneity control 0.24** 0.21** 0.03 0.16** 0.05 ––

Note: N = 1101. GDMM refers to gender diversity of middle management. Organizational performance (T3) was available from 644 organizations. The unit

of management's salary is one million won. T1 = Time 1; T2 = Time 2; and T3 = Time 3. *p < 0.05; **p < 0.01.

JOO ET AL. 9

gender diversity of middle management in an organization and the

level of HPWS improvement, such that this relationship is stronger

when subordinate gender diversity is higher. At Step 3, we found that

the interaction of gender diversity of middle management and gender

diversity of subordinates was positively related to the level of HPWS

improvement (Model 3a: b = 15.48, SD = 4.27, p < 0.01). We found

that the respective variances explained by gender diversity of middle

management, and the interaction term for HPWS improvement were

1% and 2%. A follow-up simple slope test (see Figure 2) indicated that

the relationship between gender diversity of middle management in

an organization and the level of HPWS improvement was positive

(simple slope = 4.43, t = 5.56, p < 0.01) in the presence of higher gen-

der diversity of subordinates (+1 SD), but was not significant (simple

slope = 0.52, t = 0.77, p = 0.44) in the presence of a lower level

of gender diversity of subordinates (1 SD). Accordingly,

Hypothesis H2 was supported.

Consistent with Hypothesis H3, the level of HPWS improvement

was positively related to organizational performance at Time 3 (Model

2b: b = 0.03, SD = 0.01, p < 0.01). HPWS improvement explained

4.2% of the variance of organizational performance (Table 3).

Finally, to test Hypothesis H4, which states that the level of

HPWS improvement mediates the interactive effect of gender diver-

sity of middle management and gender diversity of subordinates in an

organization on the level of organizational performance, we per-

formed 10,000-sample bootstrapping (Hayes, 2013) and used the

bias-corrected 95% confidence intervals (CI

95%

) to determine the sig-

nificance of the conditional indirect effect. We tested the conditional

indirect effect using the SPSS PROCESS macro version 26 (Model 7;

Hayes, 2013). The conditional indirect effect was only found in the

presence of a higher (b = 0.16, SD = 0.06, CI

95%

= [0.04, 0.30]) versus

lower level of subordinates' gender diversity (b = 0.07, 0.06,SD =

95%

= [0.22, 0.03]), thus supporting Hypothesis H4 (Table ).4

2.6 | Supplementary analyses

Organizations that have greater improvement in HPWS may have

more gender-diverse middle management. To address this concern,

we used a cross-lagged approach (Finkel, 1995), which is “the optimal

way to understand causality in field settings” (Lang et al., 2011,

p. 605). This approach allowed us to compare the partial regression

coefficients between variables measured at different time points. We

found that gender diversity of middle management at Time 1 was pos-

itively related to the level of HPWS improvement at Time 2 ( 0.05,b =

SD = 0.64, p < 0.05), but HPWS improvement at Time 1 was not sig-

nificantly related to gender diversity of middle management at Time

2 (b = 0.03, SD = 0.00, p = 0.28). The linkage between HPWS and

organizational performance has been criticized for ambiguity of cau-

sality (Shin & Konrad, 2014 2005; Wright et al., ). Thus, a reversed

causal relationship between organizational performance and the

TA B L E 2 Results of hierarchical regression predicting high performance work system (HPWS) improvement

HPWS improvement (T2-T1)

Predictor Model 1a b (SD) Model 2a b (SD) Model 3a )b (SD

Intercept 0.39 (0.58) 0.59 (0.58) 0.66 (0.58)

Control variables

Locality (proxy of corporate culture) 0.04 (0.23) 0.03 (0.23) 0.11 (0.23)

Management's salary 0.00 (0.00)** 0.00 (0.00) 0.00 (0.00)** **

Service 0.19 (0.23) 0.02 (0.24) 0.05 (0.24)

Finance 0.42 (0.62) 0.35 (0.62) 0.31 (0.62)

Organizational age 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 

Organizational size 0.65 (0.10)*** 0.66 (0.10) 0.70 (0.10)*** ***

Foreign stake 0.03 (0.01)*** 0.03 (0.01) 0.03 (0.01)*** ***

Proportion of female executives 1.18 (0.76) 1.74 (0.79) 1.82 (0.78) **  **

HPWS (T1) 0.55 (0.03) 0.55 (0.03) 0.56 (0.03)***  ***  ***

Endogeneity control 6.66 (3.61) 5.22 (3.64) 4.88 (3.62)*

Independent variables

Gender diversity of middle management (T1) 1.95 (0.73)*** 1.26 (0.74)*

Gender diversity of subordinates 2.50 (0.71)***

Gender diversity of middle management Gender

diversity of subordinates

15.48 (4.27)***

F-value 43.77*** 40.67 36.99*** ***

0.293 0.298 0.314

Note: N = 1,101. GDMM refers to gender diversity of middle management. T1 = Time 1; T2 = Time 2; T3 = Time 3.

*p < 0.10; **p < 0.05; ***p < 0.01.

10 JOO .ET AL

Bấm Tải xuống để xem toàn bộ.

| 1/21

| | Tải xuống

Preview text:

