Journal of Business Research 56 (2003) 1009 – 1019

Gender diversity in management and firm performance:

the influence of growth orientation and organizational culture$
Sean Dwyera,*, Orlando C. Richardb, Ken Chadwickc
a
Department of Management and Marketing, College of Administration and Business, Louisiana Tech University, P.O. Box 10318, Ruston, LA 71272, USA
b
Department of Management, School of Management and Administration, University of Texas-Dallas, Dallas, TX, USA
c
Department of Management and Marketing, College of Business Administration, Nicholls State University, Thibodaux, LA, USA
Received 7 August 2000; accepted 6 September 2001

Abstract

This study examines the influence of gender diversity in management on firm performance. The management group examined was
composed of all firm members considered to be managers and officials, a broader level of analysis than past management-level diversity
research that has primarily focused on groups composed of top management team (TMT) members. Adopting contingency and
configurational approaches, gender diversity’s interactions with two key organizational variables—organizational culture and growth
orientation—were evaluated against organizational-level performance measures. Supporting contingency theory and configurational theory,
the results suggest that gender diversity’s effects at the management level is conditional on, that is, moderated by, the firm’s strategic
orientation, the organizational culture in which it resides, and/or the multivariate interaction among these variables. These findings help
reconcile conflicting results of past diversity – performance research by suggesting that an appropriately configured and supportive
organizational environment may need to be in place before the beneficial aspects of gender diversity can be fully realized.
D 2003 Elsevier Science Inc. All rights reserved.

Keywords: Diversity; Gender diversity; Strategy; Organizational culture; Culture; Growth orientation; Management; Performance

1. Introduction In addition, past top management group research has

generally limited its examination of group diversity effects
Considerable progress has been made with respect to to the top management team (TMT) (Finkelstein and Ham-
gender diversity in management as women increasingly brick, 1996). Yet, studies have shown that middle managers
move into the managerial ranks (Elsass and Graves, 1997). play a key role in influencing strategic decision-making and
Yet, gender diversity in management has received insufficient implementation (e.g., Burgelman, 1994; Floyd and Wool-
attention in the research literature. Strategic management dridge, 1992; Kanter, 1982). We seek to extend this research
researchers have studied diversity primarily in terms of and more fully examine gender diversity’s effect at the
tenure, education, and functional background. The domain managerial level. We do so by broadening our examination
of work force diversity, however, encompasses a broader of the management group to include not only senior exec-
demographic perspective (Cox, 1994). Furthermore, scholars utives who set corporate strategy and policies but also those
have recently noted that different types of diversity have managers who implement these policies—middle manage-
divergent effects within organizations (Pelled, 1996; Pelled et ment, department managers, and salaried supervisors.
al., 1999). In particular, Pelled (1996) argues that nonvisible Empirical support for the diversity – performance link
attributes have quite different effects on performance than do has, in general, been mixed (Williams and O’Reilly,
visible attributes. Thus, knowledge of one type of diversity’s 1998). This suggests that the influence of diversity on firm
impact on the firm may provide little, if any, insight into performance may, at least in part, depend on the organiza-
another type’s influence. tional context. That is, the effect of diversity on performance
may lie in the interaction of diversity with contextual
$
variables. However, little empirical research has been com-
The first two authors contributed equally to this article.
* Corresponding author. Tel.: +1-318-257-3584; fax: +1-318-257-4253. pleted in this regard. Consistent with a growing body of
E-mail addresses: [email protected] (S. Dwyer), curtae69@ organizational demography research adopting a contingency
aol.com (O.C. Richard), [email protected] (K. Chadwick). approach (e.g., Hambrick et al., 1996; Neale et al., 1999;

0148-2963/03/$ – see front matter D 2003 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0148-2963(01)00329-0
1010 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019