1099050x, 0, DOI: 10.1002/hrm.22159 D o w n loaded
O R I G I N A L A R T I C L E from https://online
Gender diversity advantage at middle management: library.wile
Implications for high performance work system y .com/doi/
improvement and organizational performance 1 0 .1002/hrm.22159 b Min-Kyu Joo1 | Jeong-Yeon Lee 2 | Dejun Tony Kong 3 | Phillip M. Jolly4 y R M IT UNIVER
1Discipline of Organisational Studies, S ITY
University of Sydney Business School, Abstract L IB
University of Sydney, Sydney, Australia R
Research on women in leadership positions has largely focused on the board or top A R 2 Y
Department of Management, School of ,
management team (TMT) leadership level, showing that increased female representa- W i
Business, Seoul National University, Seoul, ley O South Korea
tion at these levels can benefit organizational performance. However, the strategic n line 3
Division of Organizational Leadership and L
implications of female representation at middle management have been largely ibra
Information Analytics, Leeds School of ry o
Business, University of Colorado, Boulder,
neglected. The current study addresses this issue in relation to High Performance n [05 Colorado, USA /
Work System (HPWS) improvement and organizational performance (profitability). 0 3 /2 4 0
Hospitality Management College of Health 2
By analyzing the multi-wave (2009, 2011, and 2013) Workplace Panel Survey (WPS) 3 ].
and Human Development, Pennsylvania State S ee t
University, University Park, Pennsylvania, USA
data collected from 1101 organizations in South Korea, we found a gender diversity h e Ter
advantage at middle management; that is, a higher level of gender diversity of middle m s Correspondence and
Min-Kyu Joo, Discipline of Organisational
management translated into a higher level of organizational performance due to C o n d
Studies, University of Sydney Business School, it
HPWS improvement. However, this advantage appeared only in the presence of a ion
University of Sydney, Sydney, Australia. s (htt
Email: minkyu.joo@sydney.edu.au
high level of subordinates' gender diversity. Our findings have important implications p s://on
for gender diversity and strategic human resource management. linelibrary.w K E Y W O R D S iley.
gender diversity, high performance work system, leadership, middle management, organizational com/t performance erms-and-conditions)
Companies in the top quartile for executive teams' gender diversity
executives (chairman, president, CEO, COO, etc.) are women (Eagly & o n W i
are 21% more likely to financially outperform their counterparts in the Carli, 2007) le
—it is not surprising that research has largely focused on y O n
bottom diversity quartile (Hunt et al., 2018). Post and Byron (2015)
the strategic implications of female representation at the top leader- line L
found that female board representation is positively related to
ship level (e.g., Leslie et al., 2017; Tinsley et al., 2016). ibrary
(a) organizational accounting-based performance in all countries and
Yet compared to what we know about organizational outcomes for rul
(b) organizational market performance in countries with greater gen-
and female representation in top leadership (board or TMT), our knowl- es of u
der parity (i.e., more equal access to resources and opportunities for
edge of female representation at middle management, and its implica- se; OA
women), presumably because female board representation benefits
tions for organizational performance, is rather limited. This gap is artic
monitoring and strategy involvement. Recent research has found that
glaring, given the importance of middle management to the develop- les are
when women join top management teams, organizations may shift
ment and implementation of organizational strategy. Middle managers, gover
from a buying to a building strategic approach (Post et al., 2022).
including but not limited to general line managers, functional line man- n ed by
These are but a few recent studies that have investigated the effects
agers, and team or project-based executives (Wooldridge et al., 2008), the a
of female executive leadership. Given that women are severely under- are p
“coordinator(s) between daily activities of the units and the strate- p licab
represented at the top of the corporate ladder l —only 6% of C-suite
gic activities of the hierarchy” (Floyd & Wooldridge, 1994, p. 48), and e Cre
are closer to daily operations but are often defined by their “lack (of) ative C
Jeong-Yeon Lee and Dejun Tony Kong are equal second authors.
the formal role authority held by their seniors to act strategically” o m m o n s L
Hum Resour Manage. 2022;1–21.
wileyonlinelibrary.com/journal/hrm © 2022 Wiley Periodicals LLC. 1 icense 109 2 JOO 9 ET AL. 0 5 0 x , 0, Dow
(Rouleau & Balogun, 2011, p. 954). Indeed, Floyd and Wooldridge
expect a gender diversity advantage at middle management to be n load
(1997) noted that middle managers play crucial strategic roles when
strong in this collectivistic culture characterized by significant gender ed from
they “mediate, negotiate, and interpret connections between the orga-
inequality. In addition, the potential confounding effect of racial diver- https
nization's institutional (strategic) and technical (operational) levels”
sity would not be of great concern in our study using archival data ://onli
(p. 466). Moreover, middle managers also play key roles in developing
from South Korea, given the country's racial homogeneity, allowing us n elibr
and motivating employees (Kehoe & Han, 2020).
to focus more on gender effects. ary.wi
Therefore, middle managers can have a positive influence on
Our study makes several contributions to the gender diversity ley.com
the strategic operations and positioning of their organization
and strategic HRM literatures. First, the current study sheds critical /doi/
(e.g., improved decision making, strategy/system development, and
light on research on gender diversity. Previous research has focused 1 0 .100
implementation), thereby improving organizational performance
on how gender diversity at the top or bottom levels of the organiza- 2 /hrm
(Floyd & Wooldridge, 1992, 1994, 1997; Wooldridge & Floyd, 1990).
tion influences organizational performance (e.g., Carter et al., 2010; .22159
Though still underrepresented at middle management levels, women
Gonzalez & DeNisi, 2009). Although middle managers, who handle by RM
have made greater gains at this level than in the upper echelons of
the day-to-day operations of their organization, play central roles in IT U
organizations (Einarsdottir et al., 2018). Thus, it is important to shed
enhancing organizational performance, we have very little knowledge N IV E
greater light on the impacts that female representation at lower levels
about how and when gender diversity of middle management can R S ITY
of management may have on organizations. We introduce the term
enhance organizational performance. Our study provides some evi- L IB R
gender diversity advantage at middle management to indicate the
dence regarding the critical role that gender diversity of middle man- A R Y
potential enhancement of organizational performance from increased
agement plays in enhancing organizational performance via HPWS , W iley
gender diversity of middle management.1 Specifically, we argue that
implementation. Second, very little work has looked at the anteced- O n lin
increased gender diversity at middle management can enhance orga-
ents of HPWS systems or influences on their adoption and implemen- e Libr
nizational performance through strengthened HPWS implementation.
tation, and those that have focused largely on external or top-down ary on
However, simply increasing gender diversity of middle management is
drivers (Kim et al., 2021). Our work provides insights into the impor- [05/0
not beneficial for organizational performance; its advantage can be
tant ways in which the demographic characteristics of middle manage- 3 /2023
realized only when subordinates are gender diverse.
ment, namely gender diversity, can impact the possible development ]. See
In the current study, drawing from research suggesting that com-
and implementation of HR policies and practices that can contribute the Te
prehensive HR systems, known as high performance work systems
to an overall HPWS and organizational performance, contributing to rm s a
(HPWS), are positively associated with organizational performance
the gender diversity literature. Third, we attempt to clarify how and n d C o n
(Jiang et al., 2012), we argue that gender diversity of middle manage-
when gender diversity of middle management can be a competitive d ition
ment is indirectly and positively related to organizational performance
advantage for organizations and provide a new mechanism through s (http
via the improvement or development of HPWS. HPWS improvement
which a gender diversity advantage at middle management for HPWS s://onl
refers to the strengthening of existing HPWS through the implemen-
improvement and organizational performance is realized. Lastly, this inelibr
tation of additional high-value HR practices, which results in positive
study contributes significantly to the understanding of context in stra- ary.w
changes in the functions of HPWS and subsequent improvement in
tegic HRM research (i.e., South Korea) by exploring this phenomenon iley.co
organizational capabilities essential for performance in dynamic envi-
using a large sample of South Korean panel data and identifying a con- m /term
ronments (Jiang et al., 2012; Roberson et al., 2017).
textual factor (subordinate gender diversity) that may serve as a s-and
Specifically, we argue that an increased female representation at
boundary condition of a gender diversity advantage at middle -condi
middle management can facilitate the generation of information and management. tions)
knowledge around employee needs and desires, leading to more dif- o n W i
ferentiated solutions (Wiersema & Bantel, 1992). Given middle man- ley On
agers' important role in managing strategic and operational activities, 1 |
T H E O R E T I C A L F O U N D A T I O N S A N D line L
gender diversity of middle management has the potential to facilitate H Y P O T H E S E S ibrary
organizational performance indirectly through HPWS improvement, for rul
contingent upon nonmanagerial employees' gender diversity. HPWS 1.1 |
Strategic role of middle management es of u
is a comprehensive HR system intended to enhance organizational se; OA
performance and is arguably among the most effective means that
Strategy is often considered to be solely the domain of top manage- artic
middle managers can use to achieve strategic and operational goals.
ment. However, from the perspective of strategy as processual les are
By considering (a) HPWS improvement as an intermediary mechanism
(Whittington, 1993), stakeholders outside senior management, such gover
and (b) nonmanagerial employees' gender diversity as a boundary con-
as those in middle management, play a critical role in realizing strate- n ed by
dition, we help delineate how gender diversity of middle management
gic change because “strategy is a mix of deliberate intent and emer- the a
can indirectly enhance organizational performance. We term this key
gent actions and decisions, as a result of which formulation and p p licab
benefit a “gender diversity advantage at middle management.”
implementation of strategy are intertwined rather than being separate le Cre
We integrate research on gender diversity and strategic human
processes” (Currie & Procter, 2001, p. 55). Therefore, middle man- ative C
resources (HR) for our theorizing in the country of South Korea. We
agers are “not just passive recipients, but also active interpreters, o m m o n s License 109 JOO 9 ET AL. 3 0 5 0 x , 0, Dow
mediators, and intermediaries in implementing and reforming strategy
managers' impact on HR strategy is a function of their ability to influ- n load and policy e
” (Xie et al., 2013, p. 3481). In other words, middle managers
ence decision makers and exchange information with them in a pro- d from
are involved in organizing, directing, synthesizing, and controlling imple-
ductive way (Raes et al., 2011), middle managers must be able to https
mentation of strategic plans and thus have a substantial influence on
persuasively present and advocate for recommendations based on ://onli
strategic development processes (Floyd & Wooldridge, 2000; Huy,
collected information. Given the extant evidence that female middle n elibr
2001; Kehoe & Han, 2020; Rouleau & Balogun, 2011; Wooldridge
managers are more likely to engage in participative/democratic lead- ary.wi et al., 2008).
ership and collaborative negotiation, we propose that the presence of ley.com
We focus specifically on the key linking role that middle managers
higher gender diversity of middle management is likely to lead to /doi/
may play in connecting HR strategies and employees (Currie & greater improvement in HPWS. 1 0 .100
Procter, 2001). HR strategies are considered to be “third order' strate- 2 /hrm gies .2 …flowing
from, but also upwardly influencing…wider corporate 2 1 5 9
strategies” (Purcell, 2000, p. 62). Middle managers can play a signifi- 1.2 |
Gender diversity of middle management and by RM
cant role in HR strategy (Kehoe & Han, 2020), not only in terms of HPWS improvement IT U
downward implementation, but also through active identification of N IV E
beneficial practices and policies and the promotion of these policies
In order to effectively champion new ideas and alternatives regarding R S ITY
to organizational decision makers. In identifying, implementing (on a
HPWS, middle managers should pay attention to the needs of internal L IB R
local level), and providing resources to test emergent ideas, middle
human capital and provide suggestions for systems that will appropri- A R Y
managers engage in championing new ideas and alternatives, which is
ately address these needs. Middle managers should also be proficient , W iley
one of the primary pathways for middle managers' strategic involve-
at selling HPWS-related issues to top executives to obtain buy-in and O n lin
ment (Floyd & Wooldridge, 1992).
necessary financial resources (Dutton et al., 1997; Dutton & e Libr
The main purpose of championing new ideas and alternatives is
Ashford, 1993). Finally, they also must have an ability to encourage ary on
to adjust and improve the current content of strategy to benefit orga-
suggestions from employees and a strong willingness to implement [05/0
nizational functioning (Floyd & Lane, 2000; Mantere, 2005, 2008).
HPWS, helping their organization meet its employees' needs. In light 3 /2023
From an HR perspective, middle managers can serve as organizational
of these assertions, we identify a gender diversity advantage at middle ]. See
champions, persistently and persuasively advocating for entrepreneur-
management in that such diversity contributes to HPWS improve- the Te
ial or innovative options, initiatives, and strategies to senior manage-
ment; this is because of women's participative/democratic leadership rm s a
ment and HR management. They are implementers of organizational
in the process of gathering feedback and suggestions from subordi- n d C o n
strategy and observers of its effectiveness (Raes et al., 2011), and “act
nates and their collaborative negotiation in the process of solving d ition
as an initial screen” (Floyd & Wooldridge, 1994, p. 50) of HPWS
problems with senior managers and HR managers. Said differently, s (http
implementation in day-to-day operations as well as subordinates'
gender-diverse middle management (i.e., an increased proportion of s://onl
reactions to HPWS. If HPWS implementation is ineffective or subordi-
female middle managers) should lead subordinates to speak up about inelibr
nates have adverse reactions to HPWS, middle managers can select
issues (Eagly & Johnson, 1990) and should also possess a greater abil- ary.w
from a wide range of business opportunities, novel procedural pro-
ity to integrate diverse sets of information in ways that might lead to iley.co
posals, and administrative changes suggested at operating levels
innovative HR policy and practice solutions and then advocate for m /term
(Floyd & Wooldridge, 1994) and provide this information to strategic
those solutions (Eagly et al., 2003; Eagly & Johnson, 1990). Our argu- s-and
decision makers (Raes et al., 2011). Because middle managers “(live) in
ments to this effect are similar to those arguing for the benefits of -condi
the organizational space between strategy and operations,” they “are
gender diversity in new venture teams in the entrepreneurship litera- tions)
uniquely qualified to make such judgments” (Floyd & Wooldridge, 1994, ture (see Dai et al., 2019). o n W i
p. 50) and decide to which issues senior managers and HR managers
First, well-documented differences in information processing ley On
should pay attention (Wooldridge et al., 2008).
between males and females should lead to greater collection and integra- line L
As alluded to above, middle managers play two primary strategic
tion of suggestions from subordinates by middle management that is ibrary
roles: (1) collection of information regarding both the effectiveness of
more gender diverse. Women are likely to pay more attention to subtle for rul
the implemented strategy and potential suggestions and alternatives
information (Darley & Smith, 1995), are less likely to adopt problem- es of u