Pelled et al., 1999; Richard, 2000; Smith et al., 1994) and 1994; Kanter, 1982). Studies of mid-management’s influ-
strategic human resource management research espousing a ence on the strategic management process have shown, in
configurational approach (e.g., Baker and Cullen, 1993; fact, a positive relationship between their involvement in
Delery and Doty, 1996; Dess et al., 1993), we position strategy and firm performance (Schilit, 1987; Wooldridge
diversity within a broader set of interrelated strategic and and Floyd, 1990). Indeed, as Floyd and Wooldridge (1992,
organizational variables. Overall, this study attempts to p. 155) noted, ‘‘Implementation of top management’s strat-
answer the following research question: In what context egy is often considered the key strategic role of middle-level
does gender diversity in management impact firm perform- managers.’’ Firms’ strategic performance thus relies not
ance? In doing so, we focus on two key contextual varia- only on top managers but the entire pool of managers and
bles—growth orientation and organizational culture. officials (Floyd and Wooldridge, 1994, 1997).
In sum, it may be that the benefits of diversity in general,
and gender diversity in particular, are more fully realized
2. Conceptual background and hypotheses when the breadth of perspectives imbued in a diverse
management group benefit not only the TMT-led strategic
2.1. Gender diversity in managerial ranks decision-making process but also the strategic implementa-
tion phase in which middle managers are primarily
According to upper-echelons theory, senior-level man- involved. We address this potentially significant gap in the
agers (i.e., the TMT) have a critical impact on firm literature by integrating upper echelons and middle man-
performance given the significant organizational decisions agement perspectives as they relate to gender diversity and
they are empowered to make (Finkelstein and Hambrick, firm performance.
1996). However, research focused at the TMT level of
analysis has reported conflicting results with regard to the 2.2. Contingency approach
relationship between diversity and performance. Some
researchers report a negative relationship between TMT Some researchers investigating the impact of diversity on
heterogeneity and firm performance. They conclude that performance have taken a contingency approach by exam-
diverse TMTs are more difficult and costly to coordinate ining factors that interact with diversity (Neale et al., 1999;
and control than homogeneous teams and that the added Richard and Johnson, 1999). Jackson (1992) proposes that
costs impede performance (Ancona and Caldwell, 1992; heterogeneity is beneficial for unstructured, novel tasks but
Murray, 1989). Other studies, however, have found a not for routine tasks. Richard and Johnson (1999) posit that
positive association between TMT diversity and perform- diversity’s beneficial effects are more likely to be realized
ance, indicating that diversity may enhance the breath of when firms’ business strategies and organizational cultures
perspectives, cognitive resources, and overall problem- are compatible. Richard (2000) found that firm-level racial
solving capacity of the team (Bantel and Jackson, 1989; diversity benefited firms pursuing growth strategies but was
Hambrick et al., 1996; Smith et al., 1994). detrimental to downsizing firms. The breadth of perspec-
In an attempt to explain these contradictory results, we tives and creativity provided by a diverse work force thus
explore whether diversity’s influence on firm performance is appear favorable in certain contexts. Similarly, in the appro-
more fully realized when the management group’s level of priate setting, gender diversity in the management ranks
analysis is broadened to include middle managers and other may contribute to performance gains (Williams and
midlevel officials and supervisors along with TMT mem- O’Reilly, 1998). Next, we explore the role that firm growth
bers. Past TMT studies have generally not taken into and organizational culture may play in the gender diversity–
account the perspective that the strategy process involves performance relationship.
multiple levels of management (Burgelman, 1994; Floyd
and Wooldridge, 1992). For example, Kraut and Pedigo 2.2.1. Growth orientation
(1986) found that while top-level managers monitored the Organizations pursuing growth seek to gain economies
environment, middle managers coordinated departments of scope by spreading risks over a broader base and gaining
and/or groups, and lower-level managers supervised sub- the benefits of skill transfer such as technical expertise or
ordinates. Furthermore, although top-level managers are managerial know-how (Pearce, 1982). Growth is thus best
responsible for defining the overall strategic context, middle pursued when a firm has appropriate and sufficient human
management’s contribution to the strategy process is also capital to manage an expanding organization. Cultural
crucial because they are often first to identify strategic diversity can provide the breadth of experience and know-
problems and opportunities (Burgelman, 1994; Wooldridge ledge (McLeod et al., 1996) that is likely to benefit a firm
and Floyd, 1990), serving as an important source of strategic pursuing such growth. A growth-oriented, culturally diverse
alternatives (Burgelman, 1994; Van Der Velde et al., 1999). organization benefits from employees who are flexible in
In addition, TMT strategic initiatives and directives are their thinking and who are less likely to be concerned about
ineffectual without lower-level managers directing their departing from the norm (Richard, 2000; Schuler and
staffs’ activities to implement the strategy (Burgelman, Jackson, 1987). Furthermore, if an organization overcomes
 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019 1011

resistance to change with regard to accepting diversity, it 2.2.2. Organizational culture

should be well positioned to handle other types of change as An organization’s culture can be viewed as a pattern of
well (Iles and Hayers, 1997), including expansion. shared values and beliefs that help its members understand
Growth can be achieved by selling current products or organizational functioning and provides them norms for
services in new markets through geographic expansion and/ behavior in the organization. Organizational culture, by
or through mergers and acquisitions (Suresh and Orna, providing a framework through which employees intern-
1989). Growth through geographic expansion requires cre- alize expectations about corporate roles and behaviors,
ativity, knowledge, and a variety of experience and insights serves to a large extent as an organizational control
to design the appropriate expansion path. Growth through mechanism (Deshpandé and Webster, 1989). It is in this
mergers and acquisition also requires these assets in order to manner that organizational culture may influence diversi-
effectively integrate organizational activities on a post- ty’s impact on firm performance. For example, Williams
acquisition basis. In either case, gender diversity can offer and O’Reilly (1998), in their extensive literature review on
market-related advantages. As companies enter new markets the effects of diversity, suggest that organizational culture
or acquire firms in new markets, women, particularly at the may be an effective way for managers to encourage
managerial level, may bring added knowledge, experience, solidarity.
and flexibility as well as cultural insight, understanding, and A widely accepted conceptualization of organizational
sensitivity critical to serving the needs of new market culture is the Competing Values Framework based on the
segments (Cox, 1994). work of Quinn and his colleagues (see Cameron and
In the banking industry, the setting for the current study, Quinn, 1999, for references). This framework assumes that
rapid growth, particularly through mergers and acquisitions, organization’s cultures can be differentiated by their dom-
has resulted in less efficient operations that, for many firms, inant organizational attributes, bonding mechanisms, lead-
has negatively affected performance (Bird, 1991; Hopkins ership styles, and overall strategic emphases (Cameron and
and Hopkins, 1997). Firms seeking growth must draw upon Freeman, 1991; Cameron and Quinn, 1999; Deshpandé et
additional resources to compensate for initial operational al., 1993). The framework is operationalized across two
inefficiencies. Human capital, including the array of skills, dimensions, the first of which is formal –informal organ-
knowledge, and abilities of its employees, represents a izational processes. The extremes of this continuum reflect
resource that growth-oriented firms can potentially exploit. the competing demands of flexibility and spontaneity
In particular, the creativity and flexibility provided by versus stability, control, and order. The second dimension
diversity may be beneficial to firms pursuing growth reflects the conflicting demands of the internal organiza-
(Richard, 2000). tion and the external environment. One end of this
In contrast, a firm pursuing a downsizing strategy seeks continuum represents a focus on internal integration,
operational efficiencies. Although diversity in human organizational processes, and structural stability and con-
resources may contribute to, among other things, the quality trol; the other end is anchored by an emphasis on com-
of ideas, it can also create additional costs stemming from petition, interaction with the environment, and a focus on
the need for increased coordination and control required for outcomes. The resulting four culture types—clan, hier-
a more heterogeneous group (Jehn, 1995). The additional archy, market, and adhocracy—represent aspects of firms’
costs associated with diversity may be a potentially sig- underlying assumptions about motivation, leadership, and
nificant disadvantage for downsizing firms. For example, effectiveness. Although a firm’s culture will be composed
Richard (2000) found that the relationship between firm- of values found in each of the four culture types, a
level racial diversity and firm performance was moderated dominant culture type will typically emerge to form an
by a growth strategy. Racially diverse firms that pursued identifiable corporate culture (Cameron and Freeman,
growth experienced performance gains while those pursuing 1991; Cameron and Quinn, 1999). A brief description of
a downsizing strategy experienced losses. Gender diversity each culture type follows (Cameron and Quinn, 1999;
has been posited to have similar effects as other visible Denison and Spreitzer, 1991; Zammuto and Krakower,
diversity attributes such as race (Pelled, 1996) and thus its 1991).
relationship with firm performance should be similarly
influenced by growth. Furthermore, it is plausible that this Clan culture type: The clan culture type is internally
relationship will intensify at the management level where oriented, emphasizes informal governance, and is in
critical firm-wide decisions are both formulated and imple- general a friendly place to work. The organization is
mented. As such, the following hypothesis is offered: held together through employee loyalty, morale, and
commitment. The development of human resources
Hypothesis 1: Gender diversity in management and a firm’s and employee participation in decision-making are
growth orientation will influence firm performance through highly valued. Emphasis is placed on teamwork and
their interaction effect. Gender diverse firms pursuing high cohesiveness.
growth will have higher performance compared to gender Adhocracy culture type: The adhocracy culture type
diverse firms pursuing low growth. combines informal governance with an external
1012 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019