to existing strategy; and (2) identification, evaluation, and presentation
solving strategies that conform to existing beliefs or heuristics (Chung & se; OA
of new ideas and alternatives. Accordingly, we argue that two mana-
Monroe, 1998), retain more detailed information (Seidlitz & Ed, 1998), artic
gerial capabilities should be particularly important for successful
and have better recall after encountering new information (Erngrund les are
middle managers' effectiveness of influencing strategic HR decisions.
et al., 1996) than men. All of these findings support women's tendency gover
First, creating an environment where information regarding employee
to manage subordinates in a more participatory and democratic manner, n ed by
needs flows freely up to middle managers from subordinates is instru-
which promotes employee voice and input (Book, 2000; Greene the a
mental to information gathering (Kehoe & Han, 2020), because such
et al., 2003), thus providing the information necessary to understand the p p licab
an environment will make subordinates more willing to provide
needs of employees with respect to HR policies and practices. le Cre
insights and ideas on what employees believe would be valuable
Second, women tend to engage in greater levels of relational ative C
from an HR practice perspective. Second, considering that middle
information processing than men, meaning that they focus more on o m m o n s License 109 4 JOO 9 ET AL. 0 5 0 x , 0, Dow
the relationships between sets of individual pieces of information
which includes various measures such as pay, childcare costs, and per- n load
(Chung & Monroe, 1998), such as diverse suggestions from multiple
centage of senior positions (Economist, 2018). In South Korea, female ed from
subordinates as to HR policy preferences and needs. Such information
employees' contributions to their organization have been less valued or https
processing increases the likelihood that gender-diverse middle manage-
appreciated than those of their male counterparts, which results in gen- ://onli
ment will be able to craft innovative and norm-challenging HR policies
der inequality in promotion to top management (Cooper-Jackson, 2001) n elibr
and practices. However, this is only the first step to HPWS improve-
and a lack of women in upper management positions (Easton, 2015). ary.wi
ment. Middle managers must also persuade their senior management
Several critical factors constitute barriers to women's career advance- ley.com
to adopt new HR practices in order to accommodate the needs of
ment in South Korea, such as Confucianism, traditional norms, and gen- /doi/
employees (Ashford & Detert, 2015; Mirabeau & Maguire, 2014). Nota-
der stereotypes and roles (e.g., about married women with children) 1 0 .100
bly, despite its potential benefits at the operational level, improving
(Ibarra et al., 2013). Because women are burdened by both work and 2 /hrm
HPWS is not cost free to organizations, and therefore, finding a solu-
family demands and responsibilities, they feel the need to put extra .22159
tion that benefits all parties is important for the success of such negoti-
effort toward standing out in the workplace via their work performance by RM
ation (cf. Raes et al., 2011); it requires collaboration and problem
(Kim et al., 2020) while resolving the conflict between the roles of IT U
solving among all negotiation parties (Kong et al., 2014; Walton &
breadwinners and homemakers/caregivers (Cho et al., 2019). In addi- N IV E McKersie, 1965).
tion, many women in South Korea have limited networking and leader- R S ITY
Middle management's gender diversity can facilitate exploration
ship development opportunities, which make them struggle in rising to L IB R
of creative ideas and collaboration with senior managers and HR man-
management positions (Cho et al., 2017; Rowley et al., 2016). A R Y
agers, as women have been found to focus more on problem solving,
We focus on HPWS as an important mediator between gender , W iley
conflict resolution, and finding a mutually beneficial solution during
diversity of middle management and organizational performance in O n lin
negotiation (Brahnam et al., 2005; Henderson et al., 2013; Walters
South Korea because Bae and Lawler (2000) pointed out four reasons e Libr
et al., 1998). Taken together, these findings support the argument that
why HPWS could be important to implement in a collectivistic coun- ary on
gender-diverse middle management can help senior managers and HR
try such as South Korea. First of all, South Korea's collectivism facili- [05/0
managers formulate better HPWS strategy by directing attention
tates cooperation, loyalty, and harmony, which are aligned well with 3 /2023
toward strategic issues in regard to HPWS improvement and advocat-
HPWS (Lee & Johnson, 1998). Second, recent Korean culture can be ]. See
ing for resources for HPWS improvement. Hence, gender diversity of
characterized as a composite of Asian and Western values. That is, the Te
middle management can be a crucial driver of successful HPWS
Korean organizational culture shows that individualism and group har- rm s a
improvement (Gilbert et al., 2015).
mony are both equally strong. After the financial crisis in 1997, many n d C o n
In sum, since more gender-diverse middle management tends to
organizations operated their organizations with dynamic collectivism, d ition
focus on development, learning, individual/communal goals, and
reflecting multidimensional and paradoxical subcomponents: in-group s (http
voices (Paustian-Underdahl et al., 2014), with lateral and upward sup-
harmony, optimistic progressivism, and the hierarchical principle s://onl
port, they likely better serve as champions who gather information
(Cho & Yoon, 2001). Dynamic collectivism is defined as a mixture of inelibr
from subordinates about what issues require new ideas and alterna-
harmony and change, face-saving and aggressiveness, and emotional ary.w
tives that help improve the existing HPWS, and negotiate collabora-
community and impersonal achievement (Cho et al., 2014). iley.co
tively with senior managers and HR managers, thereby facilitating
The third reason is globalization (Bae et al., 2007). In order to m /term
HPWS improvement. In addition, HPWS can provide a context where
achieve better performance under globalization, organizations should s-and
female middle managers find a good fit with their cooperative man-
be more flexible to adapt to uncertainty in their environment. After -condi
agement style and soft skills (Kato & Kodama, 2017). Therefore, we
the 1997 financial crisis, the Korean Chaebols ti —large conglomerates— o n s)
predict a gender diversity advantage at middle management, such that
westernized their HRM systems (e.g., from seniority- to performance- o n W i
as more women are represented there, the greater HPWS improve-
based compensation) and increased employment flexibility (e.g., from ley On
ment will be. As such, we propose the following hypothesis:
high job security to more part-time or contingent job positions). Chae- line L
bols started emphasizing teamwork, employee empowerment, voice, ibrary
Hypothesis H1. Gender diversity of middle management
and merit-based hiring, incorporating many characteristics of HPWS for rul
in an organization is positively related to the level of HPWS
(Bae et al., 2007). Many South Korean organizations emphasize inno- es of u improvement.
vation, market culture, and risk-taking and implement HPWS (Rhee se; OA
et al., 2016). Relative to Western culture, collectivistic organizations artic
are more likely to encourage people to collaborate to better coordi- les are 1.3 |
Gender, management positions, and HPWS
nate and integrate organizations' skills and resources (Lee & gove in South Korea r
Miller, 1999). Knowledge exploitation and exploration through knowl- n ed by
edge transfer within the organization can be accomplished by such the a
Gender differences may play more critical roles in implementing HR
collaboration, which is aligned well with the core function of HPWS p p licab
practices in male-dominant cultures (Ciancetta & Roch, 2021; Hofstede (Miller & Shamsie, 1996). le Cre
et al., 2010; Williams & O'Reilly, 1998). South Korea is a relatively
Finally, because HPWS involves continuous training, it fits well ative C
unfriendly place for working women in terms of the glass-ceiling index,
with the fact that South Korean people value education and o m m o n s License 109 JOO 9 ET AL. 5 0 5 0 x , 0, Dow
development as a means to be competitive (Bae & Lawler, 2000). The
because they are more likely to have and accept unequal power distri- n load
above reasons explain the transformation of South Korean cultures,
butions within social institutions and for organizational representatives ed from
leading to more acceptance and willingness to work under HPWS
to invoke legitimate authority (Hofstede, 1984). Managers can flaunt https
(e.g., selective staffing, comprehensive training, outcome-based com-
authority more and enjoy greater hierarchical inequality in the higher ://onli
pensation, employee empowerment, participation in a decision making
power distance context, which thereby enhances the possibility that n elibr
process, etc.), and pointing to the need to understand the increasing
managers' strategic choices concerning low levels of HPWS will be ary.wi
role of HPWS in South Korean organizations. accepted. ley.com
Set against this context, we argue that the effect of gender diver- /doi/
sity of middle management on HPWS improvement is bounded by 1 0 .100 1.4 |
Subordinate gender diversity as a critical
gender diversity of subordinates for two reasons. First, while gender- 2 /hrm contingency
diverse middle management tends to have more engaging leadership .22159
and collaborative negotiation skills than gender-homogenous middle by RM
As previously discussed, national culture plays important role in imple-
management, the former faces challenges due to gender role incon- IT U
menting HR practices and perceiving leadership (Schuler, 2013;
gruity (i.e., a mismatch between their gender and middle manager role) N IV E
Stoeberl et al., 1998). We also argue that the national context of our
in South Korea. Whether female middle managers can effectively play R S ITY
data collection (South Korea) may influence the degree to which inter-
the champion role that facilitates HPWS improvement depends on L IB R
nal context (subordinate gender diversity) can shape the relationship
the internal environment; one important aspect of the internal envi- A R Y
between gender diversity of middle management and HPWS improve-
ronment is organizational demography. We argue that gender diver- , W iley
ment. Employees who are in individualistic countries may have an
sity of subordinates is a demography-related boundary condition. O n lin
independent view, while employees who are in collectivistic countries
South Korea ranked dead last in terms of gender egalitarianism in e Libr
such as South Korea, Japan, and China emphasize an interdependent
House et al.'s (2004) study; the culture values a masculine leadership ary on
view stressing connectedness and relationships (e.g., Guanxi; Chen
style, and thus, female middle managers in South Korea likely face [05/0
et al., 2013) as well as high vertical collectivism 3 / —high collectivism and
prejudice and discrimination from their male subordinates (cf. Eagly & 2 0 2 3
high power distance (Singelis et al., 1995). For example, organizations in
Carli, 2007; Eagly & Karau, 2002). ]. See
Japan still maintained traditional East Asian HR system (e.g., seniority-
For example, female middle managers, who are not common in the Te
based compensation and promotion, long-term employment; Bae
South Korea, may be viewed by male subordinates as less competent rm s a
et al., 2010). Traditional HR management in East Asian countries relied
or effective and thus be less accepted by male subordinates. In a n d C o n
more on informal personal relationships and people's judgment than on
male-dominant internal environment, female middle managers may be d ition
official and objective criteria and regulations (Lam et al., 2002; Xian
compelled to adjust their leadership behaviors to more dominant and s (http
et al., 2019). Confucian values and the Chinese family business model
assertive ones to match their male subordinates' preferences for mas- s://onl
based on traditional authority are pervasive in South Korea as well
culine leadership styles. By doing so, gender-diverse middle manage- inelibr (Claessens et al., 2000).
ment loses its advantage in participative/democratic leadership and ary.w
Traditional South Korean HR management has prevented
collaborative negotiation. As a result, the benefit of gender diversity iley.co
employees from becoming involved in decision making processes as
of middle management for HPWS improvement is less likely to hap- m /term
information was passed down from senior management (Deyo, 1989;
pen. By contrast, in an internal environment characterized by a higher s-and
Frenkel & Lee, 2010) and emphasized employee loyalty and long
level of subordinate gender diversity, female middle managers are not -condi
working hours. In these contexts, individuals perceived typical male
as threatened by negative gender stereotypes or role incongruity tions)
leadership styles as more appropriate than female leadership
(Paustian-Underdahl et al., 2014). Therefore, gender diversity of mid- o n W i
(Triandis, 1995); a majority of organizations did not have any women
dle management can translate into more widespread participative/ ley On
on their board in those counties (Low et al., 2015; The Japan
democratic leadership and collaborative negotiation, facilitating a fer- line L
Times, 2021; The Korea Herald, 2020). Despite high levels of technol-
tile environment for HPWS improvement. ibrary
ogy and economic development, traditional social and cultural norms
Second, the new ideas and alternatives that gender-diverse mid- for rul
are still alive, influencing beliefs about gender roles in those countries
dle management champions for HPWS improvement can come from es of u (Brooks, 2006).
subordinates. Female and male subordinates have different life and se; OA
While employees in countries where individualism is highly valued
work experiences. Due to such differences, they likely have different artic
seek opportunities and incentives for themselves, collectivism would
views on how the existing HPWS is working and how it should work. les are
lead employees to reorient their goals toward collective benefits.
Gender diversity, despite its potential for engendering conflict, can be gover
According to Schuler (2013), in countries where collectivism is strong,
a pivotal source of novel and useful ideas (Egan, 2005; Jackson n ed by
such as East Asian countries, HR practices that facilitate information
et al., 2003). Stated differently, gender diversity is a source of creative the a
sharing may be less prevalent. High power distance also prevents the
ideas and alternatives in that male and female employees have a p p licab
drive to seek personal benefits and would be well aligned with auto-
greater range of viewpoints and lived experiences (Mateos de Cabo le Cre
cratic leadership (Dorfman et al., 2012). In high power distance cultures,
et al., 2012) from which to draw to create suggestions for changes ative C
employees may agree with the strategic choices of management
and/or improvements in HR policies and practices. With these ideas o m m o n s License 109 6 JOO 9 ET AL. 0 5 0 x , 0, Dow
and alternatives, gender-diverse middle management is in a better
and comprehensive training opportunities are intended to help n load
position to collect information used to understand the issues and
employees acquire and develop their work-related knowledge, skills, ed from
potential solutions based on HPWS improvement.
and competencies. Practices such as developmental feedback, competi- https
Taken together, the above arguments lead us to hypothesize that
tive compensation, performance-based incentives, high-quality bene- ://onli
gender diversity of subordinates strengthens the link between gender
fits, clear promotion and career development opportunities, and job n elibr
diversity of middle management and the level of HPWS improvement.
security are designed to enhance employees' work motivation. Finally, ary.wi
practices such as team-based work, employee involvement systems, ley.com
Hypothesis H2. Subordinate gender diversity modifies
flexible job designs, and pathways for information sharing provide /doi/1
the relationship between gender diversity of middle man-
employees with the opportunity to use skills and motivation in ways 0 .100
agement in an organization and the level of HPWS
that achieve organizational objectives (Boon et al., 2019). 2 /hrm
improvement, such that this relationship is stronger when
As HPWS is improved, employees will have stronger ability, .22159
subordinate gender diversity is higher.
stronger motivation, and more opportunities to use their ability and by RM
motivation for their work (Huselid, 1995; Ramsay et al., 2000). Thus, IT U
employees can more efficaciously participate in collaborative decision N IV E 1.5 |
HPWS improvement and organizational
making processes and work with others in delivering high-quality R S ITY performance
output, thereby increasing their work efficiency (Boon et al., 2019) L IB R
and ultimately facilitating organizational performance. A R Y