orientation. It is a dynamic and creative place to Zammuto and Krakower, 1991) would serve to further
work where firm members take risks. Individual support a diverse work group. As such, the following
initiative and spontaneity are valued. Individuals are hypothesis is offered:
motivated by the ideological appeal of growth,
flexibility, and variety. Effectiveness criteria revolve Hypothesis 2: Gender diversity in management and a clan
around new market development and resource organizational culture type emphasis will influence firm
acquisition. performance through their interaction effect. Gender diverse
Hierarchy culture type: The hierarchy culture type adopts firms with a clan emphasis will have higher performance
an internal orientation and an emphasis on mecha- compared to gender diverse firms with little or no clan
nistic governance such that formal rules and policies emphasis.
are closely followed. It focuses on stability and
Like the clan culture type, the adhocracy culture type is
smooth-running operations. Firm members are moti-
broadly characterized by its flexibility, spontaneity, and
vated by security and rewards for accomplishments.
individualism. Such traits are likely to provide a context
Effectiveness is defined by permanence and achiev-
in which a diversity of opinions and ideas can thrive. With
ing clearly defined goals.
dominant attributes of entrepreneurship, creativity, and
Market culture type: The market culture type has a
adaptability, the adhocracy culture type is likely to more
formal governance structure and an external orienta-
fully utilize the wide range of perspectives found in a
tion. Members are goal-oriented and concerned with
diverse work group and, as a result, more fully elicit its
getting the job done. This culture is achievement-
potential benefits. In addition, its focus on innovation,
focused and emphasizes planning, performance, and
variety, and growth would provide a setting in which a
efficiency. Individuals are motivated by competition
creative work group would prosper. As such, the following
and in achieving market share.
hypothesis is offered:
Several studies indicate the potential influence that Hypothesis 3: Gender diversity in management and an
firms’ organizational culture may have on the diversity – adhocracy organizational culture type emphasis will influ-
performance relationship. Gaertner et al. (1990) note that ence firm performance through their interaction effect.
when cooperation and teamwork are introduced to the larger Gender diverse firms with an adhocracy emphasis will have
group, intergroup bias across dimensions such as race and higher performance compared to gender diverse firms with
gender can be reduced. In fact, Chatman and Barsade little or no adhocracy emphasis.
(1995) discuss how organizational cultures may be created
to foster cooperation and commitment. Using a business The hierarchy and market culture types’ influence on
simulation, Chatman et al. (1997) found that organizational gender diversity is less clear. The hierarchy culture type,
culture moderated the effects of diversity such that conflict like the clan, has an internal orientation with concern
arising from group heterogeneity was seen as more bene- placed on stability and employee job security. Such
ficial for groups with a collectivist organizational culture. emphases should benefit a diverse work group. Its gov-
More recently, O’Reilly et al. (1997) found that an organ- ernance system, however, is highly regulated and mecha-
izational culture supporting racial diversity improved firm nized where rules and procedure govern employee duties.
performance. Such a context is, conversely, likely to dampen diversity’s
These studies suggest that a diverse workforce will have creative elements.
greater potential to thrive in certain organizational cultures The market culture type’s external orientation that values
than in others. For example, work force diversity has been differentiation and achieving market share through cus-
found to provide varied opinions that have the potential to tomer-focused segmentation may complement the talents
improve the quality of decisions (McLeod et al., 1996). of a diversified work force that closely matches its cus-
There is also evidence that diversity has the potential to tomer base. However, this culture also stresses a formal
introduce creativity and problem-solving abilities (Cox, control system driven by planning and structure with an
1994; Cox and Blake, 1991). Thus, a diverse work force overarching goal of market dominance. Such an envir-
will likely be more effective in informal organizational onment is one in which diversity’s benefits are less likely
cultures characterized by flexibility, spontaneity, and indi- to surface and thus one in which a diversified work group
viduality—general characteristics of the clan and adhocracy is less likely to realize its full potential.
culture types. Thus, the relationships between the hierarchy and market
It seems reasonable that the clan culture type’s dom- culture types and diversity are uncertain. This ambiguity
inant attributes of cohesiveness, participation, and team- results from the market and hierarchy culture types’
work would provide an environment conducive to a emphasis on various values, ideals, and practices that can
diverse work group. Its employee-focused orientation with be expected to positively and negatively influence the impact
an emphasis on satisfaction and consensual problem-solv- of diversity on performance. As such, no hypotheses are
ing and decision-making (Denison and Spreitzer, 1991; offered in this regard.
 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019 1013

2.3. Configurational theory a separate study 6 months earlier. The respective bank
presidents and senior human resource executives had
Organizational theorists argue that multivariate configu- completed surveys about their firms’ organizational cul-
rations may offer more useful and complete explanations of ture and performance as well as key firm background
organizational phenomenon than those provided by simple information. For validation purposes, we compared the
contingency approaches using bivariate interactions (Baker presidents’ responses to those of the senior human
and Cullen, 1993; Dess et al., 1993; Miller, 1986). Config- resource managers and found an acceptable level of
urational theory departs from contingency theory by taking a interrater reliability ( P < .001). A second questionnaire
more holistic stance, positing that parts of a social entity take was sent to the senior human resource manager of the
their meaning from the whole and cannot truly be understood 535 banks requesting work force demography information
in isolation (Doty and Glick, 1994). Configurational that most banks must report annually per federal law. A
approaches take into account past advances of contingency total of 177 usable questionnaires were received, resulting
studies, synthesize them, and ground them in multivariate in a 33% response rate. Logistic regression found no
descriptions (Meyer et al., 1993). These higher-order inter- evidence of nonresponse bias between responding and
actions cannot be captured by traditional bivariate approaches nonrespondent banks across key publicly available varia-
(Delery and Doty, 1996). In fact, configurations of key bles (bank age, number of branches, holding company
variables have been found to have more predictive power affiliation, asset size, and return on equity) (Delery and
than bivariate contingency relationships (Dess et al., 1993). Doty, 1996).
For example, Baker and Cullen (1993) found that different
configurations of firm age, size, and growth affect structural 3.2. Independent variables
change. They conclude that the effects of these variables
cannot be understood alone or in additive combinations. 3.2.1. Level of gender diversity
Similarly, Ostroff and Schmitt (1993) determined that various The banks provided gender information from their
configurations of organizational processes and external Federal EEO-1 Standard Form 100 in which the gender
resources are related to organizational effectiveness. and race of employees is reported across nine hierarchical
Following configurational theory reasoning, we suggest job categories: officials and managers, professionals,
that greater order and understanding emerge from the inter- technicians, sales workers, office and clerical, craft work-
action of the parts—gender diversity, growth orientation, and ers, operatives, laborers, and service workers. For this
organizational culture. We posit that firms pursuing growth study, the officials and managers category representing
objectives will do so more effectively in a setting character- all levels of management was used. This broad job
ized by flexibility in organizational processes and a focus on category includes administrative and managerial personnel
the external environment. These characterizations correspond who set broad policies, exercise overall responsibility for
closely to the adhocracy culture type that combines an the execution of these policies, and direct individual
informal governance structure with an external orientation. departments or special phases of a firm’s operations
In addition to its supportive context for a diverse work group, (Equal Employment Opportunity Commission, 1997).
in particular for individual initiative, variety, and spontaneity, The mean number of managers was 32, reflecting this
the adhocracy culture type’s dominant attributes of entre- broader level of analysis.
preneurship, creativity, and adaptability position it well to Consistent with previous operationalizations of categor-
match an organization intent on growth and expansion. ical data (Bantel and Jackson, 1989; Hambrick et al., 1996;
Indeed, adhocracy culture types largely define their effective- Murray, 1989), Blau’s (1977) index of heterogeneity was
ness based on market development and resource acquisition used to assess the level of homogeneity – heterogeneity. The
(Denison and Spreitzer, 1991). As such, the following hypo- index was calculated as follows:
thesis is offered:
X
Hypothesis 4: The configuration of gender diversity in ð1  Pi2 Þ
management, a growth orientation, and an adhocracy culture
type emphasis will positively influence firm performance where P is the proportion of group members in a category
through its interaction effect. and i is the number of different categories represented in the
firm.

3. Methodology 3.2.2. Growth

Each bank’s record of growth was used to operation-
3.1. Sample alize growth orientation. This was done by averaging the
banks’ asset growth for fiscal years 1996 and 1998 (as
The sample frame for this study consisted of 535 reported on their FDIC financial statements), an operation-
banks that had responded to an earlier questionnaire for alization consistent with recent bank studies’ measurement
1014 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019