HPWS, as a unique, inimitable, causally ambiguous, and synergetic , W iley
mechanism, can deliver organizations a sustainable competitive
Hypothesis H3. The level of HPWS improvement is posi- O n lin
advantage (Lado & Wilson, 1994). Wright et al. (1994) argued that
tively related to the level of organizational performance. e Libr
individual HR practices are not a competitive advantage as HR ary on
practices can easily be copied by competitors. However, it is difficult
Autonomous strategic behaviors are defined as activities that [05/0
to imitate bundles of HR practices due to the interdependence of
create substantial internal variations by diverging from or substituting 3 /2023
multiple practices and their synergistic effect (Lado & Wilson, 1994).
for existing strategic plans (Burgelman, 1991; Mirabeau et al., 2018). ]. See
Characteristics of HPWS bundles, such as unique historical paths, can
For reasons explained above, we argue that gender diversity of the Te
make them difficult for competitors to copy (Becker & Huselid, 2006).
middle management may create significant content-based variance rm s a
Indeed, Saridakis et al. (2017) found that HPWS, as a bundle of multi-
(i.e., implemented HR practices) between organizations by actively n d C o n
ple HR practices, facilitated organizational performance to a greater
facilitating autonomous strategic behaviors such as championing new d ition
extent than individual HR practices, and this effect appeared for both
ideas and initiatives related to programs and practices that constitute s (http
financial performance (e.g., accounting returns) and operational
HPWS (Noda & Bower, 1996). As a result, gender diversity of middle s://onl
performance (e.g., workforce productivity). The positive implication of
management represents a potential source of heterogeneity among inelibr
HPWS for organizational performance has been demonstrated using
organizations, which is the primary driver of “organizations' ability to ary.w
various indicators of organizational performance (e.g., Huselid, 1995;
derive rents from complex HR systems such as HPWS i ” (Chadwick & ley.co
Jiang et al., 2012; Messersmith et al., 2011; Saridakis et al., 2017) in
Flinchbaugh, 2021, p. 35), and thus creates a gender diversity advan- m /term
various countries, such as China, Japan, South Korea, Spain, Canada,
tage at middle management. HPWS is arguably among the most effec- s-and
the U.K., and the U.S. (Choi, 2014; Huselid, 1995; Prieto &
tive means for middle managers' achievement of strategic goals, as -condi
Santana, 2012; Rabl et al., 2014; Sun et al., 2007; Takeuchi
HPWS fosters human capital and employee motivation, which are two tions) et al., 2007).
important drivers of organizational performance (Jiang et al., 2012; o n W i
Following previous research (e.g., Jiang et al., 2012; Lepak
Messersmith et al., 2011) (see Figure 1 for our conceptual model). ley On
et al., 2006), we argue that HPWS has ability-, motivation-, and line L
opportunity-enhancing (i.e., AMO-enhancing) functions, satisfying
Hypothesis H4. The level of HPWS improvement medi- ibrary
employees' psychological needs for competence, relatedness, and
ates the interactive effect of gender diversity of middle for rul
autonomy (Ryan & Deci, 2000), and thus can help improve organiza-
management and subordinate gender diversity in an orga- es of u
tional performance. HPWS refers to “organizational actions or pro-
nization on the level of organizational performance. se; OA
cesses and job characteristics that focus on attracting, developing, artic
and motivating employees and providing opportunities to contribute” les are
(Boon et al., 2019, p. 2518). Lepak et al. (2006) conceptualized gover
systems of HR practices as falling into three distinct dimensions n —skill-, 2 | M E T H O D ed by
motivation-, and opportunity-enhancing HR practices. Stated the a
differently, HPWS provides three components to foster employee 2.1 | Sample p p licab
performance: ability, motivation, and opportunity (Jiang et al., 2012; le Cre
Lepak et al., 2006; Messersmith et al., 2011). Practices such as stan-
To test our hypotheses, we used the Workplace Panel Survey (WPS) ative C
dardized and well-developed recruitment practices, rigorous selection,
data2 collected in 2009, 2011, and 2013. This survey was conducted o m m o n s License 109 JOO 9 ET AL. 7 0 5 0 x , 0, Dow F I G U R E 1 Conceptual model n loa Gender Diversity d ed of Subordinates from (T1) h ttps://onlinelibrary.wile Gender Diversity High Performance y .c Organizational o m of Middle Work System /d Performance o i Management Improvement /10 (T3) .1 (T1) (T2-T1) 0 0 2 /hrm.22159 by RM
by the Korea Labor Institute, a government-funded policy research
(1977). We calculated this index by subtracting the sum of the respec- IT U
body, in tandem with the South Korean Ministry of Labor. Approxi-
tive squared proportions of male and female employees from 1; the N IV E
mately 1700 firms with 30 or more employees across the country
score ranged from 0 (most homogeneous) to 0.5 (most diverse) R S ITY
were randomly selected using stratified sampling and surveyed.
(Gonzalez & Denisi, 2009; Richard et al., 2004). L IB R
Organization-level data in 2009 and 2011 were available for 1737 A R Y
and 1770 organizations, respectively, while organization-level data in , X , W iley
2013 were available for 1621 organizations. B:I: ¼ 1 Pi2 O n lin
Individual interviews (Blaise system-based Computer-Assisted e Libr
Personal Interviewing) were conducted to obtain information related
where, P = the proportion of subordinates in the organization (male ary on
to human resources and labor relations. Mail and web surveys were
or female); i = the number of different categories represented in the [05/0
conducted to obtain data on employment status and financial status. organization (=2). 3 /2023
Organizations in the forestry, mining, fishing, and agriculture sectors,
HPWS improvement. We included HR practices to measure ]. See
as well as public service organizations, were excluded from the sam-
HPWS in the current study, which are highly overlapped with those in the Te
pling frame. The organization-level survey was completed by the HR
previous studies. According to the recent comprehensive review con- rm s a
manager and a labor union representative of each organization. The
ducted by Boon et al. (2019), HR practices most commonly used to n d C o n
data of the endogeneity control variables were collected in 2007
build HPWS are categorized into the following: training, participation/ d ition
(Time 0), those of the predictor and control variables in 2009 (Time 1);
autonomy, incentives, performance appraisal, recruiting, and selection, s (http
those of the mediator (i.e., HPWS improvement) in 2011 (Time 2), and
which are typically identified as core HR practices for HPWS s://onl
those of organizational performance (profitability) in 2013 (Time 3).
(Posthuma et al., 2013). We selected five of these six HR practice cat- inelibr
We matched the organization-level panel data across the 3 years
egories for our HPWS construct (Boon et al., 2019; Lepak ary.w
using unique identification numbers. The resulting final sample com-
et al., 2006). This HPWS composition was also similarly used across iley.co
prised 1101 organizations that provided suitable and usable data for
countries (e.g., the United States, China, Spain, the United Kingdom, m /term
analysis. After we matched identified numbers in each year, we used
Canada, and South Korea) (Rabl et al., 2014). s-and
1066 organizations to test Hypotheses H1 and H2. 621 organizations
Specifically, to create an index of HPWS in each organization, we -condi
were used to test Hypotheses H3 and H4. The response rates for
used twelve HR practices that encompass the types of practices iden- tions)
original samples were 70.3% (2009), 62.4% (2011), and 54.9% (2013),
tified by prior studies (cf., Huselid, 1995; Jiang et al., 2012): training, o n W i respectively.
recruiting, a regular and formal performance appraisal system, annual ley On
salary, profit sharing, incentives, employee suggestion program, small line L
group activities, regular team meetings, job rotation, multi-functional ibrary 2.2 |
Measures of key variables
training, and autonomy. For all practices except autonomy, HR man- for rul
agers responded to items (1 = yes, 0 = no), indicating whether or not es of u
Gender diversity of middle management (proportion of female middle
their respective organizations had implemented the respective HR se; OA
managers) (Time 1). Female middle managers included female team- practices. artic
or project-based executives, deputy general managers, and depart-
For autonomy, HR managers indicated the levels of autonomy in les are