of growth (e.g., Richard, 2000). The 3-year period was size, operationalized as the logarithmic transformation of
employed to help smooth year-to-year fluctuations and to the total dollar value of assets (Osterman, 1995), was also
provide a broader horizon over which the firms’ growth controlled because it has been found to have a direct effect
orientation could be inferred. Such a growth measure was on financial performance based on economies of scale and
appropriate for the objectives of this study in that a bank’s market power (Shepherd, 1975). It is an established means
growth in assets over a 3-year period should reflect its of accounting for differences in organizational outcomes
strategic intent toward growth as well as its success in its that has been used in past bank studies (e.g., Hopkins and
implementation. Asset growth for 1996 represents the Hopkins, 1997; Richard, 2000).
percentage growth in total assets from year-end 1995 to Also employed in previous bank studies (e.g., Delery
year-end 1996, and asset growth for 1998 covers year-end and Doty, 1996; Zimmerman, 1996) and controlled for
1997 to year-end 1998. Positive percentages of this indic- was the banks’ holding company affiliation (1 = holding
ator reflect a growth orientation, while negative percen- company; 0 = other). Banks that are part of a holding
tages reflect a downsizing orientation. company are likely to have access to considerably greater
resources that enhance their ability to compete in the
3.2.3. Organizational culture marketplace (Zimmerman, 1996). Interaction variables of
Organizational culture was measured using the Com- Hierarchy culture  Gender and Market culture  Gender
peting Values Scale developed by Quinn and his col- were also introduced as controls to account for the
leagues (cf. Cameron and Quinn, 1999). The scale indeterminate effects that these culture types and their
identifies the relative influence of four culture types that interaction with gender might have and to better isolate
comprise a firm’s organizational culture. The scale used the hypothesized effects of the clan and adhocracy culture
Likert-type items scored on a seven-point scale ranging types.
from 1 = strongly disagree to 7 = strongly agree. This scale
has been widely used in past research and acceptable 3.4. Dependent variables
levels of reliability and validity have been reported across
numerous studies (Cameron and Quinn, 1999). The clan Two measures of firm performance were employed.
culture type (a=.71) and the adhocracy culture type Employee productivity, an intermediate-level measure of
(a=.77), based on bank presidents’ responses, served as output, is an important performance criterion in service
independent variables. organizations such as banks because of their relatively high
labor costs (Mehra, 1996). Productivity is calculated as the
3.3. Control variables logarithmic transformation of net income per employee for
year-end 1998 (Richard, 2000). The second measure,
Management group size, operationalized as the number return on equity, represents the central measure of the
of officials and managers, was used as a control in strength of a financial institution (Earle and Mendelson,
consonance with previous demography research that has 1991). The average return on equity for years-end 1996–
found group size to influence group dynamics and per- 1998 as reported to the FDIC was employed to determine
formance (Hambrick et al., 1996; Pelled et al., 1999). Firm the return on equity.

Table 1
Means, standard deviations, and correlations
1 2 3 4 5 6 7 8 9 10 11
Mean 32.24 11.70 0.87 2.00 1.84 0.41 0.21 5.63 4.62 3.24 13.00
S.D. 66.44 1.26 0.34 0.84 0.84 0.16 0.41 0.83 0.98 0.64 6.27
1 Management size –
2 Firm size .63* * –
3 Holding company .06 .11 –
4 Hierarchy  Gender .14 .26* * .03 –
5 Market  Gender .17 * .35* * .12 .86* * –
6 Gender diversity .20 * .35* * .03 .89* * .88* * –
7 Growth orientation .16 .24* * .00  .02 .05 .05 –
8 Clan culture  .05  .05  .018 * .15 .03 .05 .02 –
9 Adhocracy culture .01 .15 .24* * .09 .23* * .04 .07 .16 * –
10 Productivity .11 .28* * .13 .10 .17 * .13 .13  .07 .12 –
11 Return to equity .18 * .30* * .18 * .03 .13 .08 .04  .04 .13 .58* * –
* p < .05.
* * p < .01.
 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019 1015

4. Results

Table 1 provides the means, standard deviations, and

correlations of the study variables. Hypotheses 1 – 3 state
that gender diversity will influence firm performance
through a positive interaction effect with firms’ growth
orientation (Hypothesis 1), clan culture type (Hypothesis
2), and adhocracy culture type (Hypothesis 3), respectively.
Table 2 displays the hierarchical regression results for these
hypotheses. Analyses were completed for both the produc-
tivity and return on equity measures of firm performance.
Hypotheses 1 – 3 were tested simultaneously. Control
variables were entered into the analysis first, followed by
the main effect variable, gender diversity, and then by the
growth, clan culture type, and adhocracy culture type varia-
bles. The interaction variables of Gender diversity  Growth
(Hypothesis 1), Gender diversity  Clan (Hypothesis 2), and
Gender diversity  Adhocracy (Hypothesis 3) were then
entered.
The interaction between gender diversity and growth was
significant for productivity (b = 2.17, P=.000). Fig. 1 depicts
this interaction after splitting the sample at the median for
the growth and gender variables. Firms with high levels of Fig. 1. Gender diversity – growth interaction.
gender diversity and growth had the highest productivity
(x = 3.37). For return on equity, though, the interaction was A significant association was found between the Gender
significant at only a level of P < .10 (b = 0.87, P=.064). diversity  Clan interaction and productivity (b = 1.18,
Thus, partial support was found for Hypothesis 1. P=.041), though not for return on equity (b = 0.66,

Table 2
Results of hierarchical regression analysis: productivity and return on equity
Hypothesis Variable Productivity Return on equity
tested Two-way Three-way Two-way Three-way
interaction model interaction model interaction model interaction model
Step 1 DR2=.112 * * DR2=.112 * * DR2=.142* * * DR2=.142* * *
Management group size  0.12  0.12  0.038  0.038
Firm size 0.34 * * 0.34 * * 0.32 * * 0.32 * *
Holding company 0.09 0.09 0.12 0.12
Hierarchy culture  Gender  0.10  0.10  0.23  0.23
Market culture  Gender 0.14 0.14 0.21 0.21
Step 2 DR2=.001 DR2=.001 DR2=.000 DR2=.000
Gender diversity  0.07  0.07  0.01  0.01
Step 3 DR2=.009 DR2=.009 DR2=.002 DR2=.002
Growth 0.08 0.08  0.05  0.05
Clan culture  0.07  0.07 0.00 0.00
Adhocracy culture 0.03 0.03 0.02 0.02
Step 4—two-way interaction DR2=.113* * * DR2=.038y
Hypothesis 1 Gender diversity  Growtha 2.17* * * – 0.87y –
Hypothesis 2 Gender diversity  Clana 1.18 * – 0.66 –
Hypothesis 3 Gender diversity  Adhocracya  0.85y –  1.54 * –
Step 4—three-way interaction DR2=.115* * * DR2=.003
Gender diversity  Growth  Clana –  0.58 – 0.18
Hypothesis 4 Gender diversity  Growth  Adhocracya – 2.06 * * – 0.06