ment managers. We aggregated them to derive the total number of
their organizations with respect to task, speed, recruiting, and training, gover
female middle managers in an organization. Then, we calculated gen-
ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item n ed by
der diversity of middle management in each organization by dividing
was “the extent to which a major operating unit (e.g., team) of your the a
the total number of female middle managers by the total number of
workplace has autonomy regarding how to work at the unit. p ” We p licab middle managers (Kalev, 2009).
transformed these four scales into dummy codes (1 l –2 = 0, 3–4 = 1) e Cre
Gender diversity of subordinates (Time 1). Subordinates'gender
and then aggregated all of them to create an autonomy index. We ative C
diversity in each organization was indicated by the Blau's index
again transformed this value into dummy codes (0 o –2 = 0, 3–4 = 1), m m o n s License 109 8 JOO 9 ET AL. 0 5 0 x , 0, Dow
with 1 representing the effective implementation of autonomy prac-
organization were held by foreigners, using a percentage ranging from n load
tices and 0 representing the ineffective implementation or lack of 0% (no foreign stake) to 100%. ed from autonomy practices.
Proportion of female executives. As noted earlier, the proportion https
Ten items were summed to measure training. These items were
of female strategic leaders was found to predict organizational perfor- ://onli
corporate education, external group training, internet-based education
mance (Post & Byron, 2015). Therefore, we controlled for the propor- n elibr
and training, support for meetings to learn work-related skills,
tion of female executives in each organization, calculated by dividing ary.wi
employee training on the technical guidance, technical guidance, and
the number of female executives by the total number of executives. ley.com
employee training from the parent company/trust companies, paid or
In order to address the concern about firm-specific unobservable /doi/
unpaid day release training, tuition support for training institutions
characteristics, we included a variable related to management salary 1 0 .100
(including school), financial aid/loans for educational institutions, and
and a locality variable as the proxy of corporate culture in the models. 2 /hrm
participation in employee training programs during working hours.
Regarding management salary, we included the total salary of a middle .22159
Two items were used to measure recruiting: multi-media (e.g.,
manager before tax (first-year salary). Organizations in Seoul are more by RM
newspaper, radio, and television) recruiting and internet-based recruiting
likely to recruit and retain female employees (i.e., higher levels of gen- IT U
(1 = yes, 0 = no). We used four items to measure what types of incen-
der equality for all positions, including top management) and have N IV E
tives were in place (1) group performance-based incentive, (2) company
progressive and innovative corporate culture than organizations else- R S ITY
performance-based incentive, (3) business unit performance-based incen-
where in South Korea. It is plausible that organizations in Seoul have a L IB R
tive, and (4) department or team performance-based incentive
more progressive corporate culture, and therefore, organizations in A R Y (1 = yes, 0 = no).
Seoul are more likely to implement HPWS (Bae & Lawler, 2000) and , W iley
We summed each item to create indices of training (10 items),
promote gender diversity to a greater extent than organizations else- O n lin
recruiting (2 items), and incentives (4 items). It is appropriate to sum
where. Corporate cultures that emphasize gender equality or devalue e Libr
all indicators of HR practices because these practices' synergistic
discrimination toward women should be more conducive to overall ary on
effects on organizational performance are claimed to be realized from
gender diversity and female employees' career advancement, helping [05/0 their complementarity in bundles (Becker & Gerhart, 1996;
more female employees be promoted to management positions. For 3 /2023
MacDuffie, 1995). To obtain the improvement of HPWS from Time
example, most Chaebols where their culture is progressive and inno- ]. See
1 to Time 2, we aggregated all the practices and then subtracted
vative are in Seoul (dummy coded, 1 = Seoul, 0 = other areas). Our the Te
HPWS at Time 1 from HPWS at Time 2 (i.e., formative measure).
correlation table supports our arguments that locality could be a proxy rm s a
Organizational performance (Time 3). We focused on organiza-
for corporate culture. Being an organization in Seoul is positively cor- n d C o n
tional profitability. Following prior research, we divided the logarithm
related with gender diversity of middle management (r = 0.11, d ition
of net profit for 2013 (fiscal year) by the organizational size in 2013
p < 0.01), HPWS level at Time 2 (r = 0.07, p < 0.05), and gender diver- s (http
(indicated by the logarithm of the total number of full-time
sity of subordinates (r = 0.10, p < 0.01) as stated. s://onl employees; Shaw et al., 2013). inelibrary.w 2.4 | Endogeneity control iley.co 2.3 |
Measures of control variables m /term
We addressed the endogeneity of the relationship between gender s-and
We controlled for several variables theoretically predictive of organi-
diversity of middle management (i.e., the proportion of female middle -condi
zational performance. The data of these control variables were avail-
managers) and HPWS improvement by providing the fixed effect esti- tions)
able from the organization-level survey described above.
mates on the effect of gender diversity of middle management on o n W i
Industry differences. Industries affect organizational performance
HPWS improvement. We controlled for the possibility that gender ley On
(Datta et al., 2005). To account for the industry differences, we used
diversity of middle management is more likely to be related to HPWS line L
industry dummy variables created based on WPS-provided industry
improvement. Following prior research (Tang et al., 2018), we first ibrary codes.
regressed gender diversity of middle management at time t against a for rul
Organizational size. The larger an organization, the more effica-
set of antecedent variables measured 2 years prior, that is, year t 2. es of u
cious HR practices it may have (Arthur, 1994). Thus, an organization's
These instrumental variables included HR strategy s —whether the pri- e; OA
size could predict its performance. We operationalized the organiza-
mary goal of HR is to maximize savings on fixed labor costs and artic
tional size as the logarithm of the number of full-time employees.
whether HRM is based on individual performance (a five-point Likert les are
Organizational age. More established organizations tend to be
scale from 1 to 5). In order to account for the possibility that prior gover
less vulnerable to performance pressure (Hannan & Freeman, 1984).
birthrate in each organization's location might drive a higher level of n ed by
Thus, we controlled for the age of an organization, which was calcu-
gender diversity of middle management in the areas in which each the a
lated as the difference between 2009 and the founding year.
organization operates, we included birthrate ratio (the ratio of boys p p licab
Foreign stake. An organization's foreign stake can predict its per-
per one hundred girls) measured in 1990 (this year is the earliest one le Cre
formance (Douma et al., 2006). To measure an organization's foreign
we could obtain) by using the Korean government database (https:// ative C
stake, its HR managers indicated the extent to which the shares of the
kostat.go.kr/portal/eng/index.action).3 In addition, we included other o m m o n s License 109 JOO 9 ET AL. 9 0 5 0 x , 0, Do TA B L E 1 w
Descriptive statistics and correlations n loaded Variable M SD 1 2 3 4 5 6 7 8 9 from h 1 Locality (proxy of 0.52 0.50 tt –– p s: corporate culture) //online 2 Management's salary 35.51 11.46 0.06* l –– ibrary 3 Service 0.50 0.50 0.20** 0.04 .w –– iley. 4 Finance 0.04 0.19 0.12** 0.20** 0.20** c –– o m /d 5 Organizational age 24.92 17.47 0.06* 0.07* 0.02 0.01 o i –– /10.1 6 Organizational size 4.98 1.24 0.10** 0.28** 0.01 0.07* 0.27** 0 0 –– 2 /hrm 7 Foreign stake 4.21 15.99 0.01 0.09** 0.13** 0.06* 0.01 0.16** –– .2215 8 Proportion of female 0.05 0.15 0.02 0.06* 0.07* 0.03 0.02 0.11** 0.01 9 –– by executives R M IT 9 GDMM (T1) 0.10 0.16 0.11** 0.02 0.31** 0.00 0.03 0.03 0.01 0.29** U –– N IV E 10 HPWS (T1) 8.63 4.56 0.04 0.43** 0.02 0.11** 0.05 0.31** 0.12** 0.09** 0.03 R S IT 11 HPWS (T2) 9.19 4.50 0.07* 0.32** 0.01 0.10** 0.07* 0.38** 0.19** 0.11** 0.04 Y L IB R 12 HPWS improvement 0.56 4.25 0.03 0.12** 0.01 0.01 0.02 0.07* 0.08* 0.01 0.08* A R (T2 Y –T1) , W il 13 Gender diversity of e 0.31 0.16 0.10** 0.10** 0.03 0.15** 0.02 0.09** 0.00 0.07* 0.18** y O n subordinates (T1) line Li 14 Organizational 1.63 0.55 0.11** 0.14** 0.07 0.06 0.09* 0.22** 0.10* 0.04 0.05 b rary performance (T3) on [0 15 Endogeneity control 0.10 0.03 0.28** 0.11** 0.16** 0.13** 0.02 0.16** 0.13** 0.04 0.20** 5 /03/20 Variable 10 11 12 13 14 15 2 3 ]. Se 10 HPWS (T1) e –– the T 11 HPWS (T2) 0.56** er –– m s an 12 HPWS improvement (T2 d –T1) 0.48** 0.46** –– C o n 13
Gender diversity of subordinates (T1) 0.14** 0.14** 0.00 d –– itions 14
Organizational performance (T3) 0.09* 0.18** 0.10* 0.06 ( –– h ttps: 15 Endogeneity control 0.24** 0.21** 0.03 0.16** 0.05 // –– o n line
Note: N = 1101. GDMM refers to gender diversity of middle management. Organizational performance (T3) was available from 644 organizations. The unit librar
of management's salary is one million won. T1 = Time 1; T2 = Time 2; and T3 = Time 3. *p < 0.05; **p < 0.01. y .w iley.com/term
HR variables associated with work-life balance (i.e., selective work
conditional mediation model (Hayes, 2013). Before analyzing the data, s-and
hours and flexible work hours; yes = 1, no = 0) (Avery &
we centered the components for the interaction term around their -condi
McKay, 2006; Kato & Kodama, 2017) and family-friendly HR policies
respective means. Table 1 presents the descriptive statistics and cor- tions)
(maternity leave, parental leave, workplace childcare facilities, support
relations. We included the control variables at Step 1. Next, we exam- o n W i
for childcare cost, leaves for regular doctor visits during pregnancy,
ined the main effect of gender diversity of middle management at ley On
stillbirth vacation, and infertility leave; yes = 1, no = 0). We aggre-
Step 2. At Step 3, we added the moderating variable of gender diver- line L
gated all scores related to family-friendly HR policies. We also
sity of subordinates and its associated two-way interaction term ibrary
included favored recruitment processes for women (yes = 1, no = 0).
(i.e., gender diversity of middle management gender diversity of for rul
We then generated a predicted gender diversity of middle manage-
subordinates). As shown in Table 2, at Step 1, some of the control var- es of u
ment score and included it as an endogeneity control in the models.4
iables (i.e., management salary, organizational size, foreign stake, se; OA
HPWS at Time 1, and endogeneity control) reached (marginal) signifi- artic
cance, which accounted for 29% of the variance. Hypothesis H1 pre- les are 2.5 | Results
sents that gender diversity of middle management in an organization gover
is positively related to the level of HPWS improvement. At Step n ed by
Since all data have been conducted at an organizational level, multile-
2, gender diversity of middle management significantly predicted the the a
vel analysis such as hierarchical linear modeling or generalized latent
level of HPWS improvement (Model 2a: p
b = 1.95, SD = 0.73, p licab
and mixed modeling (GLLAMM) is not necessary for our study. We
p < 0.01), which supported Hypothesis H1. Hypothesis H2 states le Cre
believe that the PROCESS macro is a good tool to explore our
that subordinate gender diversity modifies the relationship between ative Commons License 109 10 JOO 9 ET AL. 0 5 0 x , 0, Do TA B L E 2 w
Results of hierarchical regression predicting high performance work system (HPWS) improvement n loaded
HPWS improvement (T2-T1) from htt Predictor
Model 1a b (SD)
Model 2a b (SD)
Model 3a b (SD) p s://on Intercept 0.39 (0.58) 0.59 (0.58) 0.66 (0.58) linelibr Control variables ary.wil
Locality (proxy of corporate culture) 0.04 (0.23) 0.03 (0.23) 0.11 (0.23) ey.com Management's salary 0.00 (0.00)** 0.00 (0.00)** 0.00 (0.00)** /doi/1 Service 0.19 (0.23) 0.02 (0.24) 0.05 (0.24) 0 .1002 Finance 0.42 (0.62) 0.35 (0.62) 0.31 (0.62) /hrm.22 Organizational age 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 1 5 9 b Organizational size 0.65 (0.10)*** 0.66 (0.10)*** 0.70 (0.10)*** y R M I Foreign stake 0.03 (0.01)*** 0.03 (0.01)*** 0.03 (0.01)*** T U N IV
Proportion of female executives 1.18 (0.76) 1.74 (0.79)** 1.82 (0.78)** E R S IT HPWS (T1) 0.55 (0.03)*** 0.55 (0.03)*** 0.56 (0.03)*** Y L IB Endogeneity control 6.66 (3.61)* 5.22 (3.64) 4.88 (3.62) R A R Y Independent variables , W ile
Gender diversity of middle management (T1) 1.95 (0.73)*** 1.26 (0.74)* y O n li
Gender diversity of subordinates 2.50 (0.71)*** n e Libr
Gender diversity of middle management Gender 15.48 (4.27)*** ary o diversity of subordinates n [05/0 F-value 43.77*** 40.67*** 36.99*** 3 /202 R2 0.293 0.298 0.314 3 ]. See th
Note: N = 1,101. GDMM refers to gender diversity of middle management. T1 = Time 1; T2 = Time 2; T3 = Time 3. e Te
*p < 0.10; **p < 0.05; ***p < 0.01. rm s and Condition
gender diversity of middle management in an organization and the
bias-corrected 95% confidence intervals (CI ) to determine the sig- s 95% (http
level of HPWS improvement, such that this relationship is stronger
nificance of the conditional indirect effect. We tested the conditional s://onl
when subordinate gender diversity is higher. At Step 3, we found that
indirect effect using the SPSS PROCESS macro version 26 (Model 7; inelibr
the interaction of gender diversity of middle management and gender
Hayes, 2013). The conditional indirect effect was only found in the ary.w
diversity of subordinates was positively related to the level of HPWS presence of a higher ( i
b = 0.16, SD = 0.06, CI = [0.04, 0.30]) versus le 95% y .co
improvement (Model 3a: b = 15.48, SD = 4.27, p < 0.01). We found
lower level of subordinates' gender diversity (b = 0.07, SD = 0.06, m /term
that the respective variances explained by gender diversity of middle CI s 95% = [
0.22, 0.03]), thus supporting Hypothesis H4 (Table 4). -and
management, and the interaction term for HPWS improvement were -condi
1% and 2%. A follow-up simple slope test (see Figure 2) indicated that tions)
the relationship between gender diversity of middle management in 2.6 | Supplementary analyses o n W i
an organization and the level of HPWS improvement was positive ley On
(simple slope = 4.43, t = 5.56, p < 0.01) in the presence of higher gen-
Organizations that have greater improvement in HPWS may have line L
der diversity of subordinates (+1 SD), but was not significant (simple
more gender-diverse middle management. To address this concern, ibrary slope = 0.52, t =
0.77, p = 0.44) in the presence of a lower level
we used a cross-lagged approach (Finkel, 1995), which is “the optimal for rul of gender diversity of subordinates ( 1 SD). Accordingly,
way to understand causality in field settings” (Lang et al., 2011, es of u Hypothesis H2 was supported.
p. 605). This approach allowed us to compare the partial regression se; OA
Consistent with Hypothesis H3, the level of HPWS improvement
coefficients between variables measured at different time points. We artic
was positively related to organizational performance at Time 3 (Model
found that gender diversity of middle management at Time 1 was pos- les are
2b: b = 0.03, SD = 0.01, p < 0.01). HPWS improvement explained
itively related to the level of HPWS improvement at Time 2 (b = 0.05, gover
4.2% of the variance of organizational performance (Table 3).5
SD = 0.64, p < 0.05), but HPWS improvement at Time 1 was not sig- n ed by
Finally, to test Hypothesis H4, which states that the level of
nificantly related to gender diversity of middle management at Time the a
HPWS improvement mediates the interactive effect of gender diver- 2 ( p
b = 0.03, SD = 0.00, p = 0.28). The linkage between HPWS and p licab
sity of middle management and gender diversity of subordinates in an
organizational performance has been criticized for ambiguity of cau- le Cre
organization on the level of organizational performance, we per-
sality (Shin & Konrad, 2014; Wright et al., 2005). Thus, a reversed ative C
formed 10,000-sample bootstrapping (Hayes, 2013) and used the
causal relationship between organizational performance and the o m m o n s License

Gender diversity advantage at middle management:Implications for high performance work systemimprovement and organizational performance

Thông tin :

Tài liệu liên quan:

Cuộc thi “Nhà quản trị nhân sự tương lai” - Tài liệu tham khảo | Đại học Hoa Sen

L1-An Introduction to Human Resource Management - Tài liệu tham khảo | Đại học Hoa Sen

Models for warehouse management Classification and - Tài liệu tham khảo | Đại học Hoa Sen

Tổng hợp trắc nghiệm - Tài liệu tham khảo | Đại học Hoa Sen

Các phúc lợi của nhân viên tại Vinamilk trong quản trị nhân lực - Tài liệu tham khảo | Đại học Hoa Sen