F statistic 3.50* * * 3.90* * * 2.52 * * 2.15 *

Total R2 (total adjusted R2) 0.235(0.168) 0.237(0.176) 0.182(0.110) 0.147(0.079)
a
One-tailed test.
* P < .05.
** P < .01.
*** P < .001.
y
P < .10.
1016 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019

diversity’s impact on performance at the TMT level have

reported mixed and often contradictory results, suggesting
the critical influence of other contextual variables on this
relationship. Adopting contingency and configurational
approaches, this article posits that a broader set of organ-
izational variables interacts with diversity to influence firm
outcomes. Two such variables, firms’ growth orientation
and organizational culture, were explored in this regard
with respect to employee productivity and return on equity
outcomes.
The management group examined in this study was
composed of all firm members considered to be managers
and officials. The research focus is thus set at a broader
level of analysis than past TMT research that primarily
focused on groups composed of senior-level managers.
Diversity’s positive influences in general, and gender diver-
sity’s in particular, may be more fully realized when
examined not only across senior-level executives who
Fig. 2. Gender diversity – clan interaction.
formulate strategy but also across midlevel managers who
implement strategy.
P=.173), providing only partial support for Hypothesis 2. This study’s results suggest that the relationship between
Fig. 2 depicts the interaction of gender diversity and clan management group demography and organizational out-
culture type after splitting the sample at the median for the comes may be more complex than scholars have previously
clan and gender variables. Firms with high levels of gender realized. First, as hypothesized, a positive association was
diversity and a strong emphasis on the clan culture type had found between the gender diversity – growth orientation
the highest productivity (x = 3.33) among the groups. interaction and performance. The potential benefits of
The test of Hypothesis 3 found a significant, but contrary gender diversity such as novel insights, perspectives, cre-
to that hypothesized, negative association between the ativity, and experience appear to foster and support expan-
Gender diversity  Adhocracy variable and return on equity sion into new markets. These results support and extend
(b =  1.54, P=.011) and productivity (b =  0.85, P=.091). previous diversity research that found a similar association
Thus, Hypothesis 3 was not supported. between racial diversity, growth, and firm performance
Hypothesis 4 stated that firms with higher levels of gender (Richard, 2000).
diversity in management, an orientation toward growth, and Additionally, this is the first field study of which we are
an adhocracy culture type emphasis will be associated with aware to find a link between diversity and organizational
higher levels of performance than will firms with lower culture with respect to performance. More specifically, some
levels of these variables. Step 4 in the three-way interaction support was found for the interaction between gender
model shows the results of the analysis for the configuration diversity and the clan culture type in relation to firm
variable (Gender diversity  Growth  Adhocracy). Gender performance. The clan’s core values of teamwork, participa-
diversity in conjunction with an adhocracy culture type and a tion, and cohesiveness appear to provide an environment
growth orientation was found to be significantly associated that fosters the varied skills and abilities of a diverse work
with productivity (b = 2.06, P=.005), though not for return on force. This collectivist setting may be the prescription
equity (b = 0.06, P=.478). Thus, Hypothesis 4 received necessary to prevent the formation of in-groups, polariza-
partial support. tion, and discrimination predicted by social identity theory
In summary, hypotheses testing gender diversity’s inter- and observed in the work place. Once the potentially
action with key organizational variables and their sub- detrimental effects of diversity are suppressed by a support-
sequent influence on firm-level outcome measures were ive context, the beneficial aspects of diversity may prevail
tested. While no support was found for Hypothesis 3, partial and contribute to firm success.
support was found for Hypotheses 1, 2, and 4. Interesting and potentially revealing is the fact that
gender diversity did not provide benefits to firms with an
adhocracy culture type as hypothesized. Specifically, the
5. Discussion interaction of gender with the adhocracy culture type was
found to be significantly and negatively associated with the
5.1. Implications return on equity performance measure. Like the clan culture
type, the adhocracy culture type emphasizes informal organ-
This study sought to develop a deeper understanding of izational processes such as flexibility, individuality, and
the effects of gender diversity in management. Studies of spontaneity. What distinguishes the two culture types,
 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019 1017

though, is their orientation to the conflicting demands of the a positive effect that the bivariate contingency prediction
internal and external environments. The clan culture type did not detect. This underscores the potential importance of
focuses inward on the firm and employees, placing high taking a configurational approach to studying diversity’s
value on integration, teamwork, and other internal pro- effects within the firm. The results suggest that gender
cesses. The adhocracy culture type, in contrast, has an diversity’s impact in management is conditional upon its fit
external focus that places priority on competition and within the overall organizational context.
interaction with the environment. Thus, the adhocracy’s
results-oriented external emphasis and relatively lower level 5.2. Contribution of study
of employee-focused orientation may tend to overshadow its
informal processes in its interaction with gender diversity. This study offers several potential contributions to the
This may result in a relatively less supportive environment management literature. First, by broadening the scope of
that potentially dampens an atmosphere of open-minded management under study, we extend previous TMT research
consideration necessary to fully elicit gender diversity’s to include managers who not only formulate strategy but
benefits such as sensitivity and breadth of perspectives. also those who implement it. Our results indicate that the
Furthermore, employees in adhocracy cultures are moti- potential benefits of gender diversity may be realized across
vated by competition and individual initiative. However, all management levels and are not constrained to any one
studies have found that women are more cooperative than echelon of management.
men, encourage participation, prefer to lead through inclu- Next, the results indicate that gender diversity’s impact
sion, and avoid competition. Men, conversely, embrace on firm performance is dependent on the organizational
competitive individualism (Eagly et al., 1992; Grant, context in which it resides. The findings suggest that
1998). These contrasting styles may tend to hinder perform- gender diversity in management should enhance perform-
ance, particularly within an adhocracy culture. ance for firms seeking growth. Similarly, for firms seeking
The contingency results suggest that the potentially to diversify their managerial ranks, the results imply that
beneficial aspects of gender diversity appear to require a the appropriate culture should be in place to fully realize
more nurturing, employee-focused setting before they can gender diversity’s benefits. Together, these results provide
be fully realized. Firms planning to increase their current a possible explanation for previously conflicting diversity
level of gender diversity may be advised to cultivate a set of findings and offer direction for ongoing research in this
organizational values that focus closely on employee affili- field.
ation and consensus-building, paternal leadership styles, and Finally, we found that examining gender diversity in
strategic emphases that are internally oriented and conjunction with only one contingent variable may be
informal—all aspects of the clan culture type. Organizations insufficient to accurately reflect its influence. Instead, we
that fall short in these areas may fail to fully realize the provide evidence that more complex, configurational
benefits that a gender diverse management group offers. approaches may need to be taken with regard to assessing
These findings may help reconcile conflicting results of past the effects of gender diversity in management.
diversity – performance research by suggesting that the
influence of diversity on organizational outcomes is depend- 5.3. Limitations and future research
ent on the cultural context in which it resides.
Finally, past diversity research has largely examined This study extends previous cultural diversity research
diversity’s effects on organizational outcomes in isolation but is not without limitations. In light of much group
or, at most, relative to one additional variable. The final demography research that has been conducted in labor-
relationship explored in this study examined gender diver- atory-like settings that limit external validity, the current
sity’s effects in a context that more closely simulates its study is noteworthy in its examination of diversity issues in
impact in a real-world setting. Guided by configurational a workplace setting. However, it was set in a single industry
theory, the interaction of diversity and organizational and, as such, the generalization of results to other settings
growth, set together in an adhocracy culture type, was should be exercised with caution.
investigated. Gender diversity’s interaction with an adhoc- Although relationships between various interaction var-
racy culture type and a growth orientation was found to be iables and firm performance were found in this study, the
positively associated with performance as measured by firm cross-sectional nature of the data dictates that no causal
productivity. Thus, a gender-diverse management group relationships should be inferred. This implies the need for
appears to provide benefits to a growth-oriented firm in a longitudinal research designs that track organizational
culture that values innovation, flexibility, and interaction efforts and experiences over time. Such research should
with the environment. encourage and monitor active management interventions
That the adhocracy culture type’s interaction with gen- and/or experiments to enhance this understanding.
der diversity was negatively associated with performance Additionally, this study should be extended to include
without considering growth orientation renders this result other visible aspects of cultural diversity such as race and
even more profound. The higher-order interaction identified age. Finally, diversity was examined only at the manage-
1018 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019

ment group level. Similar focus should be placed at firm- Equal Employment Opportunity Standard Form 100—Employer Informa-
tion Report EEO-1. Washington (DC): Equal Employment Opportunity
wide and industry levels of analysis.
Commission, 1997. p. 1 – 7 [March].
Finkelstein S, Hambrick DC. Strategic leadership: top executives and their
effects on organizations. St. Paul (MN): West, 1996.
Acknowledgements Floyd SW, Wooldridge B. Middle management involvement in strategy and
its association with strategic type. Strategy Manage J 1992;13:153 – 67.
Floyd SW, Wooldridge B. Dinosaurs or dynamos? Recognizing middle man-
The authors thank the anonymous reviewers who agement’s strategic role. Acad Manage Exec. 1994;47 – 57 (November).
provided insightful guidance in helping us more fully Floyd SW, Woolridge B. Middle management’s strategic influence and
develop and refine this article. organizational performance. J Manage Stud 1997;34:465 – 85.
Gaertner S, Mann J, Dovidio J, Murrell A, Pomare M. How does cooper-
ation reduce intergroup bias? J Pers Soc Psychol 1990;59:692 – 704.
Grant J. Women as managers: what they can offer to organizations. Organ
References Dyn. 1998;56 – 63 (Winter).
Hambrick DC, Cho TS, Chen M. The influence of top management team
Ancona DG, Caldwell D. Demographic and design: predictors of new heterogeneity on firms’ competitive moves. Adm Sci Q 1996;41:
product team performance. Organ Sci 1992;3:321 – 41. 659 – 84.
Baker DD, Cullen JB. Administrative reorganization and configurational Hopkins WE, Hopkins SA. Strategic planning – financial performance rela-
context. Acad Manage J 1993;36:1251 – 78. tionships in banks: a causal examination. Strategy Manage J 1997;18:
Bantel KA, Jackson SE. Top management and innovations in banking. 635 – 52.
Strategy Manage J 1989;10:107 – 29. Iles P, Hayers PK. Managing diversity in transnational project teams:
Bird A. 1990s twist on strategic planning. Bankers Mag. 1991;66 – 9 a tentative model and case study. J Managerial Psychol 1997;12:
(March/April). 95 – 117.
Blau PM. Inequality and heterogeneity. New York: Free Press, 1977. Jackson S. Consequences of group composition for the interpersonal
Burgelman RA. Fading memories: a process theory of strategic business dynamics of strategic issue processing. Adv Strategic Manage 1992;8:
exit in dynamic environments. Adm Sci Q 1994;39:24 – 56. 345 – 82.
Cameron KS, Freeman SJ. Cultural congruence, strength and type: relation- Jehn K. A multimethod examination of the benefits and detriments of intra-
ships to effectiveness. In: Woodman RW, Passmore WA, editors. Re- group conflict. Adm Sci Q 1995;40:256 – 82.
search in organizational change and development. Greenwich (CT): JAI Kanter RM. The middle manager as innovator. Harv Bus Rev 1982;604(4):
Press, 1991. p. 23 – 58. 95 – 106.
Cameron KS, Quinn RE. Diagnosing and changing organizational culture. Kraut A, Pedigo P. The role of the manager: what’s really important in
New York: Addison-Wesley, 1999. different management jobs. Acad Manage Exec 1986;3:286 – 93.
Chatman J, Barsade S. Personality, organizational culture, and cooperation: McLeod PL, Lobel SA, Cox TH. Ethnic diversity and creativity in small
evidence from a Business Simulation. Adm Sci Q 1995;40:423 – 43. groups. Small Group Res 1996;27:246 – 64.
Chatman J, Polzer J, Barsade S, Neale M. The influence of demographic Mehra A. Resource and market based determinants of performance in the
composition and organizational culture on work processes and out- US banking industry. Strategy Manage J 1996;17:307 – 22.
comes in a business simulation. Working Paper, University of Califor- Meyer AD, Tsui AS, Hinings CR. Guest co-editors’ introduction: config-
nia, Berkeley, 1997. urational approaches to organizational analysis. Acad Manage J 1993;
Cox T. Cultural diversity in organizations: theory, research, and practice. 36:1175 – 95.
San Francisco (CA): Berrett-Koehler, 1994. Miller D. Configurations of strategy and structure: toward a synthesis.
Cox T, Blake S. Managing cultural diversity: implications for organiza- Strategy Manage J 1986;7:233 – 49.
tional competitiveness. Acad Manage Exec 1991;5:45 – 56. Murray A. Top management group heterogeneity and firm performance.
Delery JE, Doty DH. Theoretical frameworks in strategic human resource Strategy Manage J 1989;10:125 – 41.
management: universalistic, contingency, and configurational perspec- Neale MA, Northcraft GB, Jehn KA. Exploring Pandora’s box: the impact
tives. Acad Manage J 1996;39:802 – 35. of diversity and conflict on work group performance. Perform Improv Q
Denison DR, Spreitzer GM. Organizational culture and organizational de- 1999;12:113 – 26.
velopment: a competing values approach. In: Woodman RW, Passmore Osterman P. Work/family programs and the employment relationship. Adm
WA, editors. Research in organizational change and development. Sci Q 1995;40:683 – 98.
Greenwich (CT): JAI Press, 1991. p. 1 – 21. O’Reilly C, Williams K, Barsade S. Demography and group performance:
Deshpandé R, Webster FE. Organizational culture and marketing: defining does diversity help? Presented at the annual meetings of the Academy
the research agenda. J Mark 1989;53:3 – 15. of Management, Boston (MA), 1997.
Deshpandé R, Farley JU, Webster FE. Corporate culture, customer orienta- Ostroff C, Schmitt N. Configurations of organizational effectiveness and
tion, and innovativeness in Japanese firms: a quadrad analysis. J Mark efficiency. Acad Manage J 1993;36:1345 – 61.
1993;57:23 – 37. Pearce JA. Selecting among alternative grand strategies. Calif Manage Rev
Dess G, Newport S, Rasheed A. Configuration research in strategic man- 1982;26:87 – 110.
agement: key issues and suggestions. J Manage 1993;19:775 – 95. Pelled LH. Demographic diversity, conflict, and work group outcomes: an
Doty DH, Glick WH. Typologies as a unique form of theory building: intervening process theory. Organ Sci 1996;7:615 – 31.
toward improved understanding and modeling. Acad Manage Rev Pelled LH, Eisenhardt KM, Xin KR. Exploring the black box: an analysis
1994;19:230 – 51. of work group diversity, conflict, and performance. Adm Sci Q 1999;
Eagly AH, Makhijani MG, Klonsky BG. Gender and the evaluation of 44:1 – 28.
leaders: a meta-analysis. Psychol Bull. 1992;3 – 22 (January). Richard OC. Racial diversity, business strategy, and firm performance: a
Earle DM, Mendelson M. The critical mesh in strategic planning. Bankers resource-based view. Acad Manage J 2000;43(2):164 – 77.
Mag 1991;174:48 – 53. Richard OC, Johnson NB. Making the connection between formal human
Elsass PM, Graves LM. Demographic diversity in decision-making groups: resource diversity practices and organizational effectiveness: beyond
the experiences of women and people of color. Acad Manage Rev management fashion. Perform Improv Q 1999;12:77 – 96.
1997;22:946 – 73. Schilit WK. An examination of the influence of middle-level managers in
 S. Dwyer et al. / Journal of Business Research 56 (2003) 1009–1019 1019

formulating and implementing strategic decisions. J Manage Stud 1987; Williams K, O’Reilly C. Demography and diversity in organizations: a
24:271 – 93. review of 40 years of research. In: Staw BM, Cummings LL, editors.
Schuler RS, Jackson SE. Linking competitive strategies with human re- Research in organizational behavior. Greenwich (CT): JAI Press, 1998.
source management practices. Acad Manage Exec 1987;1:207 – 19. p. 77 – 140.
Shepherd WG. A treatment of market power. New York: Columbia Univ. Wooldridge B, Floyd SW. The strategy process, middle management in-
Press, 1975. volvement, and organizational performance. Strategy Manage J. 1990;
Smith KG, Smith KA, Olian JD, Sims HP, O’Bannon DP, Scully JA. Top 231 – 41 (March/April).
management team demography and process: the role of social integra- Zammuto RF, Krakower JY. Quantitative and qualitative studies of organ-
tion and communication. Adm Sci Q 1994;39:412 – 38. ization culture. In: Woodman RW, Passmore WA, editors. Research in
Suresh K, Orna D. Generic manufacturing strategies: a conceptual synthe- organizational change and development. Greenwich (CT): JAI Press,
sis. Strategy Manage J 1989;10:211 – 31. 1991. p. 83 – 114.
Van Der Velde M, Jansen P, Vinkenburg C. Managerial activities among top Zimmerman GC. Factors influencing community bank performance in
and middle managers: self versus other perceptions. J Appl Manage California. Econ Rev—Fed Reserve Bank San Francisco 1996;1:
Stud 1999;8(2):61 – 174. 26 – 42.